初创公司尽调
尽调报告 robotics/hardware Series B / late-stage private 2026-06-10

Sunday

家用机器人登月式项目,技术履历顶尖、融资推进很快,但公开经济性仍缺位

Sunday 把差异化真实家庭数据闭环、顶级机器人背景和蓝筹融资放在一起,但公开记录仍缺少支撑 $1.15 billion 估值所需的价格、客户证据和单位经济。

封面要素

成立时间 01
2024 [CO002]
总部 02
Mountain View, California [CO001]
Series B 03
165 USD M [CO021]
隐含估值 04
1150 USD M [CO021]
已披露总融资 05
200 USD M [CO023]
计划 beta 家庭 06
50 households [CO010]
报道等候名单 07
~1000 people [CO027]

公司概况

Sunday 是一家总部位于 Mountain View 的私营家用机器人公司,由 Tony Zhao 和 Cheng Chi 于 2024 年创立。公司在 打造 Memo:一款带轮式底盘的家庭机器人,瞄准重复性家务;同时搭建全栈训练系统,包括 Skill Capture Glove、 Skill Transform、ACT-1,以及从真实家庭获取数据的 Memory Developer 数据闭环。公开证据显示,这家公司技术 履历罕见、融资节奏很快,并在 2026 年 3 月完成估值 $1.15 billion 的 $165 million Series B,且计划在 2026 年末受控 beta;但公司仍未披露足以清晰承保商业化的数据。

官网
www.sunday.ai
成立时间
2024-01-01
创始人
Tony Zhao, Cheng Chi
创立地点
Mountain View, California, USA
总部
Mountain View, California, USA
产品
Sunday 正在打造 Memo:一款带轮式底盘的家用机器人,面向清桌、洗碗、洗衣、收纳等重复性家务及相关家庭流程。可见 产品栈还包括 Skill Capture Glove、Skill Transform、ACT-1,以及 Memory Developer 网络,目标是把真实家庭 行为转成机器人训练数据。
客户
目标客户是忙碌家庭,以及想减少重复家务负担的家庭;触达方式是直接面向家庭的 beta 和未来消费者推广,而不是企业买家 或渠道伙伴。
商业模式
尚未商业化的直接面向家庭硬件加支持模式,配合受控的 Founding Family Beta、公司主导的上门引导,以及反哺未来产品 发布的数据采集闭环。公开定价、订阅、融资和续约条款仍未披露。
阶段
Series B / late-stage private
融资情况
Sunday 于 2025 年 11 月带着 $35 million 启动融资走出隐身状态,随后在 2026 年 3 月宣布完成超额认购的 $165 million Series B,估值 $1.15 billion;Tracxn 暗示两轮合计已披露融资约 $200 million。
[CO001, CO002, CO003, CO010, CO021, CO023, CO026, CO043]

执行摘要

主要优势

  • Sunday 拥有差异化真实家庭数据护城河,公开称覆盖 500-plus 户家庭、约 10 million 段家庭 episode,并已出货超过 2,000 副 Skill Capture Gloves。
  • 创始团队在 ALOHA、ACT 和 Diffusion Policy 上有罕见强势的公开机器人履历。
  • 2026 年 3 月由 Coatue 领投的 Series B,以及约 $200 million 已披露融资,为 Sunday 运行真实世界 beta 提供了有分量的资本。
  • 具名的 50 户 Founding Family Beta 和据报道约 1,000 人 waitlist,显示大规模发布前已有真实家庭兴趣。

主要风险

  • Sunday 仍未披露 Memo 价格、付费客户证据、收入或 ARR,也没有给出毛利率桥梁。
  • 作为入户联网机器人,Sunday 对机器人特定安全、隐私、合规和数据治理细节的披露仍明显不足。
  • 首批 50 户 beta 可能暴露精选 demo 看不到的隐性人工、高支持强度或可靠性不足。
  • 消费者为家庭机器人付费的意愿,可能远低于支撑 Sunday 硬件加服务经济性所需的价格。
  • 融资条款、清算优先权和更广泛的治理安排仍足够不透明,headline valuation 未必反映真实风险。

未决问题

  • Memo 价格、融资方案、是否收取订金,以及 beta 转付费预期。
  • 计划部署规模下的硬件 BOM、毛利率、支持成本、现金消耗和现金跑道。
  • Beta 家庭使用数据、干预或遥操作比例、现场可靠性和更换指标。
  • Series B 及此前轮次条款堆叠,包括清算优先权、治理权利和稀释保护。
  • 机器人特定的数据保留、入户感知政策、安全测试和认证路线图。

目录

Chapter 01

01公司概览

1.1 身份、产品与运营模式

Sunday 是一家总部位于 Mountain View 的家用机器人初创公司,由 Tony Zhao 和 Cheng Chi 于 2024 年 创立。从官网、发布材料到第三方报道,公司始终把 Memo 描述为一款有用的家用机器人,能够承担清桌、装洗碗机、 叠衣服、日常整理等重复性家务。公开记录也表明,Sunday 并不把自己定位为纯软件供应商。公司材料强调,它在软件、 硬件、数据采集、训练和模型评估上掌握全栈;发布新闻稿则强调 Memo 采用轮式底盘和柔软硅胶包覆设计,而不是双足 人形形态。两者合在一起,更适合把 Sunday 理解为一家垂直整合的机器人公司,目标是最终把机器人部署到家庭,而非 一个狭义 AI 工具初创公司。 目前可见的核心商业模式是尚未商业化的部署,而不是成熟的经常性收入。Sunday 称 beta 部署将于 2026 年末启动, Founding Family Beta 覆盖 50 个家庭;但已检索来源没有披露定价、收入或客户合同。运营论点转而落在数据积累: 公司称 Skill Capture Glove 及相关工具能把人类家庭行为规模化转成机器人训练数据。官方帖子称已出货超过 2,000 只手套,训练数据来自超过 500 个家庭和约 1000 万次家庭场景片段,Memo 也从 2024 年 12 月的单臂摆鞋 系统推进到 2025 年 10 月更复杂的家务。这些都是有意义的产品进展信号,但相比大规模部署的消费者硬件业务,仍处在 商业化曲线更早的位置。[CO001, CO002, CO003, CO004, CO005, CO006]

快照 KPI 表
指标数值 / 状态日期信心缺口 / 备注
成立20242024年份已被佐证,但手头公开来源未披露确切注册日期。
总部Mountain View, California2026-03-12招聘证据也显示 Redwood City 有活跃岗位,因此更广义的运营足迹横跨不止一个 Bay Area 标签。
当前阶段Series B / 后期私营2026-03-12官方和市场数据来源支持这一阶段判断。
Series B$165M,估值 $1.15B2026-03-12官方公告和多家媒体佐证了金额和估值。
累计融资两轮合计 $200M2026-03-18总额来自 Tracxn 推算,而非公司股权结构表披露。
Series A$35M 发布融资2025-11-19发布新闻稿与 Tracxn 对 2025 年 11 月金额一致。
训练数据>500 个家庭和约 10M 段情境2025-11-19这些是公司声称的运营指标,而非独立审计数字。
Beta 家庭计划 50 个 founding-family 家庭2025-11-19部署目标公开,但完成仍属前瞻。
员工数据报道 127 名员工;据报道 >70 名工程师和研究人员2026-03-26来源提供的是代理指标,而非经审计的当前人数。
收入 / ARR / 具名客户2026-06-10保留来源中没有公开收入、ARR、客户数或具名客户证据。

混合了直接披露的融资和部署指标,并在公开经济性缺失处明确标出缺口;null 表示保留证据中未公开披露。

[CO001, CO002, CO008, CO010, CO011, CO021]
FO002: 公司快照逻辑

展示创始人研究、数据采集系统、产品部署、资本和关键依赖如何在 Sunday 模型中连起来。

[CO004, CO005, CO006, CO020, CO021, CO023]
FO003: 快照 KPI

公开可支撑的成熟度和部署指标,不包括未披露的收入和客户指标。

只使用保留来源明示的数据,或可从公开网站直接观察的数据;未披露时,经济指标有意留空。

[CO007, CO008, CO010, CO021, CO023, CO029]

1.2 领导层、治理与关键人物集中度

公开领导层披露仍很窄,且高度围绕创始人。Tony Zhao 是可见的联合创始人兼 CEO,Cheng Chi 是联合创始人兼 CTO。Zhao 的个人网站把他直接关联到 Stanford 的 ALOHA 和 ACT,并描述其曾在 Google DeepMind、Tesla Autopilot 和 Google X Intrinsic 工作,这解释了外部投资人为何不断强调他的研究履历。Cheng Chi 是 Diffusion Policy 论文作者,又被标注为 Sunday 的 CTO,进一步说明公司的技术叙事建立在两位创始人近期且可识别 的机器人学习资历之上。 治理能见度远低于融资头条。保留来源中,最清楚的公开董事会披露是 Thomas Laffont 因 Coatue 领投 2026 年 3 月 Series B 而加入董事会。除此之外,Sunday 未在已检索材料中发布完整董事名单、股权拆分或更广治理结构。这意味着 公开公司画像仍暴露于关键人物依赖:Zhao 和 Chi 不只是高管,也是研究可信度、产品愿景和投资人信心的主要证明。 这种不对称很重要,因为后期资本已经到位,但高级运营人才、董事会监督或 GTM 领导层的公开披露包并未同步变得详细。 后续尽调应把创始人集中度和治理不透明视为真实承保问题,而不是表面缺漏。[CO013, CO014, CO015, CO016, CO017, CO018]

领导层与创始人表
人物职务背景创始人-市场匹配 / 职能覆盖关键人物依赖
Tony Zhao联合创始人、CEOStanford 机器人 PhD 背景;参与 ALOHA 和 ACT;曾任职 Google DeepMind、Tesla Autopilot 和 Google X Intrinsic。公开层面连接前沿 manipulation 研究、投资人融资和 Memo 商业化叙事的主要桥梁。高——Sunday 的公开案例仍高度依赖 Zhao 的技术声誉和创始人愿景。
Cheng Chi联合创始人、CTODiffusion Policy 共同作者,并被公开识别为 Sunday CTO。在创始团队内部锚定机器人学习可信度和模型系统架构。高——创始人技术深度是 Sunday 差异化故事的核心。
Thomas LaffontCoatue 联合创始人;Series B 后董事会成员官方和媒体报道均称其随 2026 年 3 月融资加入 Sunday 董事会。代表正式外部监督和后期投资人影响力。中——重要治理信号,但融资之外的公开职责没有详细说明。
Eric VishriaBenchmark 普通合伙人和早期支持者在发布材料中被引用,用来阐述 Sunday 的数据收集逻辑和早期投资案例。代表在商业证明出现前,知名 venture sponsor 已有早期 conviction。中——投资人支持可见,但保留来源未披露正式治理席位。

仅为部分公开名单。保留来源清楚识别创始人,并点名与投资人相关的监督角色,但未披露完整董事会或完整高管组织图。

[CO013, CO014, CO015, CO016, CO017, CO018]

1.3 融资基础、利益相关方与公开规模信号

Sunday 的资本形成速度异常快。公司于 2025 年 11 月 19 日带着与 Benchmark 和 Conviction 相关的 $35 million 启动融资走出隐身状态,随后在 2026 年 3 月 12 日宣布完成超额认购的 $165 million Series B, 估值 $1.15 billion。官方和独立来源一致将 Coatue 标为 Series B 领投方,并将 Bain Capital Ventures、 Fidelity Management & Research Company、Tiger Global、Benchmark、Conviction 和 Xtal Ventures 列为 参与方。Tracxn 的画像暗示两轮已披露融资约 $200 million,与 $35 million 启动轮和 $165 million Series B 在算术上吻合。 公开规模信号方向积极,但仍不完整。Tracxn 在 2026 年 3 月下旬报告员工 127 人,TechFundingNews 则称公司有 70 多名工程师和研究员。Sunday 自己的 Series B 帖子称,随着公司转向部署,工程团队扩大三倍、研究团队扩大四倍、 数据运营增长五倍。招聘页显示 Redwood City 和远程岗位仍在招软件与数据运营人才,说明其 Bay Area 足迹可能比 Mountain View 新闻稿措辞更宽。与此同时,公开记录没有披露收入、ARR、客户数、债务额度、二级出售或具名商业客户。 因此融资故事很强,但经济画像仍大多私有。[CO021, CO022, CO023, CO024, CO025, CO026]

利益相关方或投资人图谱
利益相关方角色控制 / 经济重要性尽调要求
CoatueSeries B 领投方领投 $165M Series B,并为 Thomas Laffont 获得一个董事席位;很可能是当前轮次中最可见的外部权力中心。确认持股、董事会权利、pro rata 结构和任何保护性条款。
BenchmarkSeries A 支持者和 Series B 参与方早期支持者,在发布信息中被大量引用;即便没有公开披露席位,也可能在创始人支持和早期治理上有影响力。核实当前持股、任何董事会或观察员权利,以及 Benchmark 是否 bridge 了 B 轮。
ConvictionSeries A 支持者和 Series B 参与方显示公司从 stealth 退出到后续融资的连续性,并强化创始人网络支持。澄清支票规模、reserve 参与,以及任何特殊治理权利。
Bain Capital VenturesSeries B 参与方为股权结构表增加 go-to-market 和公司建设品牌信号。判断资本之外的战略支持承诺,以及 BCV 是否拥有信息权。
Tiger GlobalSeries B 参与方大型 crossover / growth 投资人出现,可能影响后续融资信号和预期。询问节奏预期、follow-on appetite 和治理参与。
Fidelity Management & Research CompanySeries B 参与方偏公开市场的投资人参与,可能影响后期融资观感和未来退出预期。确认这是否只是 balance-sheet participation,还是更广泛 crossover syndicate strategy 的一部分。
Xtal VenturesSeries B 参与方被列为本轮一部分,但相比更大品牌公开描述不足。澄清持股、战略相关性,以及 Xtal 是否带来特定行业运营支持。

仅列出已披露轮次中公开具名的利益相关方;股权结构表百分比、清算优先权,以及任何债务或老股部分仍未披露。

[CO011, CO019, CO021, CO022, CO023, CO024]

1.4 里程碑、证明点与反向背景

里程碑记录显示技术推进很快,叙事升温也同样快。Sunday 称其花了一年多搭建 Skill Capture Glove、Skill Transform、Memo 和 ACT-1 基础模型背后的核心基础设施。研究帖子称 Skill Transform 能以 90% 成功率把手套数据 转成机器人等效数据,ACT-1 可执行从餐桌到洗碗机的任务,并借助地图条件导航泛化到未见过的家庭。每个技术主张能否 经受独立基准测试仍不清楚,但从 2024 年原型进展,到 2025 年退出隐身,再到 2026 年独角兽融资,这条序列被压缩 得很短。 主要反向背景不是保留材料中的公司特定诉讼或执法行动,而是行业怀疑。Robot Report 引用 Ken Goldberg 的观点: 家庭管家时间表带有炒作,灵巧性和机器人数据缺口还要多年才能克服。IEEE Spectrum 又给出更市场化的警示:调查 参与者在家庭中通常更偏好专用机器人而非人形机器人,即便文章乐观情景也假设了安全、监管批准和保险条件,而这些 条件可能距离现实还有数十年。这些怀疑并不能证伪 Sunday 的论点,但确实正确界定了核心尽调问题。Sunday 具备强 产品动能、蓝筹资本和独特的数据获取故事,但在客户采用、定价,以及大规模家庭部署所需运营可靠性上,仍缺公开证明。[CO033, CO034, CO035, CO036, CO037, CO038]

里程碑表
日期事件类型金额 / 估值 / 状态参与方含义
2024公司成立成立已披露年份;具体日期未公开Tony Zhao;Cheng Chi说明公司作为非常年轻的创业项目,快速进入后期融资。
2024-12Memo 有一只手臂,并学会整理鞋子产品原型里程碑Sunday展示 stealth 发布前的早期能力证明。
2025-10Memo 学会袜子、玻璃器皿和 espresso 任务产品扩展家庭技能集Sunday证明公开发布前,公司正加速覆盖更灵巧的家务。
2025-11-19Sunday 走出 stealth 并发布 Memo规模宣布 $35M 发布融资Sunday;Benchmark;Conviction公开亮相把产品叙事和即时 venture 背书绑定在一起。
2025-11-19Founding Family Beta 开放申请产品计划 2026 年底进行 50 户 betaSunday把产品故事从纯演示推进到结构化真实家庭部署。
2026-03-12宣布超额认购的 Series B融资$165M,估值 $1.15BCoatue、Bain Capital Ventures、Fidelity、Tiger Global、Benchmark、Conviction、Xtal Ventures(投资方)确认公司在走出 stealth 约四个月内获得独角兽估值和重要资本基础。
2026-03-12Thomas Laffont 加入董事会治理董事会新增成员Coatue;Thomas Laffont提供保留证据集中最清晰的公开治理披露。
2026-03-12公司宣布「不再演示」并转向部署规模Beta 交付将在数月内开始Sunday表明执行风险已从 R&D 叙事转向真实世界可靠性。
2026-03-12职能团队快速扩张规模Engineering 3x、Research 4x、Data Ops 5x(团队扩张)Sunday暗示广泛商业化发布前,公司正在进行重度部署前组织扩张。
2026-03-26第三方 profile 报告 127 名员工规模员工数代理指标Tracxn在确切当前员工数未披露时,提供粗略规模信号。
2026-06-10招聘页面显示 Redwood City 和远程招聘规模多个软件和数据岗位开放Sunday支持公司在 3 月融资后仍处于快速 build-out 模式的判断。
2025-09-03 至 2025-09-03围绕家用 humanoids 的独立行业怀疑加剧反向无公司特定行动;仅为行业谨慎Robot Report;IEEE Spectrum将主要下行界定为品类执行和采用风险,而不是已知公司特定法律事件。

这是保留来源中的日期化记录年表。若干条目是公司声称或第三方报道,而非独立审计的运营里程碑。

[CO002, CO009, CO010, CO011, CO019, CO021]
FO001: 公司里程碑时间线

追踪 Sunday 从 2024 年创立,到退出隐身、融资、团队扩张,以及仍在塑造家用机器人赛道的品类质疑。

只有月份或年份公开时,时间线保留较粗颗粒度日期,而不编造具体日期。

[CO002, CO009, CO010, CO011, CO019, CO021]

1.5 图表

Chapter 02

02市场分析

2.1 市场边界、纳入支出与现状替代品

Sunday 的相关市场不是“所有人形机器人”,甚至也不是所有家用自动化硬件。最接近且可辩护的边界,是在家庭内部运行、 用于家务、监测或陪伴的家用服务和个人护理机器人。Sunday 自己的产品材料反复回到厨房、餐厅、洗衣空间和其他真实 居家环境中的家务辅助:清桌、装洗碗机、叠衣服、收鞋,以及类似移动操作任务。ISO 13482 对移动仆人和身体辅助 机器人的框架,与这条边界方向一致,因为它明确覆盖改善生活质量、处理人类接触风险、并在人、家养动物和财产周围使用 的个人护理机器人。 因此,纳入支出从清洁及相关家务的家用机器人开始,再延伸到相邻的居家辅助品类;这些品类的买家重视便利、劳动替代 或居家养老支持。它排除工业机器人、医疗设备、军用机器人、玩具,以及那些把医疗、酒店、制造、教育、国防和住宅用例 混在一个数字里的宽泛人形机器人预测。更宽预测可作为相邻背景,但会高估 Sunday 当前有证据支撑的机会。实际现状也 比竞争机器人供应商更宽:家庭已经靠无偿劳动、雇佣服务和扫地机器人等专用设备完成这项工作。这很关键,因为 IEEE 调查证据显示,出于安全、隐私和舒适感,用户仍偏好专用家用机器人而非人形机器人;所以 Sunday 不只要击败其他初创 公司,还要击败根深蒂固的非机器人习惯和更简单设备。[CM001, CM002, CM003, CM004, CM005, CM011]

市场定义表
细分 / 品类纳入支出排除支出买方 / 付款方与 Sunday 的相关性
核心家务机器人为清洁、整理、餐具处理、洗衣辅助、监控或其他家内服务任务而销售的家用机器人工业、医疗和军用机器人从消费者预算支付的家庭购买者核心边界;这是最接近 Memo 的公开品类
单一用途清洁机器人扫地机器人、拖地机器人、割草机器人、泳池清洁机器人超出专用任务的通用 manipulation从家电预算支付的家庭购买者作为当前收入中心和现状基准纳入
陪伴 / 老年辅助家用机器人家内辅助、陪伴、监控、提醒和 aging-in-place 支持机构医疗机器人家庭或照护者;保险方预算在此未被公开证明相邻近期细分;相关,但尚非 Sunday 已表述的产品重点
广义 humanoid 机器人仅纳入住宅辅助讨论医疗、酒店、零售、教育、工厂、国防和物流 humanoids混合企业和机构买方对 Sunday TAM 过宽;仅可作为相邻背景
个人护理机器人监管框架在人、动物和财产周围使用的移动仆役和身体辅助机器人机器人玩具、工业机器人、医疗设备、军用机器人、飞行或水上机器人取决于部署场景重要的安全和品类边界参照
现状替代品无偿家务劳动、雇佣服务和专用设备N/A家庭时间预算或服务支出真正的竞争对手往往是「自己做」或「买一台更简单的机器人」

该表混合了产品市场边界和替代品逻辑,因为 Sunday 同时与相邻机器人品类和非机器人家庭解决方案竞争。

[CM001, CM002, CM003, CM004, CM005, CM031]
FM003: 买方 / 证明到采用地图

买方地图不只回答谁付钱,也回答通用家用机器人走出 beta 前,每个细分需要什么证明。

该矩阵是定性地图,展示谁购买、谁受益,以及每个细分规模化前必须出现哪些证据。

[CM018, CM020, CM021, CM024, CM026, CM039]

2.2 市场规模视角、地理收窄与矛盾估计

自上而下的公开市场数据只有在品类边界明确时才有用。Mordor Intelligence 估计家用服务机器人市场 2025 年为 $14.62 billion,到 2031 年升至 $34.93 billion,CAGR 为 15.62%;Global Market Insights 则估计更宽的 household robots 市场 2025 年为 $14.7 billion,并预计 2035 年达 $85 billion,CAGR 为 19.2%。这些估计在 当年头条数字上接近,但分类和长期预测形态分化。Mordor 更明确指出,地面清洁和吸尘 / 拖地占据当下收入池主导; GMI 强调的家庭框架更宽,仍以清洁为主,但也包括陪伴、安全、草坪和泳池应用。没有一个来源单独拆出 Sunday 真正 需要赢下的细分:超越单一清洁功能的高端多步骤家庭操作。 更受约束的测算路径是地理范围。Mordor 称北美在 2025 年占家用服务机器人收入 38.3%,套用其 $14.62 billion 全球基数,意味着北美支出约 $5.6 billion。GMI 另行估计美国 household robot 市场 2025 年为 $3.6 billion。 对 Sunday 来说,这些比全球消费者机器人数字更适合做近期锚点,因为 Sunday 的公开部署、训练家庭和 beta 信息都 以美国为中心。相比之下,Future Market Insights 关于人形机器人的预测——2025 年 $7.8 billion、2036 年 $248.9 billion——最好视为相邻天花板,而不是 Sunday TAM,因为它混合了医院、工厂、酒店、国防、教育、零售和 居住辅助。 因此结论刻意保守。广义家用机器人 TAM 显然是数十亿美元级,北美和美国提供了更窄但仍有意义的支出池,家庭家务 也是确实存在的经济痛点。但公开数据无法可信拆出 Sunday 专属 SAM 或 SOM,因为没有披露价格、没有公开转化漏斗、 没有独立验证的任务频次经济性,也没有公开证据证明消费者会以当前隐含的硬件价位购买通用家用机器人。矛盾本身就是 重要尽调信号:慷慨的市场报告描绘了大品类,而真实采用证据仍像一个高端、受信任限制的楔子市场。[CM006, CM007, CM008, CM009, CM010, CM011]

TAM / SAM / SOM 或规模测算视角表
发布方 / 视角年份地理数值CAGR方法论信心局限
Mordor Intelligence 家用服务机器人2025 / 2031全球2025 年 $14.62B;2031 年 $34.93B15.62%面向家用服务机器人的自有分析师估计品类宽于 Sunday,且仍以清洁为主
Global Market Insights 家庭机器人2025 / 2035全球2025 年 $14.7B;2035 年 $85B19.2%面向家庭机器人的分析师估计长期预测反映的是更宽品类,且 taxonomy 不同于 Mordor
Global Market Insights 美国家庭机器人2025美国2025 年 $3.6BN/A家庭机器人市场报告中的国家拆分美国数字包含超出 Sunday 当前高端 manipulation wedge 的品类
Mordor Intelligence 北美份额2025北美全球家用服务机器人收入的 38.3%N/A家用服务机器人报告中的区域份额份额没有隔离美国或 Sunday-like 产品
推导的北美家用服务机器人2025北美~$5.6B(38.3% x $14.62B)N/A作者根据 Mordor 全球数值和区域份额计算推导代理值,不是已发布的独立市场估计
Future Market Insights humanoid 机器人2025 / 2036全球2025 年 $7.8B;2036 年 $248.9B37.0%覆盖多个终端市场的宽口径人形机器人预测该预测把工厂、医院、国防、酒店等场景都算进去,夸大了与 Sunday 的相关性
Altman Solon 支付意愿视角2025美国消费者69% 不愿支付超过 $5,000;25% 任何价格都不愿支付N/A超过 1,000 名美国消费者的调查这是采用意愿视角,不是市场规模数字;反映态度而非品类收入

前五行是评估 Sunday 最有用的公开规模测算视角。人形机器人一行刻意保留为过宽的相邻市场,支付意愿一行则是采用上限,而不是 TAM。

[CM006, CM007, CM008, CM009, CM010, CM013]
FM001: Sunday 市场规模视角

宽口径公开市场是家庭或家政服务机器人;更有用的近期视角收窄到北美和美国,而 Sunday 专属 SAM 和 SOM 仍无法从公开证据量化。

前三层是有来源支撑的公开估算或透明算术。最后两层有意不填数字,因为没有公开证据切出 Sunday 的可服务或可获取份额。

[CM006, CM007, CM008, CM009, CM010, CM014]
FM002: 市场估算区间与口径漂移

Sunday 所在品类的公开市场数字,更适合作为边界区间阅读,而不是单一精确 TAM。

所有行均使用十亿美元。中点只是简单算术便利项,不是单独的出版方估算。各行有意保留口径漂移,而不是混合不可比品类。

[CM006, CM007, CM008, CM010, CM012, CM013]

2.3 买家、用户、付款方与采用路径

Sunday 的公开 GTM 图景异常直接:家庭似乎同时就是买家、用户和付款方。不同于企业机器人或可报销医疗设备,当下 公开记录中看不到保险、雇主、房东或政府预算所有者。因此近期销售更像高客单价消费耐用品或高端家电,而不是采购驱动 的系统。Sunday 的 Founding Family Beta 让这一点尤其清楚:采用从志愿家庭开始,这些家庭愿意接纳早期机器人、 提供反馈并容忍迭代。最强的早期采用者很可能是富裕、技术前瞻型家庭:他们已经在家务上投入可观时间或金钱,重视智能 家居试验,也能承受硬件风险。 在这个宽泛家庭买家内部,用户和价值故事因细分而异。双职工家庭和父母面对最清楚的时间压力论点,因为常规清洁、整理、 洗碗和洗衣会与工作和育儿时间竞争。BLS 时间使用数据显示,家庭活动平均每天已经消耗约两小时;Altman Solon 调查 显示,最受期待的机器人工作是清洁、安全、做饭、洗衣和送物,而不是新奇功能。老年人和照护者家庭是相邻但尚未被证明 的细分:人口顺风确实存在,Mordor 和 GMI 都预计 elder-care 或陪伴应用会增长,但 Sunday 尚未披露针对养老的 产品、报销渠道,或面向虚弱用户的安全案例。 因此采用路径更可能是顺序推进,而非大众市场一拥而上。先靠 demo 和媒体建立认知,再用严格管理的 beta 证明居家 可靠性,随后由独立评测和口碑降低信任摩擦;只有到那时,更广泛的消费者购买才变得可信。Altman 调查说明了为什么 这个顺序重要:兴趣很高,但熟悉度低,安全担忧显著,多数受访者不愿支付超过 $5,000。Sunday 未来可能受益于 居家养老或订阅模式,但今天的公开证据支持的是一条消费者自费采用路径,由信任意愿、支付意愿和常规节省时间证明共同 约束。[CM017, CM018, CM019, CM020, CM021, CM022]

细分市场 / 买方地图
细分市场买方用户付款方工作流预算负责人采用触发点
Founding Family 测试家庭自愿参与的早期采用家庭整个家庭;通常是主要家庭管理者同一家庭申请、接待测试机器人、提供反馈、容忍迭代有可支配收入的家庭决策者好奇心,加上愿意一起打磨产品
双职工、时间紧张家庭家中的在职成年人负责洗碗、整理和洗衣的成年人同一家庭把机器人评估为省时家电或服务家庭运营预算 / 大件耐用品预算相比家务和外包,能清楚节省每周时间
有孩子的家庭父母或监护人父母、年龄较大的孩子或家庭整体同一家庭用机器人减轻育儿需求下的日常清洁负担主要照护者或家庭共同预算有偿工作、育儿和家务叠加带来的痛点
高端智能家居爱好者技术前沿型房主或租客同一买方或同住家庭同一家庭作为更广泛智能家居组合和试验的一部分购买可自由支配的消费电子预算新能力,加上对集成和支持的信任
老年人 / 照护者家庭老年人、成年子女或照护者老年住户或照护者家庭或家族支持网络评估居家辅助、提醒、轻家务和安全舒适度家庭预算;公开证据未显示保险方或政府付款方原地养老需求,以及对居家监控的接受度
现状清洁设备买家正在比较扫地机器人或服务的家庭同一家庭同一家庭拿通用机器人与专用设备或雇佣劳动力比较家电预算或服务预算证明更宽口径机器人在信任和 ROI 上胜过更简单工具

现有公开证据支持的是消费者自费路径。老年人和照护者家庭属于相邻细分市场,而不是 Sunday 已清楚披露的上市重点。

[CM017, CM018, CM020, CM021, CM022, CM023]
FM004: 采用路径与反馈循环

Sunday 可能的采用路径,是从广泛消费者兴趣进入窄 beta;之后只有真实家庭中的信任、安全和 ROI 证据持续叠加,才会规模化。

[CM018, CM019, CM024, CM026, CM036, CM037]

2.4 增长驱动、采用约束与保留的尽调缺口

Sunday 所在品类的结构性增长逻辑真实存在。家庭仍把大量时间花在家务上,智能家居预期在上升,主要公开市场报告 也把品类增长归因于便利性、AI/IoT 集成、家务劳动力短缺和人口老龄化。Sunday 自己的差异化论点也对应真实市场需求: 家用机器人在杂乱、多变环境中无法泛化时就会失败;如果一家公司能采集更好的居家操作数据,就可能获得性能优势。 如果 Sunday 关于 500 个家庭和 1000 万次场景片段的数据主张在运营上站得住,公司就能攻打单一用途清洁机器人 难以解决的那部分市场。 但约束至少同样重要。安全和标准尚未解决。ISO 13482 为个人护理机器人提供基线框架,但 IEEE 的标准报道认为,当前 家用人形机器人标准仍缺面向真实家庭自主性的强制合规标准、测试方法和执行机制。NIST 同样把采用条件定义为严格验证、 安全人机协作,以及在非结构化环境中可测量的灵巧性、感知和移动表现。需求侧,Altman 展示了鲜明的支付能力和信任 缺口,IEEE 显示用户仍偏好专用机器人而非人形机器人,The Robot Report 和 UNSW 也保留了更深层怀疑:灵巧性、隐私、 遥操作依赖和“数据缺口”问题,可能让真正有用的家用人形机器人还要多年才能到来。 尽调中应保留这些矛盾,而不是把它们平均掉。公开来源支持一个数十亿美元级品类和可信的长期社会需求,但尚不足以支持 关于大众市场时点、Sunday 最终 ASP、续约经济性、事故率,或消费者长期愿意让通用机器人在家人身边自主运行的判断。 残留尽调缺口因此是商业和运营问题,不只是数字问题:价格、转化、独立任务成功指标、故障模式、支持负担,以及一个家庭 在从 Roomba 式工具升级到能力强得多、也亲密得多的机器人之前,到底需要多少信任。[CM024, CM026, CM027, CM028, CM029, CM030]

增长驱动因素与约束表
驱动因素 / 约束方向时间含义尽调追问
家务仍然消耗大量时间驱动因素立即支撑家务自动化的核心价值主张Memo 在真实测试家庭中按任务每周净节省多少小时?
智能家居和 AI 集成趋势驱动因素近期提高消费者把机器人纳入家庭自动化的意愿Memo 与其他家庭系统集成有多深,需要哪些隐私授权?
人口老龄化和辅助生活需求驱动因素中期把居家辅助的长期需求从清洁扩展到更多场景Sunday 能否安全服务老年人,还是老年照护仍只是相邻市场?
Sunday 的全栈数据飞轮驱动因素近期如果居家数据确实能复利,可能加快模型改进公司任务成功率提升中,数据量和硬件迭代各贡献多少?
消费者对实用家庭功能有兴趣驱动因素立即支撑围绕清洁、洗衣、做饭和安防的功能优先级哪些用户请求的任务在部署后能转化为重复使用和留存?
前期价格敏感度高约束立即在成本下降或融资改善前,采用可能被限制在高端家庭Sunday 的实际硬件或订阅价格是多少,与 <$5k 的调查上限相比如何?
安全、信任和标准缺口约束立即至中期需要证明它能在家人、宠物和杂乱家庭环境中安全运行Sunday 现在使用哪些测试方法、认证和事故响应流程?
隐私和远程操作顾虑约束立即居家视频、远程协助和数据收集可能劝退买家并引发监管关注远程人工干预使用频率多高,哪些数据会被存储或由人工查看?
用户更偏好专用机器人而非人形机器人约束近期宽口径多用途价值必须明显超过扫地机器人等更简单设备Sunday 如何量化自身相对单一用途设备在可靠性和总成本上的优势?
对灵巧性和数据缺口的怀疑约束中期行业热度可能跑在杂乱家庭中实际泛化能力前面在 Sunday 控制的演示和测试叙事之外,有哪些任务成功率独立证据?

时间相对于 2026-06-10 的 runDate。表格刻意同时纳入结构性市场力量和公司特有的商业化约束,因为两者都会塑造 Sunday 的真实采用曲线。

[CM022, CM024, CM026, CM027, CM028, CM029]
Chapter 03

03竞争对手

3.1 格局:直接同业、既有玩家、相邻路径与替代品

Sunday 面对的不是一个干净的同业集合。保留证据中,最接近的直接 home-first 对手是 1X 和 Figure,因为两者也把 通用人形系统营销给家庭任务,而不是单一清洁或仓库工作。Samsung Ballie 和 iRobot 的重要性在另一处:它们说明 家庭可能更偏好更窄、更可控、与既有智能家居或地面护理习惯绑定的助手,而不是完整人形机器人。Agility、Boston Dynamics、Tailos 和 LG 仍重要,但主要作为相邻路径,证明机器人公司今天在哪里变现:仓库、工业物料搬运、酒店 清洁和商业服务机器人,而不是混乱的消费者家庭。 这种分类关系到 Sunday 的真实竞争问题。一个家庭在决定是否信任家用机器人时,仍可以维持无偿劳动、外部清洁服务、 Roomba 级产品或智能家居套装,同时等待通用机器人成熟。换句话说,Sunday 并不面对当前对手的赢家通吃锁定。它面对 的是宽替代集合,以及一个可能进入者集合,里面有渠道更强、资产负债表更大或信任基础更稳的公司。因此,格局应被框定 为直接同业、窄型既有玩家、工业相邻方和现状,而不是单一人形机器人排行榜。[CP001, CP004, CP010, CP015, CP019, CP022]

竞争对手画像表
竞争对手类别规模 / 融资信号目标客户产品范围公开定价信号战略方向 / 局限
Sunday直接同业保留的竞争对手来源未披露真实家庭中的忙碌家庭通用型家庭家务机器人产品和数据循环优先服务家庭;仍需证明可靠性和经济性
1X直接同业$100M Series B;不到 12 个月累计融资超过 $125M消费者家庭通用型家庭人形机器人(NEO)明确优先服务消费者家庭,但仍在解决安全操作和高级推理
Figure直接同业>$1B 承诺 Series C,投后估值 $39B最终面向家庭和商业运营方搭载 Helix、规划进入家庭的通用型人形机器人资本雄厚且能接触住宅数据,但家庭测试仍处于 alpha 阶段
Agility Robotics相邻 / 工业声称已有生产部署;这里未披露公开消费者规模仓库和物流运营方人形机器人搬运周转箱和设施地面自动化商业证明来自工业 ROI,而非家庭使用
Boston Dynamics相邻 / 工业成熟机器人公司;这里未披露单独人形机器人融资数字工业物料搬运客户Atlas 工业人形机器人产品重心在工厂和物流,而不是家庭
Samsung Ballie既有厂商 / 可能进入者Samsung Electronics 概念项目智能家居消费者滚动式 AI 家庭陪伴设备和家电控制器渠道能力强,但保留证据未显示价格或上市细节
LG robotics可能进入者 / 相邻母公司是大型家电和企业解决方案厂商,这里未保留家庭机器人规模数字保留证据中的企业客户CLOi 商用服务机器人这里保留的公开证据指向商用,而不是已出货家庭机器人
iRobot既有替代品公开市场家庭机器人既有厂商;并购失败后重组大众市场家庭用户专用地面清洁机器人即便缺少人形机器人宽度,窄任务领先和用户控制经验仍然重要
Tailos Rosie相邻替代品12 个国家超过 1,000 台机器人酒店和住宿运营方商用扫地机器人运营导向清洁设备,明确非住宅
Everyday Robots历史反向先例Alphabet 支持、后来关闭的内部项目Google 内部运营 / 办公环境通用型移动操作原型展示了技术潜力,但未能商业化为独立业务

覆盖不完整,但刻意横跨直接同业、既有厂商、相邻公司、替代品和最强反向先例;定价为 null 表示保留证据中没有公开披露。

[CP001, CP004, CP006, CP007, CP010, CP011]
FP001: 竞争定位地图

有证据支撑的序数评分,比较各竞争对手在家庭任务宽度与分销或部署能力上的位置。

评分是方向性分析判断,不是厂商报告指标。家庭任务宽度反映保留证据中对家庭操作或协助的支持;部署能力反映装机基础、资本、渠道或可见商业进展。

[CP015, CP019, CP022, CP029, CP033, CP034]

3.2 能力、定价姿态、GTM 与信任对比

这一组公司的能力宽度仍不均衡。Sunday、1X 和 Figure 都在宣传多步骤家庭操作,但公开就绪信号差异很大。Sunday 强调真实家庭自主性和专有训练数据。1X 明确销售家务机器人概念,但其保留材料也承认高级推理依赖云端,并仍在努力让 双足操作在家庭内部变得安全。Figure 拥有最深的公开资本栈,并通过 Brookfield 获得异常大的住宅数据入口,但保留 的独立报道仍把家庭测试描述为 alpha 阶段,而工业部署更接近近期落地。 既有玩家和相邻方讲的是另一个故事。Samsung 的 Ballie 不是人形机器人,但已经对应家电控制、家庭陪伴式存在感和既有 智能家居行为。iRobot 展示了当任务足够窄时,规模化家庭机器人长什么样:基于摄像头的建图、避障、用户控制功能,以及 围绕“在信任建立前不过度自动化”的隐私姿态。Agility 和 Boston Dynamics 展示人形机器人先在哪里变现:仓库和工业 物料搬运,因为 ROI、安全边界和客户预算更清晰。定价本身也是竞争信号。本报告引用的通用家用机器人页面中,公开标价 仍未披露,说明包装和价格透明度仍落后于 demo 的雄心。[CP002, CP003, CP005, CP008, CP009, CP010]

功能 / 能力矩阵
竞争对手通用家庭操作当前商业证明家庭优先重心公开来源中的信任 / 隐私姿态买方应得出的判断
Sunday仅有早期真实家庭证据强调真实家庭中的自主性;公开认证细节仍有限Sunday 明确优先家庭场景,但仍要证明现场可靠性和支持能力
1X消费者证明有限;EVE 有企业 / 安防历史安全和有隐私意识的数据收集是明确主题最接近的家庭人形机器人同业,但解决部署仍处早期
Figure商业试点和住宅数据合作家庭操作演示很强;家庭推出仍停留在 alpha如果资本和数据访问更快复利,会成为非常严肃的竞争对手
Agility Robotics家庭使用低仓库工作流证明高仓库安全和 ROI 叙事占主导可为人形机器人商业化纪律提供相邻参照,不是直接家庭同业
Boston Dynamics家庭使用低工业早期采用者中等工业安全系统和产品契合经验明确工程可信度高,但保留证据显示家庭相关性低
Samsung Ballie环境辅助中等;操作能力低概念阶段家庭助手证据家庭助手定位明确;公开推出细节很薄争夺同一买方注意力和家电控制预算的非人形机器人路径
iRobot地面清洁之外较低已出货消费机器人中较高窄任务内较高用户控制、隐私和家庭规则是核心信任和重复居家使用的最重要既有基准
Tailos Rosie家庭使用低酒店清洁中较高商用边界明确提醒我们,窄任务自动化可以先于通用家庭机器人规模化

高 / 中 / 低是基于保留官方和独立来源作出的证据支撑序位判断;矩阵总结的是购买标准,不是经审计的基准测试分数。

[CP002, CP005, CP008, CP009, CP010, CP015]
定价 / 包装比较
竞争对手保留证据中披露的公开价格观察到的包装 / 销售模式保留证据包含的内容含义
Sunday家庭机器人产品营销;这里未保留公开商业条款家务、自主性和训练栈声明公开来源尚未证明大众市场可负担性
1X面向消费者家庭的方向;这里未保留公开标价家务辅助和自主性声明消费者意图可见,但定价不透明让采用测算不清楚
Figure家庭路线图加商业部署;这里未保留公开标价Helix 能力、Brookfield 合作和家庭 alpha 测试叙事资本充裕不能消除消费者定价不确定性
Agility Robotics企业部署和云平台仓库 ROI 和集成叙事对任何未来家庭路径而言,商业动作比经济性更清楚
Boston Dynamics面向早期采用者的工业产品销售流程物料搬运和制造用例工业买方能比家庭更早承受价值
Samsung BallieCES 概念 / 智能家居助手定位家电控制和家庭陪伴功能价格透明之前,渠道实力已经存在
iRobot消费硬件加持续软件更新建图、避障和用户控制改进既有厂商展示了如何建立信任,尽管这里未保留确切价格
Tailos Rosie面向酒店和住宿业的商业部署商用清洁价值主张和非住宅边界窄任务机器人可以靠工作流 ROI 销售,而家庭人形机器人仍隐藏价格

这是披露和包装比较,不是数值 ASP 表。Null 表示本章保留的公开证据没有披露可信标价。

[CP007, CP012, CP016, CP023, CP026, CP029]
FP002: 功能宽度 / 能力地图

序数矩阵展示竞争对手在 Sunday 最关键的买方维度上谁最强,而不是比较原始热度。

高 / 中 / 低 / 未知是来自保留来源、有证据支撑的序数判断。该图用于总结竞争就绪度,与更偏文本的表格比较是不同视角。

[CP010, CP017, CP019, CP020, CP021, CP026]

3.3 切换成本、多归属、分发权力与数据或伙伴入口

近期消费者切换成本低于第一眼所见。家庭可以试用新机器人而不放弃现状,因为既有栈是模块化的:人力劳动、清洁工、 智能音箱、扫地机器人、洗碗机和家庭日常流程都能共存。这让多归属比硬替代更可能发生。因此真正约束不是旧数据锁定, 而是一台机器人是否足够可信、可负担、可支持,并且可靠有用到足以成为家庭正常操作系统的一部分。iRobot 自身转向 用户控制的战略变化很有启发:即便是一个窄型既有玩家,也学到如果用户感觉失控,更多自主性并不自动更好。 因而分发和伙伴入口至少和原始机器人能力同样重要。Figure 与 Brookfield 的关系扩大了住宅数据采集和商业化杠杆。 Samsung 可通过家电、SmartThings 和消费电子渠道进入。iRobot 已在一个重要家庭任务中拥有持久信任和装机基础。 如果家庭经济性有一天足够有吸引力,足以推动下沉,Agility 和 Boston Dynamics 拥有集成和工业客户优势。Sunday 最可信的优势是比多数玩家更早为家庭优化;但只有当其学习闭环在更大渠道和资金更强的对手汇聚到相同工作流前变得明显 更好,这一优势才会复利。[CP026, CP027, CP034, CP035, CP036, CP037]

FP003: 护城河 / 就绪度 KPI

公开代理指标浓缩最可能影响 Sunday 2026 年耐久性的竞争事实。

数值混合公司声称事实、保留新闻报道和分析综合。它们是竞争就绪度的方向性信号,不是经审计的市场指标。

[CP028, CP031, CP034, CP039, CP041]

3.4 护城河耐久性、商品化风险与反向证据

这一品类的反向证据足够强,Sunday 的护城河不能只靠愿景承保。Figure 的家庭努力虽有惊人融资,但仍很早期。1X 之所以有吸引力,正因为它是少数明确 home-first 的同业之一;但其公开记录仍显示云端依赖和持续的安全 / 操作工作, 而不是已经解决的部署经济性。Agility 和 Boston Dynamics 证明,许多资金最雄厚或名气最大的机器人团队目前仍偏好 工业场景,这暗示家庭仍是最难的商业环境。最重要的是,Everyday Robots 说明,即便有 Alphabet 级人才、资本和内部 部署入口,只要可靠性、使命清晰度和客户经济性仍模糊,也不能保证形成商业业务。 这给 Sunday 留下一个清楚但狭窄的护城河故事:如果真实家庭数据采集和训练确实优于有家庭野心的同业能获取的数据, Sunday 可能建立可防守的学习曲线。但在通用家用人形机器人成为大众产品之前,品类也可能向下商品化为更窄助手、既有 家电生态或更便宜的单任务机器人。真正的尽调问题因此不是 Sunday 有没有竞争——显然有——而是它能否在别处的资本、 渠道权力和信任优势压缩窗口前,证明持久的家庭表现优势。[CP006, CP007, CP014, CP018, CP028, CP031]

护城河持久性 / 竞争风险登记表
护城河主张或风险威胁严重性证据缓释 / 尽调追问
Sunday 真实家庭数据循环变得持久Figure 和 1X 也在建设家庭数据引擎,同时 Figure 还获得 Brookfield 住宅访问家庭数据差异化有可能成立,但公开层面尚未证明更优要求提供真实家庭中的并排任务成功率、学习曲线和留存数据
家庭人形机器人能够定价到大众采用保留的通用型家庭页面中,公开标价仍普遍未披露包装可见度落后于能力营销在为采用率背书前,拿到价格、融资、安装和支持条款
用户会信任机器人在家中自主行动iRobot 的经验显示用户仍想保留控制,家庭感知更丰富时隐私担忧会上升即便对更窄的既有厂商,信任和隐私仍是采用门槛要求提供隐私架构、覆盖控制,以及事故或失败率报告
工业优先的头部公司会继续留在家庭市场之外Agility 和 Boston 可能在硬件与经济性改善后下探市场它们当前重心在工业场景,但技术底座可以迁移跟踪路线图表态、合作伙伴关系和组件成本下降
既有生态无法压缩 Sunday 的窗口期Samsung 和 iRobot 已经掌握更强渠道、装机基础或消费者注意力分销权力集中在早期家用机器人创业公司之外评估 Sunday 时看自有性能,不要假设渠道中立
通用助手机器人会自然商业化Everyday Robots 即便有 Alphabet 支持和内部部署入口,仍然关停即使演示跑通,可靠性、任务清晰度和单位经济性也可能失效在假设规模化前,要求运营指标、支持负担和窄用例证明

严重性标签是基于留存公开证据作出的分析判断。该登记表有意保持选择性,聚焦后果最重的竞争风险。

[CP018, CP021, CP028, CP031, CP032, CP034]

3.5 图表

Chapter 04

04财务

4.1 收入模式、定价姿态与直接 GTM 动作

公开证据支持消费者家用机器人收入模式,但尚未支持已披露的变现时间表。Sunday 直接面向忙碌家庭营销 Memo,官网和 beta 材料要求家庭申请访问,而不是把需求导向企业买家、保险方或渠道伙伴。最强商业化证据是 Founding Family Beta: 申请于 2025 年 11 月开放,50 个家庭将收到单独编号的机器人和直接支持;2026 年 3 月材料称公开部署从 beta 开始, 独立报道则称兴趣已达到至少四位数等候名单,且申请者可能有数千人。这是有用的需求信号,不是收入证明。Sunday 仍未 披露 Memo 标价、beta 费用、订阅条款、融资计划,或 beta 设备的会计处理。定价缺口很重要,因为品类调查证据显示 多数消费者不愿为家用机器人支付超过 $5,000;因此 Sunday 最终实现的 ASP 必须跨过有意义的可负担门槛,需求才会 转化为高质量收入。[CI003, CI006, CI007, CI008, CI009, CI014]

收入来源表
来源机制单位 / 定价基础当前价值 / 状态收入质量尽调要求
Founding Family Beta 部署早期采用家庭会在 2026 年末 beta 期间收到单独编号的机器人和直接支持。单户 beta 费用或补贴未披露计划 50 户;申请已开放;价格未披露低 —— 试点经济性可能由补贴、推广或非经常性因素支撑提供 beta 合同、任何收费条款、补贴结构和收入确认政策
未来 Memo 机器人直销Memo 可能会通过 Sunday 自有渠道直接卖给家庭。机器人 ASP、融资或租赁条款未公开产品营销已上线,但未披露公开销售启动条款若存在实际 sell-through 则为中;但大概率是一次性硬件收入披露标价、融资选项、实际 ASP、退货和附加率
支持、软件和技能更新持续帮助、模型改进、技能发布和家庭支持可能与机器人打包。订阅费或服务费未公开beta 材料和产品信息里能看到支持;变现方式未披露若可拆分则为中;否则就是打包收入并稀释利润率拆分服务与软件定价、留存和附加率
数据采集生态Memory Developers 和 Skill Capture Gloves 提升产品性能与训练速度。未披露外部客户定价仅是战略投入;没有第三方数据收入证据作为独立收入证据为低说明是否存在任何开发者、合作伙伴或数据授权变现
当前已确认收入 / ARR公开披露门槛null留存来源未发现公开收入、ARR 或付费客户数字Unknown按来源提供月度收入历史、付费客户数、积压订单或预约转化

该表把部署活动和实际变现拆开。Null 表示留存公开记录不支持披露运行日数字。

[CI003, CI007, CI008, CI009, CI030, CI039]
定价 / 变现表
产品 / 信号公开价格或单位来源类型解读仍未知的事项
Memo 标价null官方产品页和发布页Sunday 尚未公开发布 Memo 标价。实际 ASP、折扣、硬件融资条款以及税费或安装费仍未披露。
Founding Family Beta 商业条款null官方 beta 和发布材料beta 已公开,但未披露费用、押金或租赁结构。家庭是否付费、获得补贴,或依据测试协议参与仍未知。
购买模式还是订阅37% 的消费者偏好一次性购买;24% 偏好订阅(品类调查)Altman Solon 调查从品类视角看,家庭可能把家用机器人当作大型耐用品购买,而不是企业软件。Sunday 尚未披露 Memo 会一次性出售、融资、租赁,还是与服务打包。
家用机器人的可负担上限69% 不愿支付超过 $5,000;25% 在任何价格下都不愿支付Altman Solon 调查消费者支付意愿看起来远低于当前高端人形机器人的预期。Sunday 尚未披露目标 ASP,也未说明能否靠融资或低成本形态弥合差距。
消费机器人可比价格带中端机器人 $300-$499 MSRP;高端机器人 $500+ MSRPiRobot FY2024 业绩当下公开消费机器人价格仍远低于通用家用机器人的潜在成本结构。该可比项不能直接推导 Sunday 定价、毛利率或功能对等性。

这里的公开价格点是品类或可比信号,不是 Memo 标价。Sunday 未披露可信商业条款时,Null 值保持不变。

[CI008, CI018, CI019, CI020, CI032, CI036]
FI001: 收入模型桥

展示家庭兴趣如何转化为 beta 部署、未来硬件销售,以及任何最终支持或软件收入——同时明确保留未披露商业条款。

这是一张结构图,不是已披露的会计政策。公开证据能看出可能的商业化步骤,但看不出具体收入确认机制。

[CI003, CI006, CI007, CI008, CI009, CI039]

4.2 成本结构、利润率驱动与公开牵引缺口

即使在大规模商业化之前,Sunday 的成本结构也显得昂贵。公司称自己掌握硬件、软件、数据、训练、模型评估和制造的 全栈;2026 年 3 月帖子称,为支持部署,工程团队扩大三倍、研究团队扩大四倍、数据运营增长五倍。招聘列表仍显示 Redwood City 和远程岗位覆盖 fleet software、data-operations tooling、memory development、data annotation 和 evaluation operations。这意味着至少有四个活跃成本桶:机器人 R&D、硬件与制造、数据采集与标注、居家部署与支持。 公开牵引代理指标真实存在——数据集中超过 500 个家庭、约 1000 万次家庭场景片段、出货超过 2,000 只手套、计划 50 个 beta 家庭、1,000 多等候名单或申请信号,以及 Tracxn 报告的 127 名员工——但这些代理都没有转化为已披露 收入、毛利率、CAC 或回本周期。iRobot 是最清楚的公开可比,说明这为什么重要:其 FY2024 结果仅显示 20.9% GAAP 毛利率、$26.6 million 库存相关费用、为刺激 sell-through 而做促销,并在大幅裁员后启动战略评估。iRobot Q3 2025 发布让警示更强:收入低于预期,现金降至 $24.8 million,库存仍有 $140.9 million,管理层称没有额外资本来源。 官方 Roomba 商店价格大致从 $250 到 $1,400,并提供消费者融资,这凸显主流家用机器人价格架构仍多受约束。Sunday 未来可能跑出更好结果,但公开记录尚未显示风险更低的消费者机器人成本模型。[CI004, CI005, CI010, CI011, CI012, CI013]

单位经济性表
指标数值 / 状态置信度重要性尽调要求
收入 / ARRnull没有标准收入数字,投资人无法检验规模、增长或资本效率。提供 2025-2026 年按来源拆分的月度收入,以及任何已签约但未确认的 beta 收入。
毛利率null毛利率决定 Sunday 更像高端家电卖家、重服务运营商,还是介于两者之间。提供硬件 BOM、制造成本、支持成本和按队列拆分的毛利率。
CAC / 回本周期 / 销售效率null直接面向消费者的机器人业务,效率高低取决于付费获客、支持和转化,可能从好看到极差。披露付费获客支出、beta 转化漏斗、CAC、回本周期和留存。
需求代理指标至少 1,000 个 waitlist 名字;据报道有数千名 beta 申请者说明消费者确有好奇心,但不能证明支付意愿或转化。分享申请者到合格家庭的漏斗、是否收取押金,以及付费转化假设。
成本基础代理指标Engineering 3x、Research 4x、Data Ops 5x;据报道有 >70 名工程师和研究员;127 名员工招聘和团队规模是公开可见、最清晰的运营费用强度指标。提供当前按职能拆分的员工数、平均全包成本,以及相对现金跑道的招聘计划。
服务交付负担50 个 beta 家庭加直接支持;正在招聘数据运营和评估运营家庭部署在规模化前就可能增加现场支持、机队维护和异常处理成本。拆分每户家庭的上门支持分钟数、远程运营占比、更换成本和故障支持负担。
消费者可负担上限多数受访消费者支付意愿封顶在 $5,000 或以下Memo 经济性最终必须装进现实家庭预算,而不能只靠演示热度。提供目标 ASP 与融资假设,并与观察到的支付意愿队列对比。
公开家用机器人毛利率可比项iRobot FY2024 GAAP 毛利率 20.9%公开消费机器人可比项说明,即便规模化,硬件毛利也可能很薄。解释 Sunday 为什么能在结构上跑赢公开消费机器人毛利率。

公开记录沉默处保留 Null。非 Null 行只是代理指标,不等于 Sunday 完整单位经济性披露。

[CI006, CI009, CI011, CI012, CI013, CI018]
FI002: 单位经济桥

展示 Memo 要成为有吸引力的消费级机器人业务,哪些价值与成本驱动因素必须对齐。

这张图是定性的,因为 Sunday 尚未披露价格、CAC、毛利率或支持成本数据。公开可比公司与调研证据只作为方向性约束。

[CI013, CI017, CI018, CI021, CI030, CI031]
FI003: 财务估计区间

展示截至 2026 年中,对 Sunday 最相关且有来源支撑的公开财务与采用代理指标边界。

融资区间混合了首轮、最近一轮和累计披露资本。支付意愿区间来自 Altman Solon 品类调研回应,不是 Sunday 定价。团队规模区间使用了公司 2025 年 11 月自报的工程人数、2026 年 3 月独立来源给出的 >70 名工程 / 研究人员数据,以及 Tracxn 的 2026 年 3 月员工数。毛利率区间是公开 iRobot 可比参照,不是 Sunday 指引。

[CI002, CI012, CI018, CI019, CI027, CI032]

4.3 资本充足性与融资依赖

资本充足性应从已披露融资时间线向前看,而不是从融资头条热度向后推。Sunday 的启动融资是 2025 年 11 月的 $35 million,2026 年 3 月 Series B 又以 $1.15 billion 估值加入 $165 million;因此 Tracxn 显示两轮合计 融资约 $200 million。对一家仍没有公开价格或收入披露的公司来说,这是可观资本,但不足以解决充足性问题。同一批 2026 年 3 月材料把新资本用于真实世界部署、更快迭代,以及更大的工程、研究和数据运营团队。这些用途恰恰会加速 烧钱。没有保留来源披露手头现金、月度烧钱、跑道月数、债务额度、契约限制、客户预付款或项目融资式义务。由于 Sunday 正从 demo 走向现场部署,下一次融资触发因素可能不再主要取决于 B 轮头条规模,而取决于 2026-2027 年 beta 运营、 支持和制造在实践中有多贵。[CI001, CI002, CI011, CI012, CI013, CI027]

资本充足性表
项目公开数值 / 状态置信度重要性尽调要求
启动融资2025 年 11 月 $35M 融资,由 Benchmark 和 Conviction 支持确立公开部署前的初始股权基础。确认准确交割日期、如有投后估值,以及所有权影响。
Series B2026 年 3 月 $165M,估值 $1.15B,由 Coatue 领投最新重大资本事件,也是投资人兴趣最强的实时信号。提供本轮文件、优先权、董事会权利和 pro rata 结构。
已披露总资本 / 规模代理指标两轮合计融资约 $200M;截至 Mar. 26, 2026 据报道有 127 名员工在公开收入证据出现前,已经有大量资本投入。提供当前股权结构表、完全稀释后股数和最新按部门拆分的员工数。
手头现金 / 月度烧钱 / 现金跑道null这是资本充足性的核心变量,公开层面仍未披露。提供当前现金余额、月度烧钱历史,以及基准 / 下行情景现金跑道。
声明资金用途真实世界部署、beta 推出,以及扩大工程 / 研究 / 数据运营团队暗示公司在为商业化和现场运营融资,而不只是做研究。量化 R&D、制造、支持、销售和营运资本之间的分配。
债务 / 项目融资义务未公开披露债务额度、契约包或项目融资结构隐性义务可能实质改变稀释风险和流动性时点。披露债务、风险债、设备融资、担保,以及任何客户预付款或融资安排。

纳入历史融资只是为了框定未来充足性。公开不可得的现金、烧钱和债务数据,仍是测算融资依赖度的主要障碍。

[CI001, CI002, CI011, CI027, CI028, CI029]
FI004: 资本强度 / 现金流图

展示公开披露的股权融资、全栈建设和 beta 部署如何叠加,在公开收入证据出现前形成融资依赖。

这张图是方向性的。公开来源能看出可能的现金来源和用途,但看不出实际现金桥、债务条款,或实现经营性现金流转正所需的跑道。

[CI001, CI002, CI011, CI027, CI029, CI041]

4.4 财务结论与承保阻碍

以当前公开记录看,Sunday 的财务故事高承诺、低可承保。收入质量无法判断,因为没有披露价格、没有付费设备销售证据、 没有客户集中度数据,也没有 beta 或未来部署的流级收入确认政策。利润率路径同样开放:公司可能受益于专有数据闭环和 未来规模,但今天可见模型在证明持久消费者支付意愿之前,就背负沉重 R&D、制造、数据运营和支持义务。最强正面因素是 资本可得性、申请兴趣和异常强的数据规模信号。最大阻碍仍是经济性:实现 ASP、硬件 BOM 和支持成本、按 cohort 的 毛利率、beta 到付费转化、现金消耗、跑道,以及业务究竟会变成高端家电销售、服务很重的机器人产品,还是介于两者之间。 更广文献也支持谨慎:NBER 发现机器人投资仍是设备支出中很小且集中的一部分,学术智能家居和家用机器人采用研究也不断 提出成本、隐私和安全担忧是采用刹车。在这些披露之前,Sunday 应被承保为一个资本密集型家用机器人实验:融资强劲, 但收入质量未证实。[CI008, CI009, CI018, CI019, CI020, CI021]

公开财务缺口表
缺失的私有指标对分析的影响当前公开状态精确尽调路径
Memo 标价和实际 ASP无法建立严肃的需求、利润率或融资模型。未公开披露。获取定价备忘录、融资选项、折扣政策以及实际 beta 或预购经济性。
按来源拆分的收入无法检验 Sunday 到底是硬件销售、服务层、订阅,还是混合模式。未公开披露。要求按月收入桥,拆分硬件、服务、软件和任何试点对价。
毛利率和成本栈无法评估长期单位经济性。未公开披露。要求按队列提供硬件 BOM、制造成本、物流、上门支持、云 / 推理和保修成本。
CAC、回本周期和转化无法把 GTM 效率与其他消费业务比较。未公开披露。提供 waitlist 到 beta 漏斗、付费获客支出、推荐占比、CAC 和回本周期假设。
现金、烧钱和现金跑道没有这些数字,无法测算融资依赖度。未公开披露。提供当前现金、月度烧钱、招聘计划,以及基准 / 下行情景现金跑道桥。
债务、风险债或契约包可能实质改变稀释和流动性风险。未发现公开披露。披露贷款方、金额、契约、抵押品和再融资时点。
Beta 合同经济性缺少这些条款,beta 牵引力无法转化为收入确认或成本回收。申请和支持是公开的,经济性未公开。提供 beta 协议、家庭付款条款、服务级别承诺,以及更换 / 召回义务。
机队可靠性和服务指标支持负担是现实家庭机器人经济性的核心。未发现公开故障率、在线率或服务成本披露。提供利用率、任务成功率、介入率、保修索赔和每户支持分钟数。

每行都列出具体数据请求,因为当前公开记录远不足以支持标准消费硬件或机器人投资评估。

[CI008, CI009, CI029, CI030, CI039, CI040]

4.5 图表

Chapter 05

05产品与技术

5.1 产品定义与家庭工作流

Sunday 公开销售的是一个家庭工作流故事,而不是抽象机器人平台。从官网、发布报道到 beta 材料,Memo 被描述为面向 忙碌家庭的家用机器人,承担清餐桌、刮食物残渣、装洗碗机、叠袜子、摆鞋,甚至拉一杯 espresso shot 等重复家务。 公司称 Memo 应能在不同家庭中开箱自主工作,并明确把价值主张定义为把时间还给家庭,而不是销售窄型工业机械臂。这一 定位很重要,因为它用客户语言定义产品:买家是家庭,job-to-be-done 是减轻家务劳动,服务承诺是居家任务执行,而 不只是实验室表现。可见商业路径仍受控。Sunday 的 Founding Family Beta 是主要部署面,计划 50 个家庭收到单独编号 的机器人和直接支持。Beta 材料还称 Memo 的大脑会持续更新,参与家庭可以为更广训练闭环提出新技能。因此,公开来源 支持一条具体初始工作流——家务、试点部署、支持和反馈——但尚未披露公开消费者结账流程、定价或已发布的可靠性 SLA。[CE001, CE002, CE003, CE004, CE005, CE007]

产品模块 / 资产矩阵
模块 / 资产主要用户 / 操作方当前状态 / 成熟度差异化尽调缺口
Memo 机器人家庭 / 家人已公开营销;计划 2026 年末 beta瞄准一般家务,而不是单一用途家电没有公开价格、结账流程或现场可靠性指标
Skill Capture GloveMemory Developers / 内部数据闭环公司称已交付 2,000+ 个规模以匹配机器人手的几何形态采集人类家庭行为没有公开 BOM、传感器规格表或故障率数据
Memory Developer 网络远程和内部数据操作员公司称采集项目已上线,参与者达数百人让 Sunday 采样真实家庭差异,而不只靠实验室演示参与者经济性、筛选和地理覆盖未披露
Skill Transform内部 ML / 数据管线核心自研管线;声称 90% 转换成功率弥合手套观察和机器人数据之间的 embodiment mismatch未公开独立基准或消融实验
ACT-1机器人策略栈Sunday 第一代基础模型;已公开演示公司称其把长时程操作和地图条件导航结合起来未公开第三方复现或部署 KPI 集
Skill Library机器人软件栈正在扩展;官网称每月新增技能把 Memo 定位成可泛化的技能平台,而非固定流程机器人没有公开发布说明或带版本的能力日志
Beta 机队 + 直接支持Sunday 部署和支持团队面向 50 户家庭的初始运营模型把现场学习与直接家庭反馈、引导式支持结合升级路径、维护模型和更换流程未披露

该矩阵把直接披露的产品资产与公开运营细节缺失处并列;成熟度反映官方和独立来源可见的信息,而非私有尽调。

[CE001, CE006, CE008, CE009, CE011, CE015]
工作流 / 用例表
用户任务当前工作流Sunday 方案可衡量收益限制
餐后清理人工清理盘碟、食物残渣和洗碗机物品演示中 Memo 会清桌、刮掉残渣并装载洗碗机官方演示序列,加上 ACT-1 从餐桌到洗碗机的叙事未公开真实客户家庭中的任务成功率
叠袜子人工分拣并折叠柔软、可变形的衣物ACT-1 演示叠袜子堆释放出超越刚性物体抓取的灵巧度信号未公开吞吐量或恢复率数据
整理鞋子人工整理家中鞋履官网称 Memo 在 2024 年 12 月最早学会整理鞋子说明一个早期窄技能正在演进为后续多技能行为只是早期里程碑;没有持久生产指标
制作 espresso人工装入并锁紧 portafilter,再启动咖啡机ACT-1 演示操作标准消费级 espresso 机器展示双手协调和对扭矩敏感的操作能力没有公开证据证明其能在公司演示之外重复完成
适应新家每个家庭都有不同布局和杂物模式Sunday 称 ACT-1 使用 3D 地图在未见过的家庭中导航直接回应家庭之间差异这一运营变量未公开设置时间、地图创建工作流或错误处理方式

用例来自官方演示和技术文章;收益是证据信号,不是经审计的客户结果。

[CE003, CE004, CE007, CE019, CE020, CE021]
FE002: 客户工作流 / 运营流

可见客户流程从一项家务开始,经过 Memo 的导航与操作循环,最后进入 beta 反馈和新技能收集。

[CE001, CE003, CE004, CE008, CE009, CE010]

5.2 运营架构与技术差异化

公开运营模式围绕数据采集和迭代速度展开。Sunday 称自己掌握软件、硬件、数据、训练、模型评估、硬件设计和制造全栈; 其最详细技术帖子认为,家庭机器人真正瓶颈不只是硬件或算力,而是缺少互联网规模的操作数据。Sunday 的答案是一串相连 资产:Skill Capture Glove 记录人类家庭行为,Skill Transform 把手套观测转成机器人等效数据,ACT-1 作为学习到的 策略栈,再加上 fleet-plus-data-operations 闭环把新数据推回系统。公司称手套和 Memo 的手共享相同几何与传感器布局, Skill Transform 转换成功率达 90%。公司还声称 ACT-1 能执行长时程餐桌到洗碗机行为,利用 3D 地图在未见过的家庭中 导航,并处理叠袜子和准备 espresso 等高灵巧任务。Sunday 的差异化故事不仅靠已披露生产架构图,也由创始人脉络增强。 公司页面、Tony Zhao 个人资料和公开 ALOHA / Diffusion Policy 材料把创始人与低成本遥操作、动作分块、基于扩散的 策略学习和大规模灵巧数据采集等开放机器人工作相连。这条脉络不能证明 Memo 复用了那些确切代码库,但说明 Sunday 正从机器人社区已经读得懂的研究传统出发建设。[CE006, CE011, CE012, CE013, CE014, CE015]

技术 / 运营架构表
层级 / 流程 / 组件角色依赖风险
Skill Capture Glove在家庭环境中采集人类示范手套 / 机器人手几何匹配,以及 Memory Developer 参与公开材料未披露传感器细节、校准负担或手套故障模式
家庭数据采集网络大规模采样真实家庭差异数百个家庭、数千只手套和数据运营人员覆盖偏差、标注质量和贡献者留存未公开
Skill Transform把手套观察转换成机器人等价数据计算机视觉和运动学对齐管线90% 转换率声称来自内部,未经独立基准验证
ACT-1 策略 / 模型层执行长时程家庭操作训练数据、模型评估闭环和 Memo 硬件执行没有针对真实家庭失败案例或维护负担的外部基准
3D 地图条件导航让 Memo 在未见过的家庭中找到任务地点准确环境地图和泛化训练数据地图工作流、刷新节奏和兜底行为未披露
Memo 机身 / 末端执行器在家庭空间里安全执行操作滚动底座、臂手设计、力敏感能力和硅胶外壳公开文案未披露负载、触达范围、续航或服务间隔
机队、数据运营与评测运营闭环支撑部署、再训练和质量控制机器人机队管理、ML 评测、标注和支持团队未公开事故指标、评审节奏或发布治理政策

架构行基于已披露组件和运营角色;未公开的内部细节明确列为风险,而不是推断成规格。

[CE011, CE014, CE015, CE016, CE018, CE020]
FE001: 产品架构图

Sunday 公开展示的技术栈从人类在环的家庭数据采集开始,经转换层和模型层,最终落到 Memo 部署与机队迭代。

[CE009, CE011, CE014, CE015, CE016, CE018]
FE003: 关键依赖图

Sunday 公开产品逻辑依赖真实家庭数据采集、数据转换、模型训练、机队运营,以及外部信任框架;家用自主场景下,这些框架尚未完全成文。

[CE006, CE009, CE015, CE016, CE018, CE026]

5.3 部署、信任与剩余产品风险

部署证据在改善,但公开信任证据仍落后于家用机器人的雄心。Sunday 2026 年 3 月帖子称,公司已从 demo-first 叙事 转向真实世界部署;招聘页用机器人 fleet management、ML 数据与评估、语音交互、memory development、标注和 eval ops 等可见岗位支持了这一转变。这说明运营模式现在包括现场支持和质量闭环,而不只是模型训练。与此同时,公开信任面仍 很窄。Sunday 有一份网站隐私政策,覆盖 beta 或 newsletter 收集、沟通、设备 / 网络信息和网站活动数据,但已访问材料 没有解释机器人传感器留存、居家视频处理或车队数据治理。外部标准也不会自动替 Sunday 补上缺口。ISO 13482 覆盖个人 护理机器人,NIST 强调测试方法和经验证的表现指标,IEEE 2026 年报道则认为家用机器人标准仍缺少真实家庭中关系性安全 的强制测试方法和执行机制。外部怀疑同样相关:Ken Goldberg 关于灵巧性和机器人数据缺口会拖慢家庭管家时间表的批评, 直接指向 Sunday 的承保挑战。结果是一款看起来技术差异化、运营严肃的产品,但在现场可靠性、供应商细节和安全验证上 仍披露不足。[CE008, CE009, CE025, CE026, CE035, CE036]

信任 / 质量 / 合规表
控制 / 信号状态范围缺口
隐私政策 + Cookie 通知已发布覆盖 beta / 通讯订阅信息收集、沟通、设备 / 网络数据和网站活动未说明机器人传感器留存、家庭视频政策或机队数据治理
滚动底座稳定性声明公司声明发布材料称 Memo 断电时仍保持稳定,并避开双足跌倒风险未发布测试协议或量化稳定性数据
软硅胶外壳声明公司声明发布材料称机器人柔软,适合家庭生活空间未公开材料标准、可清洁性或冲击测试披露
直接支持的 beta 运营已发布50 个 beta 家庭会收到单独编号的机器人和直接支持支持时段、更换政策和待命升级机制未公开
数据运营 / 评测岗位配置已发布岗位信号招聘信息显示机队、标注、ML 评测和评测运营岗位岗位说明暗示流程存在,但未给出量化质量结果或审计节奏
适用的个人护理机器人框架外部标准已存在ISO 13482 覆盖移动仆从机器人及相关个人护理机器人的危险该标准不能证明 Sunday 合规,也缺少完整冲击数据
家庭自主标准缺口重要外部警示IEEE 称现行修订仍缺少面向关系型家庭安全的强制测试方法和执行机制Sunday 未披露用于弥合该缺口的公开认证或合规计划

本表列出的控制项混合了 Sunday 官方披露和相关外部框架;未发布某项控制,并不代表内部不存在,只说明在已访问公开界面上不可见。

[CE009, CE023, CE024, CE026, CE035, CE036]
路线图 / 发布 / 开发阶段表
日期 / 阶段功能 / 里程碑状态含义来源
Dec. 2024Memo 学会用一只手臂整理鞋子已完成里程碑最早披露的能力节点,锚定从窄任务走向更广家务的进展主页
Oct. 2025Memo 学会处理袜子、玻璃器皿和 espresso已完成里程碑公开部署前,能力宽度已有可见扩展主页 / 技术页
Nov. 19, 2025退出隐身模式,Founding Family Beta 开放申请已完成公告标志着公司从内部构建转向外部试点招募发布报道
Late 2026首批 50 个家庭收到 beta 机器人和直接支持已规划公开部署仍是试点阶段,不是规模化消费者供给Beta 页 / 发布报道
Mar. 12, 2026公司称正从演示迈向真实世界部署当前战略姿态说明运营重心放在现场投放、迭代速度和支持能力上B 轮文章
当前公开路线图beta 后带日期的发布里程碑未披露投资人能看到能力快照和招聘信号,但看不到有时间表的 GA 路线图已访问公开界面观察

路线图只从有日期的公开里程碑和明确部署表述中重建;未发现更完整的发布日历。

[CE007, CE008, CE009, CE021, CE025, CE040]
FE004: 产品成熟度 / 能力图

Sunday 在数据采集和家务演示上显示出真实成熟度,但公开信任、可靠性和供应披露成熟度仍低得多。

[CE007, CE008, CE018, CE019, CE020, CE021]

5.4 图表

Chapter 06

06客户

6.1 客户分层与市场进入路径

Sunday 的公开客户故事比典型规模化消费者产品窄得多。公司始终把 Memo 定位给想获得重复家务帮助的忙碌家庭和家庭成员, 任务包括洗碗、洗衣、整理和清桌。在本报告审阅的公开界面上,买家、用户和付款方似乎都收束为同一个家庭单位:Sunday 没有披露雇主赞助、养老照护、酒店或企业动作,也没有按地理、收入或住宅类型细分家庭。这种狭窄并不天然负面,但意味着 今天可见市场证明集中在单一消费者楔子,而不是分散在多个买家原型上。 更重要的区别,是几个容易被混同的家庭侧 cohort。Sunday 的训练数据家庭、Memory Developers、等候名单成员、beta 申请者,以及最终 Founding Family 家庭,都靠近客户旅程,但并不是同一件事。500 多个家庭和约 1000 万次家庭场景 片段支撑数据护城河故事;它们不能证明 Memo 的付费使用。报道中的 1,000 人等候名单和数千份 beta 申请显示早期兴趣, 但仍是漏斗顶部需求信号。最清楚的部署 cohort 是 2026 年末 Founding Family Beta,届时 50 个家庭应收到单独编号 的机器人和直接支持。市场进入路径也显得高度集中。Sunday 公开材料强调直接 beta 申请、直接语音或 app 控制,以及 公司主导支持,而不是零售、经销商或伙伴主导分发;这带来短期聚焦,也把 onboarding 和服务执行集中在 Sunday 内部。[CU001, CU002, CU003, CU004, CU007, CU009]

客户分群表
分群 / 队列购买者 / 用户 / 付款方使用场景规模 / 证据战略价值缺口
忙碌家庭 / 家庭用户同一家庭很可能同时购买、使用和付款餐具、洗衣、整理、清桌等省时家务主页和发布材料中的官方定位定义核心消费者切入点,让产品聚焦实用家庭劳动未披露收入、住房类型或地理分群
Founding Family Beta 家庭早期采用家庭真实 beta 使用加产品反馈计划 late 2026 覆盖 50 个家庭,并提供直接支持首个真实部署队列,也是近期最强客户证据仍是预期队列;未披露具名家庭或定价
Beta 申请者 / waitlist感兴趣的潜在用户申请加入早期部署2026 年 3 月报道显示数千份申请和约 1,000 人 waitlist显示大规模出货前的漏斗顶端需求不是活跃客户,也不是留存信号
训练数据家庭 / Memory Developers数据贡献者,不是 Memo 买家记录真实家庭日常,用于模型训练公司称已有 500+ 个家庭,以及数百个家庭或 Memory Developers对产品学习和家庭泛化声明很重要这些家庭不能证明付费机器人采用
未来规模化家庭尚未公开beta 后更广泛的消费者推出B 轮和 Beta 页面称首批队列之后部署会扩大如果上线引导跑通,这是长期市场奖池未公开 GA 日期、定价或转化数据
渠道 / 合作伙伴路径未披露合作伙伴或经销路线公司直接运营获客和支持官方界面强调直接 beta 申请、app 控制和支持短期降低渠道不确定性让 onboarding 和服务能力依赖 Sunday 自营

本表把真实部署证据、训练数据和兴趣信号分开;缺少某项分群细节,意味着公开来源集未披露。

[CU001, CU002, CU003, CU004, CU009, CU013]
客户披露和支持界面表
界面公开披露内容对客户的含义新鲜度缺口
主页 / 公司页Memo 面向忙碌家庭和家务解释产品服务谁当前官方界面无客户数量或定价
Beta 项目页语音 / app 排程、新家庭自主性,以及面向 Founding Families 的直接支持显示早期家庭会如何与 Memo 互动当前官方界面无具名家庭或推出地域
B 轮文章公司称正从演示转向部署说明运营重心放在真实家庭2026-03-12无出货量
隐私 / 条款 / Cookie网站层面的法律页面在推出前已存在显示基本上线引导和法律准备度2025-09-03 生效;2026 年站点地图仍列出无机器人专属传感器或服务政策
Journal / 公开新闻室官方叙事集中在研究和融资,而非客户案例削弱后续买家的独立社会证明2026 年 6 月 Journal 仍稀疏无公开案例库或 FAQ 档案

本表记录面向客户的披露界面,而不是采用指标。Sunday 仍处于规模化前,很多承保难点来自尚未公开的信息,因此该表有用。

[CU004, CU005, CU010, CU020, CU021, CU026]
FU001: 客户旅程图

Sunday 可见客户旅程从广泛家庭兴趣走向申请、直接入门、定时任务使用,再进入产品反馈;还不是一个已验证的续约循环。

[CU001, CU005, CU006, CU009, CU011, CU013]

6.2 采用轨迹与公开证明

公开证据里的采用轨迹是真实的,但仍处在商业化前。Sunday 的发布材料和多次转载都称,产品训练数据来自 500 多个家庭、约 1000 万段家庭情境。这一点重要,因为它证明产品在部署前已接触真实家庭,但它仍是产品开发指标,不是客户指标。可见轨迹的下一步是 Founding Family Beta。Sunday 官方 beta 页面和多篇发布报道显示,申请已于 2025 年 11 月开放,计划在 2026 年底部署到 50 个家庭,并提供直接支持。到 2026 年 3 月,Sunday 的 Series B 信息已经从演示口径转向「真实世界部署」;第三方报道还补充了两个漏斗顶部需求信号:据报道约 1,000 人的 waitlist,以及「数千份」 beta 申请。这些信号有价值,因为它们说明产品在出货前已吸引关注,但它们不能证明接受、激活、使用强度或复用。 因此,具名客户证明是本章最弱的一环,而且弱点是实质性的,不是写法问题。保留的来源集中,没有发现任何具名付费家庭、公开生产用户或独立客户案例。最强的公开证明,是存在一个具体的首批队列,而不是任何家庭已经以生产深度使用 Memo 的证据。也就是说,Sunday 的公开客户证明应被描述为由需求信号和 onboarding 准备支撑的前瞻性试点证明,而不是规模化采用。投资人不应把训练数据家庭、waitlist 规模和计划中的 50 户队列当作同一种证据。三者分别支撑客户故事的不同部分,只有最后一项接近真实部署证明。[CU007, CU008, CU009, CU010, CU011, CU012]

客户增长 / 采用轨迹表
指标数值日期来源 / 依据置信度含义缺失分母
训练数据家庭500+ 个家庭2025-11-19发布报道和技术页Sunday 在 beta 部署前已有家庭场景暴露不等同于付费 Memo 部署
家庭片段约 1,000 万个片段2025-11-19公司声明的训练数据指标暗示产品背后有大规模真实家庭数据集无独立审计或任务成功率映射
Founding Family 队列计划 50 个家庭Late 2026Beta 项目和发布报道首个公开部署队列受到严格控制未披露筛选标准或地域拆分
候补名单信号约 1,000 人2026-03-12TechCrunch 和 Yahoo 援引 Bloomberg显示出货前需求信号未披露候补名单到获批 beta 家庭的转化率
申请量数千份 beta 申请2026-03-12GlobeNewswire 和 Tech Funding News暗示兴趣范围比具名队列更广未披露独立家庭数量或重叠情况
部署姿态今年秋季 / 年内向用户推出2026-03-12Sunday B 轮文章和新闻稿公司叙事已从演示转向真实使用暂无公开出货量或激活数

这是公开证据轨迹表,不是收入漏斗。数值混合了公司声明、第三方报道和明确缺口;缺失分母标出采用情况仍无法量化的位置。

[CU007, CU008, CU009, CU010, CU011, CU013]
具名客户证据表
客户 / 队列分群部署 / 使用场景生产 vs 试点结果 / 证据限制
Founding Family Beta 队列早期采用家庭late-2026 家庭内 Memo 使用,并配直接支持试点 / 预期Sunday 披露过最具体的公开部署队列暂无具名家庭或已出货机器人
Beta 申请者 / 候补名单感兴趣的消费者出货前的申请或候补名单信号部署前需求显示首批 50 个家庭之外也有兴趣不能证明真实使用或付款
训练数据家庭数据贡献家庭借助 Skill Capture Glove 捕捉真实家庭日常不是 Memo 部署显示 Sunday 出货前已有家庭语境数据这些家庭不是机器人客户
公开客户法律界面未来 beta 家庭隐私、条款和 Cookie 页面已在推出前存在就绪信号,不是部署显示 Sunday 已搭好部分上线引导 / 法律界面仍无机器人专属 SLA、保修或续约细节

公开客户证据不完整,因为 Sunday 未披露具名付费家庭。本表列出最强可用队列,同时避免把需求、数据收集和真实部署混为一谈。

[CU009, CU013, CU014, CU015, CU016, CU017]
FU002: 采用 / 部署漏斗

漏斗在认知和申请环节最强,在获准队列激活环节较弱,重复使用最弱,因为目前没有公开使用率或续约指标。

[CU005, CU009, CU010, CU011, CU012, CU013]
FU003: 客户证据矩阵

公开证据质量在计划中的 beta 队列上最好,在耐久性上最差,因为所有可见信号都停在收入、续约或具名客户证据之前。

[CU007, CU008, CU009, CU013, CU020, CU021]

6.3 耐久性与留存可见度

Sunday 的耐久性图景主要是证据缺口。公司已经发布面向客户的法律页面——隐私、条款和 cookies 页面; beta 页面也披露了配套 app、语音排程和直接支持。这些都是部署前准备信号,但不是续约证明。保留来源未披露 Memo 定价、合同期限、取消权、保修条款、活跃家庭数、重复任务频率、支持响应时间、满意度评分、流失、NRR 或 GRR。因此,最诚实的说法是:公开耐久性证据还不存在,因为首批家庭队列尚未被公开证明已规模化运行。 家庭内数据处理也存在更具体的信任缺口。Sunday 的网站级隐私和条款页面说明,公司在 rollout 前已搭好法律基础设施;对于家庭机器人,这一点重要。但这些文件没有公开解释面向未来 beta 家庭的机器人专属传感器留存、室内音频或视频处理、维护升级,或更换 SLA。这个缺口重要,因为在家用机器人里,信任和耐久性相互绑定:家庭可能愿意加入 waitlist,但如果服务承诺和数据实践仍不透明,仍可能拒绝激活或续约。现阶段,正确的分析姿态应当保守。Sunday 披露的信息足以说明客户准备工作已经启动,但还不足以支撑长期留存或 beta 后转化质量。[CU005, CU020, CU021, CU022, CU023, CU026]

留存 / 重复使用 / 满意度表
指标数值 / 状态分群置信度尽调追问
NRR / GRR / churn(留存 / 流失)null所有 Memo 客户一旦出现付费队列,要求提供 cohort 留存、续约、流失和收入留存指标
合同期限 / 续约条款nullFounding Family Beta 及后续商业队列要求提供 beta 协议期限、取消权和 beta 后商业迁移条款
重复使用频率nullBeta 家庭部署开始后,要求提供每日任务数、周活跃家庭和重复排程指标
客户满意度 / NPSnull真实使用家庭要求独立评价、支持 CSAT 以及 onboarding 后调查输出
隐私和支持耐久性通用法律页面存在,但缺少机器人专属运营细节未来 beta 家庭要求传感器数据留存政策、升级工作流、更换 SLA 和维护节奏

Null 表示留存的公开记录未披露该指标。唯一非 null 行是治理信号,不是量化耐久性 KPI。

[CU020, CU021, CU022, CU023, CU026, CU035]

6.4 扩张、集中度与采用摩擦

公开扩张模型容易描述,却难以承销。Sunday 自己给出的顺序是:先做宽泛的家庭定位,再收集 waitlist 和申请,然后管理一个高度受控的 50 户 beta,最后希望在反馈和数据改善产品后扩大 rollout。对于一家早期家用机器人公司,这是一套合理的 land-and-expand 假设,但背后没有公开转化数学。公司没有披露多少申请会变成获批家庭、多少获批家庭会激活、多少家庭会重复使用 Memo,也没有说明是否会有家庭转化为付费长期客户。集中度同样不透明:没有公开队列结构或收入结构,无法量化头部客户风险;但小规模早期队列仍可能主导学习和任何近期经济性。 外部证据也说明,从好奇心跃迁到耐久采用会很难。UNSW 强调隐私风险、隐藏远程劳动,以及家庭并不是为通用机器人设计的事实。The Robot Report 引述 Ken Goldberg 称,teleoperation 仍是持续的拐杖。Altman Solon 的调研显示品类兴趣存在——65% 的受访美国消费者表达兴趣——但认知度低,安全和隐私担忧真实存在,价格承受度也远低于高端机器人硬件通常的起点。这些并非 Sunday 的公司特定失败,但确实意味着,Sunday 的客户扩张更可能受 onboarding 质量、支持运营、信任和可负担性约束,而不是受初始媒体热度约束。简言之,Sunday 拥有一个可信的家庭 wedge 和真实的发布前需求信号,但其公开客户故事仍停在投资人通常需要看到的耐久、可规模化采用证据之前。[CU018, CU019, CU025, CU027, CU028, CU029]

扩张与集中度风险表
扩张驱动 / 风险当前公开信号影响重要性尽调路径
先落地再扩张路径候补名单 -> Founding Family Beta -> 更广泛家庭推出如果 beta 转化为可背书家庭,规模可能很大扩张只有概念描述,尚无转化率或复购数据度量要求候补名单到获批、激活和 beta 后转化数据
大客户集中度未公开账户结构或家庭收入集中度披露Unknown一个很小的早期队列可能主导使用学习和任何早期收入家庭上线后,要求队列结构、地域和收入分布
渠道依赖公司直接运营 beta,未披露经销商或零售动作短期高度依赖 Sunday 运营获客、onboarding、支持和问题解决看起来都集中在公司侧要求人员配比、支持覆盖窗口和渠道路线图
信任和 onboarding 摩擦外部来源显著关注隐私、安全和远程操作可能拖慢从好奇到活跃使用的转化家庭买家需要的信任程度高于 waitlist 注册所暗示的水平要求退出原因、异议处理和安全事故预案
定价和可负担性Memo 未披露定价,同时品类调研显示消费者对 $5,000 以上价格强烈抵触重要采用风险消费者支付意愿可能跟不上硬件经济性要求目标定价、融资路径和支持成本绑定方式

本表聚焦可能推动或阻碍扩张的机制;不编造集中度数字,而是列出承保这些判断所需的缺失证据。

[CU018, CU019, CU023, CU025, CU028, CU030]

6.5 图表

Chapter 07

07风险

7.1 风险概览:Sunday 正试图在品类控制体系成熟前,把一个亲密的家庭产品工业化

Sunday 的公开材料在许多机器人初创最弱的地方反而最强:它有具体部署时间表、可见 beta 队列,以及清晰表述的数据策略。公司称 Memo 的 beta 将于 2026 年底启动,首个公开队列只有 50 个家庭并配有直接支持,公司也已融资 $165 million,专门用于从演示走向真实世界部署。这些是有意义的积极因素,因为它们说明公司理解,家庭机器人只有在真实家庭里能工作才有意义,而不是只在搭好的演示场景里好看。Sunday 的数据故事也比典型发布前机器人公司更可信。公开材料指向来自 500 多个家庭的训练数据、约 1000 万段家庭情境,以及向 Memory Developers 发出的 2,000 多副 gloves。这确实缓解了品类最大问题:缺少混乱的真实家庭数据。 问题在于,Sunday 的公开承诺直接撞上三项外部现实校验。第一,家用机器人安全和标准体系仍明显不完整。IEEE Spectrum 2026 年 5 月报道称,家用机器人 ISO 更新仍缺少可执行的测试方法,以及面向家庭自主性的治理机制;NIST 则强调,先进机器人必须先有严格验证、指标和协议,才能在预定环境中被信任。第二,隐私和责任义务扩张速度快于 Sunday 公开法律细节。Sunday 有网站级法律页面,但外部法律和监管来源显示,AI 驱动机器人可能触发生物识别、音频、地理位置和儿童数据问题;责任也可能尴尬地落在制造商、运营方和软件提供商之间。第三,品类经济性先例偏弱。Altman Solon 的调研显示消费者有好奇心但支付意愿低;iRobot 近期的流动性、库存和 going-concern 问题则说明,消费机器人叙事能多快转成资本效率故事。换句话说,Sunday 最高严重度的风险不是缺乏想象力,而是公司可能在周边经济性、安全框架和支持模型准备好规模化之前,就已经进入家庭。[CR001, CR005, CR011, CR014, CR015, CR037]

7.2 法律、隐私和安全风险偏高,因为 Sunday 在公开合规细节成熟前,就要把联网机器人带进家庭

Sunday 已经有法律框架的开端,但还没有达到足以完成尽调的机器人专属披露水平。公开法律页面是真实存在的:隐私政策、服务条款和 cookies 政策均已上线,并自 2025 年 9 月 3 日起生效。隐私政策称 Sunday 会收集信息以提供、维护和改进服务,并明确考虑从 newsletter 和 beta 注册中收集姓名、邮箱和电话信息。 beta 页面进一步显示,Memo 将通过直接语音互动和配套 Sunday app 安排任务。这些细节重要,因为它们证明产品不是一台简单离线家电,而是一个计划在亲密家庭环境中运行的联网系统,并通过必然触及用户信息和家庭日常的软件界面互动。 公开记录仍未展示的内容同样重要。已审阅的 Sunday 材料没有公开列举 Memo 专属认证、保修、召回流程、泄露响应承诺或家庭传感器留存规则。这个缺口重要,因为外部法律背景正在收紧。FTC 指引强调,公司必须兑现隐私承诺,并维持与所持数据相匹配的安全水平;COPPA 材料则显示,在家庭内部,围绕儿童、音频、视频和地理位置的问题会很快变成合规问题。Cyber Law Monitor 对机器人更直接:视频、音频、地理位置和生物识别数据可能触发州隐私和生物识别法规、最小化义务、留存要求和监管通知义务。安全法同样不是缺席,而是不完整。ISO 13482 覆盖个人护理机器人,但 IEEE Spectrum 2026 年报道称,当前标准框架仍未把家用自主性知识转化为带有测试方法和治理机制、可执行的真实家庭互动规则。Hill Dickinson 的责任分析又补上另一半问题:如果机器人伤人或处理敏感数据不当,制造商、运营方和软件栈之间可能争执责任归属。对 Sunday 来说,结论很直接。法律风险不是部署后再解决的边角问题;任何想住进厨房、餐厅和有儿童家庭的机器人,都必须把它纳入产品市场匹配。[CR008, CR021, CR022, CR023, CR024, CR025]

监管 / 法律风险登记表
风险当前公开证据可能性影响缓释成熟度剩余暴露尽调路径
家庭隐私和生物识别合规Sunday 已有上线的法律页面和 app / 语音交互模型;FTC 与机器人专项法律指引也表明,家用机器人可能触发隐私、生物识别、留存和安全义务。低-中机器人专项数据治理细节仍未公开。要求提供机器人专项数据清单、留存时间表、供应商图谱、生物识别同意分析,以及事件响应手册。
国内安全标准缺口ISO 13482 已存在,但 IEEE 表示,国内自主性标准更新仍缺少可执行限制、测试方法,以及人机交互治理。Sunday 可能在品类形成成熟外部保障规则前就进入家庭。获取面向 Memo 的内部安全论证、第三方测试、人因验证,以及认证路线图。
事故后的责任分配Hill Dickinson 指出,类人机器人造成伤害或滥用数据时,制造商、运营方和软件提供商之间的责任仍未厘清。中-高家庭事件发生后,合同、云服务和运营方角色都可能引发争议。审查质保条款、赔偿安排、保险覆盖、运营方指引,以及事故后升级流程。
儿童与家庭数据暴露FTC 的 COPPA 材料覆盖照片、视频、音频、地理位置和信息收集限制;Sunday 面向的是忙碌家庭,而不是无菌的工业场景。家庭场景会带来混合受众和儿童数据边界问题,公开材料尚未处理。要求管理层说明儿童在场保护、家长控制假设,以及如何处理家庭场景边界问题。
加州法与公开网站法律集中度Sunday 的条款把公开争议集中到加州法律和 Mountain View 法院;已审阅的公开法律栈仍停留在网站层面,而不是机器人操作手册层面。中-高当前公开法律框架可能比投资人最终需要的机器人专项合同框架更薄。要求提供 beta 家庭协议、消费者披露,以及任何超出网站表面的机器人专项条款。

行按 Sunday 首批真实家庭部署中可能影响承销的严重程度排序,而不是按抽象法律议题的重要性排序。

[CR008, CR021, CR022, CR023, CR025, CR027]
FR003: 依赖图

Sunday 依赖数据贡献者、未披露的基础设施伙伴、监管方和家庭场景中的客户信任,才能把技术进展转化为可扩张的家庭业务。

[CR005, CR008, CR011, CR014, CR018, CR019]

7.3 运营和依赖风险仍然偏高,因为 Sunday 的数据优势仍依赖劳动力、快速迭代和脆弱的首批服务模型

从运营上看,Sunday 承诺的是机器人领域最难的事之一:机器人能自主进入新家庭,处理长周期家务,并以客户能感知的速度进步。官方材料清楚说明这需要多少基础设施。Sunday 称 Memo 从数百个独特家庭、2,000 多副 gloves 和约 1000 万段家庭情境中学习。ACT-1 文章认为 teleoperation 是规模化死锁,并称 Sunday 用 Skill Transform 和 90% 的 glove-to-robot 数据转换流程解决了其中一部分问题。这确实是缓解因素,也解释了为什么公司在招聘机器人 fleet management、语音交互、ML 数据与评估、数据运营岗位,同时还描述工程、研究和数据运营团队的大规模扩张。 但正是这套装置本身也构成风险。Sunday 自己的材料仍依赖一个配有直接支持、严格管理的 beta,这意味着服务负担还没有常规化。品类证据显示,隐藏人工劳动和支持强度仍真实存在。UNSW 描述了品类中的远程接管和录制家庭互动,The Robot Report 称 teleoperation 仓库仍存在。Sunday 可能比这条基线更先进,但其自有文本仍承认数据瓶颈,以及需要发现 Memo 在每个新家庭会遇到的独特问题。家庭不是工厂;它们多变、杂乱,充满非标准行为、宠物、儿童和易碎物品。NIST 的机器人项目也强化了这一点:在不可预测环境中实现安全自主性之前,需要经过验证的数据集、测试方法和健康监测。依赖风险同样仍高,因为多个关键输入要么部分外部化,要么仍不透明:Sunday 称受益于公开数据集和 VLM 进展,但没有公开说明制造栈、云架构或硬件供应商集中度。结果是,Sunday 可能在技术上领先同业,却在机器人离开演示循环、进入真实家庭后,被现场支持、数据运营劳动力或未披露供应商依赖困住。[CR004, CR005, CR006, CR009, CR012, CR013]

运营 / 质量 / 安全风险登记表
故障模式重要性发生可能性影响缓释成熟度剩余暴露尽调要求
凌乱家庭中的自主能力失效Sunday 承诺在新家庭里完成长周期家务,但家庭环境变化大,到处有易碎物、宠物和意外布局。前 50 户家庭仍可能暴露精选 demo 里看不到的故障模式。要求提供最新内部部署中的任务成功率、人工介入率、误停率和家庭问题日志。
隐性人力 / 远程操作泄漏品类资料显示,远程接管和遥操作仍支撑着部分家用机器人体验。中-高低-中Sunday 需要的远程支持可能仍高于自主叙事暗示的水平。要求提供遥操作率、远程辅助率,以及任何 human-in-the-loop 操作的明确披露政策。
支持模型难以走出礼宾式 beta公开队列只有 50 户家庭且配有直接支持,说明服务强度仍高。高接触启动会掩盖长期家庭经济性的薄弱。要求提供在 50、500 和 5,000 户家庭规模下预计的支持人员配比、备件计划、MTTR 目标和升级指标。
连接数据与 app 安全暴露Memo 使用配套 app 和语音交互;法律来源则表明,机器人可能牵涉音频、视频、地理位置和生物识别数据。中-高低-中机器人、app 或供应商栈的任何薄弱点,都可能演变成家庭信任事件。要求提供架构图、访问控制模型、加密标准和隐私内建评审。
数据管道转换或模型泛化失败Sunday 称 Skill Transform 能以 90% 成功率转换手套数据,基于地图的导航也能泛化到未见过的家庭。仍然可能出现数据漂移、边界案例失效或真实家庭表现下降。审查离线到在线迁移指标、回归测试,以及按家庭类型和任务拆分的失败情况。

登记表聚焦 Memo 部署后如何在真实家庭中失败,而不只看抽象机器人研究风险。

[CR007, CR008, CR009, CR018, CR030, CR031]
合作伙伴 / 依赖风险登记表
依赖项交易对手 / 系统角色失败情景严重性可见缓释剩余暴露
Memory Developers 与手套网络分布式人类数据贡献者构建自研家庭数据飞轮数据收集放慢、质量下滑,或贡献者经济性变差已发出 2,000+ 只手套,且在招聘 Data OperationsSunday 仍依赖持续的人类参与,才能让学习闭环保持速度。
公开模型与数据集世界模型、VLM 和其他上游 AI 积木加速感知与泛化上游模型停滞、条款变化,或无法顺利迁移到 Memo中-高Sunday 称自研数据仍是核心差异点核心自主能力仍部分依赖 Sunday 无法控制的外部研究进展。
未披露的硬件 / 制造栈未公开具名的供应商、合同制造商和零部件厂商生产机器人、备件和维修零件供应商瓶颈或质量问题拖慢部署或服务Sunday 刚拿到新资本,首批队列规模也小公开层面的制造集中度和应急规划仍不透明。
云 / app / 后端基础设施配套 app、语音栈、数据处理和远程服务运行排程、更新和支持工作流后端故障或供应商事件损害家庭信任中-高小规模 beta 初期应能限制影响范围公开记录没有说明冗余、供应商集中度或离线兜底行为。
监管方、标准机构和法院FTC、州隐私监管方、ISO 流程、加州法院塑造合规家庭部署的样子规则或责任预期收紧速度快于 Sunday 运营模型成熟速度Sunday 已发布基础法律页面Memo 进入家庭时,外部治理仍可能发生实质变化。

各行强调外部依赖:即便 Memo 的核心机器人栈继续改进,这些依赖也可能拖慢或重塑 Sunday 的规模曲线。

[CR005, CR006, CR018, CR020, CR025, CR027]
人员 / 执行风险登记表
职能 / 依赖观察到的信号发生可能性影响缓释成熟度剩余暴露尽调路径
研究到产品的转换Sunday 同时推进基础模型研究、手套数据、基于地图的导航和产品部署。技术野心并行太多,可能拖慢投资人最需要的单一证明点:稳定的家庭内表现。要求提供 ACT-1、车队软件和 beta 就绪工作流的里程碑延误历史。
数据运营管理招聘页面显示 Sunday 在招聘 Memory Developer、标注和 eval-ops;Series B 文章称数据运营扩大了 5x。中-高标注、QA 和贡献者管理的运营复杂度可能变成隐藏瓶颈。审查数据运营的组织架构、流失率、吞吐指标和 QA 失败率。
现场支持与车队运营Sunday 正在招聘机器人车队管理岗位,同时启动直接支持式 beta。低-中支持强度在 50 户家庭时或许可控,但还没有证明能放大。要求提供现场运营人员模型、备件规划,以及预期上门或远程支持率。
语音 / app / 全栈执行广度招聘页面显示 Sunday 同时在招语音交互、全栈软件和数据工具岗位。中-高中-高软件范围过宽,会增加安全敏感产品的 QA 和集成负担。要求提供缺陷积压、发布节奏,以及机器人、app 和后端层的责任归属图。
安全 / 合规领导深度Sunday 的公开页面展现出很强的机器人野心,但没有公开突出专门的安全、隐私或监管负责人。可见控制职能缺位,可能拖慢审批,并让后期尽调事项收不了尾。要求提供具名的合规、安全、隐私和产品法律负责人,以及外部顾问和评审机制。

执行行聚焦组织负荷:Sunday 要把研究进展变成可靠的家庭服务业务,需要承受这些负荷。

[CR018, CR019, CR020, CR039, CR045, CR046]
FR002: 风险传导图

隐私、安全、支持和可负担性风险如何叠加,进而拖慢采用并重新推高融资压力。

[CR005, CR011, CR013, CR027, CR030, CR037]

7.4 财务模型风险大多仍未关闭,因为消费者兴趣尚不等于耐久的家用机器人业务

Sunday 的公开需求故事令人鼓舞,但仍是漏斗顶部。公司据报道有 1,000 人 waitlist、数千份 beta 申请,在一个大多数成年人平均每天都会做家务的国家里,价值主张也非常直观。这些信号重要,因为它们说明 Sunday 想解决的问题是真实的。但它们还没有回答硬件加服务业务最关键的承销问题:价格、转化、毛利率、服务负载、更换成本、保修准备金,以及每个部署家庭的现金消耗。保留来源中没有发现公开 Memo 价格、单位经济性或毛利桥。这一缺口严重,因为 Altman Solon 的调研显示,该品类可能面临陡峭的消费者可负担性天花板:兴趣很高,但多数消费者不愿支付超过 $5,000,安全担忧也普遍存在。 本文件中最好的公开品类可比对象是 iRobot,不是因为产品相同,而是因为它展示了消费机器人在需求、毛利率和库存失衡后有多不宽容。iRobot 近期 filings 和发布材料显示,这家公司面临 going-concern 压力、贷款人豁免、低毛利、库存积压、战略评估、Amazon 交易失败后的大幅裁员,以及随后现金减少且已无剩余资本来源。Sunday 更早期,资本也更新鲜,因此不应机械类比。即便如此,它仍是正确的警示信号。消费机器人可以一直显得战略上令人兴奋,直到获客成本、支持负担或硬件经济性迫使融资重置。Sunday 真正可见的缓解因素,是小规模首批队列、直接支持、家庭数据护城河,以及足够在公开环境中学习的资本。击杀标准也同样清晰。如果 50 户 beta 延期,如果隐私和安全治理继续模糊,如果 Memo 的价格价值等式落在家庭愿意支付的范围之上,或如果部署需要隐藏人工劳动或不可规模化的支持比例,无论演示多有吸引力,投资逻辑都应大幅下调。[CR011, CR014, CR016, CR017, CR043, CR044]

缓释与否决标准表
风险可监控触发项阈值 / 事件重要性行动含义
Beta 执行延误Founding Family 部署时间50 户家庭 beta 明显晚于 2026 年底,或在直接支持准备不足的情况下启动这说明 Sunday 仍无法把 demo 和内部部署转化成面向客户的运营。视为一次重大去风险失败,并按比管理层说法更早期的公司重新承销。
安全 / 标准不成熟第三方安全证据部署前,管理层无法展示有意义的外部测试、认证路线图或清晰的内部安全论证标准仍不完整时,家庭信任只能押在公司叙事上。要求更低估值、里程碑融资,或暂停到安全证据产出后再推进。
隐私 / 数据治理缺口机器人专项隐私控制家庭机器人数据没有清晰的留存、删除、同意、供应商或事件响应框架缺少这些细节的联网机器人,会面对可避免的法律和声誉风险。实质上调法律风险,并把完整数据治理披露作为尽调条件。
隐性人力暴露远程介入或遥操作率Memo 在家庭中保持运转需要持续且有意义的 human-in-the-loop 操作会同时削弱自主能力逻辑、抬高支持经济性。除非介入率被证明会快速下降,否则把 Sunday 重新归类为人力增强服务模型。
可负担性 / 单位经济错配价格到价值证明Memo 全包价格和服务成本高于目标家庭愿意支付的水平,或高于公司能盈利支撑的水平单有消费者兴趣,造不出可行的机器人业务。除非 Sunday 能证明高端付费意愿和健康的家庭贡献利润率,否则大幅下调确信度。
资本效率压力现金需求轨迹Sunday 在证明持久 beta 满意度、可支持性和可信商业化路径前就需要再次融资模型仍未验证时,这会放大融资风险。推进前要求披露现金跑道、里程碑式资金用途和下行情景条款。

这些触发项有意设计成可监控指标,方便风险章节直接输入等待 / 推进 / 放弃决策。

[CR001, CR011, CR014, CR015, CR018, CR027]
FR001: 风险热力图

对 Sunday 最重要风险域的序数视图,强调部署野心目前跑在标准、服务经济性和合规成熟度的公开证据之前。

评级是截至 2026-06-10 基于留存证据综合得出的序数承销判断,不是数值预测。

[CR014, CR019, CR027, CR030, CR037, CR039]

7.5 图表

Chapter 08

08估值

8.1 建议与价格纪律

Sunday 容易让人喜欢,却难以定价。市场问题是真实的:BLS 称,80% 的人平均每天会做家庭活动,并花约两小时;Altman Solon 显示,65% 的消费者对先进家用机器人有兴趣。产品证据也强于许多更早期的机器人初创。Sunday 称 Memo 已用来自 500 多个家庭的约 1000 万段家庭情境训练,公司也已经从抽象演示推进到一个具名的 2026 年底 50 户 beta,并配有直接支持。但客户仍大多停在漏斗顶部。独立报道指向约 1,000 人 waitlist,而不是付费用户;保留来源没有披露标价、beta 费用、转化率、收入或毛利结构。 正是这个分母缺口,让建议无法变成买入。公开记录支持一个真实消费者问题、一条差异化数据收集回路,以及足够继续在现场学习的资本。它还不能支撑更难的承销问题:Memo 价格是多少、家庭是否真的愿意支付足够金额、产品需要多少支持,以及当前轮次是否嵌入了让头部估值产生误导的条款。竞争进一步强化谨慎。1X 是更接近的 home-first 同业,但披露的融资仍只是 $100 million Series B;Figure 资金实力强得多,post-money 达 $39 billion,但在家庭场景仍只是 alpha-testing;Samsung 可以用更简单的智能家居机器人争夺同一份家庭注意力预算;iRobot 则说明,即便有品牌、渠道和多年部署,消费机器人经济性也会变得多么残酷。因此,基于公开证据的正确判断是:以中等信心、高风险和完整估值立场继续研究,而不是干净利落地 yes 或 no。[CV001, CV004, CV005, CV006, CV009, CV010]

建议摘要表
维度评估决策含义
建议继续研究继续跟踪 Sunday,但不要只凭公开证据承销 2026 年 3 月价格。
置信度现有来源足以框定风险和当前估值,但还不足以结案。
风险评级最大风险是商业化和条款不透明,而不是技术野心不足。
估值立场足价当前价格或许能辩护,但公开证据没有显示它有吸引力。
入场纪律只能基于里程碑或下行保护付高价前,等待 beta 经济性、更清晰条款,或结构性保护。
持有 / 退出纪律暂无可支持的数字目标回报定价、beta 可支持性和股权结构条款披露后再重启研究;若触发否决项则退出。

这张表评估当前估值的可投资性,而不是 Sunday 这个产品愿景。

[CV001, CV002, CV009, CV034, CV035, CV039]
投资逻辑 / 反向逻辑表
论点投资逻辑什么会改变判断
市场需求家务确实耗时,消费者好奇心也真实存在。如果家庭拒绝支付足够价格,或节省时间的价值被证明过低,需求就无法货币化。
产品楔子Sunday 拥有罕见的真实家庭数据和面向家庭的产品设计。如果 beta 使用仍依赖大量隐性人力或频繁介入,数据护城河的价值就会下降。
客户证明已具名的 50 户家庭 beta 和约 1,000 人等待名单显示规模化前已有兴趣。没有订金、付费转化或持久使用,这些信号仍只是漏斗顶端。
融资信号$1.15B 轮说明可信投资人愿意为这个故事出钱。如果优先股条款很重,或下一轮融资早于商业证明到来,头部估值就会失去意义。
竞争背景Sunday 资金远强于 1X 这类家用优先同行,也没有 Figure 那么过热。如果 Figure、Samsung 或其他 incumbents 更快解决家庭编排,Sunday 的中间档估值可能守不住。
品类风险反向逻辑主要在可负担性、可支持性和信任,而不是缺少想象力。公开定价、支持配比和安全治理,可能把判断从足价推向合理。

反向逻辑由商业化证明和条款不透明驱动,而不是缺少真实家庭问题。

[CV004, CV005, CV009, CV010, CV011, CV012]
FV001: 投资建议逻辑

当前估值还能作为参照,只因为融资和产品证明都真实存在;但商业化分母仍然缺位。

[CV002, CV004, CV005, CV006, CV009, CV011]

8.2 融资背景、条款不透明与可比边界

融资故事在标题层面清晰,底层却不透明。Sunday 官方披露 2026 年 3 月 Series B 融资 $165 million,估值 $1.15 billion;TechCrunch、Yahoo Finance 和 Tracxn 均佐证了这一估值。Tracxn 还称两轮合计披露融资约 $200 million,这意味着投资人在公开变现证明出现前,已经支付独角兽价格。这个价格仍可能合理,但公开文件还没有说明为什么。保留来源没有披露 Memo 标价、付费客户数、收入、毛利率、净 burn 或清算优先权结构。也就是说,外部人无法判断本轮反映的是耐久运营证明、激进 sponsor optionality,还是两者兼有。 可比背景与其说填补了缺口,不如说给缺口划了边界。1X 证明,一个严肃的家庭 android 同业在讲家务和家庭辅助故事时,仍可能处于 $100 million 融资阶段;这让 Sunday 看起来是在投资人出价意愿上更靠前,而不一定是在已证明经济性上更靠前。Figure 位于另一端:post-money $39 billion,再加上 Brookfield 合作,可触达 100,000 多个住宅单元和大型商业空间。这证明资本愿意进入该品类,但也让 Sunday 的 $1.15 billion 估值像是介于资本轻型 aspirant 和巨额融资平台押注之间的中间地带。公开 incumbent iRobot 是最清醒的参照。其 FY2024 业绩显示收入 $681.8 million,但 GAAP 毛利率只有 20.9%、库存相关费用 $26.6 million,随后出现战略评估、going-concern 警告,以及 2025 年 Q3 关于没有额外资本来源的声明。换句话说,Sunday 当前价格没有被品类证伪,但公开证据也尚未证明该估值便宜,甚至没有证明它安全合理。[CV001, CV002, CV003, CV009, CV019, CV020]

可比估值表
可比对象指标倍数 / 估值 / 状态相关性局限性
Sunday(标的)2026 年 3 月私募轮$165M Series B 对应 $1.15B 估值;已披露累计融资约 $200M公司自身唯一硬公开价格发现。该估值背后没有公开价格、收入、毛利率或优先股条款。
iRobot上市消费机器人 incumbentsFY2024 收入 $681.8M;GAAP 毛利率 20.9%;到 2025 年进入战略评估 / 持续经营压力最好的公开参照,说明即便有品牌和规模,家用机器人经济性也可以非常苛刻。产品范围比 Sunday 窄,反映的是地板清洁 incumbent,而不是通用家用机器人。
1X私营家用 android 同行$100M Series B;不到 12 个月融资超过 $125M;NEO 定位于家务和家庭帮助留存集合中最接近的、已披露融资的家用优先类人机器人参照。未披露估值、付费需求证明或成熟消费者部署数据。
Figure私营类人机器人平台龙头已承诺 >$1B Series C,投后 $39B高端资本和品类热度参照。体量和资金基础使其更像热度与平台离群值,而不是 Sunday 的干净定价可比对象。
Apptronik私营类人机器人融资 / 工业部署$520M,估值 $5B;Apollo 已与 Mercedes-Benz 和 GXO 测试人形机器人公司一旦拿出工业部署势头,投资人胃口会提速,这一点很清楚。工业优先的姿态让它更像资本市场参照,而不是家庭需求可比公司。
Unitree低成本人形机器人定价信号G1 标价 US $13.5K;H1 定位为通用人形机器人这个参照提醒我们,未来人形机器人价格压力可能既来自高端美国创业公司,也来自更便宜的硬件新进入者。公开标价不等于已交付的家庭能力、安全性或支持经济性。
Figure + Brookfield里程碑 / 部署参照管理 100,000+ 套住宅单元,另有大型商业地产,可供数据和部署使用人形机器人公司有大伙伴时,规模化数据和部署入口可能长什么样,这一组合给了样本。合作规模不能直接转化为对 Sunday 估值的支撑。
Samsung Ballie既有智能家居替代方案未披露融资估值;大型家电生态用另一条路径进入同一家庭预算适合作为提醒:家庭机器人也可能被更简单、生态驱动的形态切入。Ballie 不是直接的通用操作机器人。
Amazon / iRobot并购参照收购在 2024 年终止;随后 iRobot 裁员 31%即便是品类龙头,战略可选性也可能快速丧失,这一点值得记住。监管和交易本身的因素让它只是宽松参照,不是直接估值可比。

这一组覆盖本章估值逻辑所用的全部参照:一个直接价格参照、一个上市消费机器人既有公司、两个私人人形机器人参照、一个智能家居替代方案,以及一个并购警示信号。

[CV002, CV003, CV019, CV020, CV022, CV023]

8.3 乐观 / 基准 / 悲观逻辑与下行触发因素

情景表应被理解为按里程碑加权的估值判断,而不是虚假的精确性。乐观情景下,Sunday 按时完成 50 户 beta,披露家庭可以直接承受或通过可信融资吸收的定价,并证明支持负载和隐藏人工劳动会随着 fleet 成熟而下降。在那个世界里,当前轮次开始像是一次早期去风险入场,而不是峰值估值。基准情景下,2026 年 3 月轮次仍是唯一硬性的价格发现点,beta 如期发生但仍像一个高接触验证场,投资人获得足够证据来 defend 当前估值,但还不足以证明大幅溢价合理。悲观情景下,可负担性问题继续悬而未决,beta 延期或需要过多 concierge 支持,投资人发现条款或单位经济性让头部估值更像宣传而非保护。 反向来源重要,因为它们比上行更清楚地定义下行触发因素。Altman Solon 称,近 70% 的消费者不会为家用机器人支付超过 $5,000,25% 的消费者无论价格多少都不会购买。IEEE 和 UNSW 显示,在许多情境下消费者仍偏好专用机器人,并担心安全、隐私和隐藏远程劳动。The Robot Report 让 teleoperation 问题变得具体;IEEE 的安全标准报道则称,家用机器人治理仍缺少可执行测试和规则制定。这些信号不意味着 Sunday 会失败;它们意味着,任何高于当前估值的价格,都需要证明 Sunday 能解决品类级约束,而不只是继承品类热情。[CV004, CV011, CV012, CV014, CV015, CV016]

乐观 / 基准 / 悲观情景表
情景概率信号假设估值逻辑示例公允价值
乐观Beta 按时启动,Memo 定价在融资支持或明显高端价值下跨过家庭可负担门槛,支持负担下降,新增尽调显示融资条款温和。相比上一轮支付溢价,因为 Sunday 开始像少数能把数据优势转化成家庭付费使用的家用机器人公司。$1.4B-$2.2B
基准中-高2026 年 3 月轮仍是最佳价格发现点,50 户家庭 beta 落地,商业化证明改善,但仍显得高接触、偏早期。估值贴近当前标记,因为融资支持真实存在,但公开经济性仍不完整。$0.9B-$1.3B
悲观Beta 延误,Memo 定价超过家庭支付意愿,隐性人力持续存在,或优先股条款比头部估值暗示得更差。下调到上一轮以下,因为公司会从接近部署的故事退回昂贵机器人期权。$0.4B-$0.8B
决策规则在定价、可支持性和条款透明度改善前,基准情景占主导。把当前标记当作参考点,而不是目标回报模型。当前标记 = $1.15B

这些区间是围绕上一轮、品类可负担性证据和阶段匹配可比参照作出的情景判断,而不是靠虚构收入倍数推导。

[CV001, CV002, CV011, CV012, CV016, CV017]
投资逻辑破裂与止损触发因素表
触发因素阈值对投资逻辑的传导行动含义
Beta 执行滑坡50 户家庭 beta 明显晚于 2026 年底,或上线时缺少可信的直接支持准备。说明公司仍比估值叙事所暗示的阶段更早。将案例重切到悲观区间,或暂停。
可负担性错配Memo 全包价格落在消费者价值区间之上,且没有融资方案或异常强的节省证明。家庭兴趣和付费需求之间的桥会断掉。要求更低入场价格,或放弃。
隐性人工劳动家务要跑通,仍需要大量人类在环操作。同时削弱自主性和利润率假设。按服务属性更重的模型承销 Sunday,或退出。
优先权悬置既往或本轮融资文件显示清算优先权、反稀释条款或治理权扭曲了表观估值。名义估值会成为经济入场价值的劣质代理。要求结构化条款,或重定价。
安全 / 信任缺口部署前,管理层无法拿出可信的安全测试、隐私和事故治理材料。家庭产品转化会放慢,责任风险会上升。上调风险评级,并把任何投资都绑定到控制措施上。
消费机器人经济性接近 iRobot利润率、支持成本或库存动态开始接近承压消费硬件经济性。战略故事会变成资本效率问题。将案例移向估值区间下端。

这些是估值止损触发因素:每一项都可能直接推翻当前价格,而不只是让故事没那么好听。

[CV011, CV016, CV017, CV018, CV020, CV021]
FV002: 估值敏感性

估值更受里程碑和商业化证明牵动,而不是任何公开收入倍数。

数值是示意性的情景中点,单位为十亿美元;锚点是当前融资估值和基于里程碑的估值判断,而不是凭空编造的销售倍数。

[CV001, CV011, CV016, CV017, CV036, CV037]
FV003: 估值 / 回报区间

当前轮次落在基准情景内;上行取决于商业化证明,下行则由可负担性或执行失误驱动。

[CV001, CV034, CV036, CV037, CV039]

8.4 最终尽调要求与退出准备度

因此,退出准备度仍处早期。Sunday 已具备商业化推进前投资人想看的要素:新资本、具名首批队列、真实家庭数据,以及专门面向家务而非工厂 retrofit 设计的产品架构。但公开记录仍停在能让委员会干净定价所需的文件之前。没有公开 beta 合同,没有公开价格表,没有披露支持比例,没有单位经济性桥,也没有公开解释清算优先权或其他股权结构保护。即便安全和信任层也仍部分开放,因为品类标准框架还在追赶真实家庭自主性。 因此,最终尽调要求并不泛泛。投资人需要商业包:Memo 价格、融资方案、如有定金则披露定金、beta 到付费转化预期、每户服务分钟数、故障和更换率,以及清晰的收入和毛利桥。还需要融资包:pro forma 股权结构表、清算 stack、发行价格、治理权,以及此前轮次中的任何 ratchet 或下行保护。破坏投资逻辑的触发条件也相应简单:如果 2026 年底 beta 延期,如果 Memo 全包价格在没有强有力融资支持的情况下高于消费者价值区间,如果家庭家务仍需要有意义的 human-in-the-loop operation,或如果条款 stack 意味着新资金排在比头部估值暗示更重的优先权 overhang 之后,就应退出或实质性重定价。[CV008, CV009, CV018, CV035, CV037, CV039]

最终尽调问题表
主题缺失证据重要性负责人 / 尽调路径
商业条款Memo 标价、融资方案、订金、如有 beta 费用,以及各队列上线时间。这是判断需求质量和消费者可负担性的基本分母。管理层定价材料加 beta 材料。
转化与使用申请漏斗、入选家庭标准、激活率、重复任务频率,以及 beta 后转化假设。等候名单规模不等于付费家庭需求。增长、产品和客户成功仪表盘。
单位经济性硬件 BOM、替换成本、支持分钟数、上门服务风险、毛利率,以及按队列划分的贡献利润率。家用机器人可以看起来很有吸引力,却仍然不赚钱。财务和运营审查。
股权结构表与优先权当前备考股权结构表、清算堆叠、发行价格、董事会权利,以及任何反稀释或 MFN 保护。表观估值可能与经济入场价值明显背离。法律顾问审查融资文件。
安全与信任控制安全测试、隐私架构、留存和删除规则、事故流程,以及保险覆盖。家庭机器人需要先拿出信任证据,再谈规模。产品、法务和外部保证审查。
规模路径从 50 户到 500 户及以上的里程碑,包括人员配比和支持自动化假设。真正的问题是 Sunday 会成为可规模化的家庭业务,还是礼宾式试点。运营计划和董事会里程碑。

这些是把 Sunday 从可信故事变成可承销价格所需的最低尽调问题。

[CV008, CV009, CV018, CV035, CV037, CV039]
FV004: 投资 KPI

核心估值 KPI 说明了 Sunday 为什么值得关注,但价格还没有彻底跑通。

[CV002, CV003, CV004, CV005, CV011, CV020]

8.5 图表

免责声明

本报告仅供参考,基于截至 2026-06-10 本次运行所留存的公开证据。

证据索引

结论
编号陈述可信度来源
CO001 Sunday is publicly described as a Mountain View-based home robotics startup. SO011, SO013
CO002 Retained sources identify Sunday as founded in 2024 by Tony Zhao and Cheng Chi, but do not disclose an exact founding day. SO013, SO008
CO003 Memo is Sunday’s home robot for repetitive household chores such as table clearing, dishwashing, laundry, and tidying. SO001, SO011
CO004 Sunday’s visible business model is pre-commercial household deployment, with beta rollout preceding any publicly disclosed broad consumer commercialization. SO001, SO002, SO011
CO005 Sunday says it owns the full stack across software, hardware, data, training, and model evaluation. SO002, SO006
CO006 Sunday’s technical differentiation centers on Skill Capture Glove plus Skill Transform to convert human household behavior into robot-training data. SO003, SO001
CO007 Sunday says it has shipped more than 2,000 gloves to Memory Developers. SO001
CO008 Sunday says Memo is trained on data from more than 500 real homes and about 10 million household episodes. SO011, SO012, SO018
CO009 Sunday says Memo progressed from a one-arm shoe-arranging system in December 2024 to folding socks, handling glassware, and pulling espresso by October 2025. SO001
CO010 Sunday publicly plans a Founding Family Beta for 50 households launching in late 2026. SO001, SO011
CO011 Sunday emerged from stealth on 2025-11-19 and paired the launch with a $35 million financing backed by Benchmark and Conviction. SO011, SO013
CO012 Memo’s launch materials emphasize a rolling base and soft silicone-clad design rather than a bipedal humanoid form factor. SO011
CO013 Tony Zhao is publicly identified as Sunday’s co-founder and CEO. SO014, SO013, SO011
CO014 Tony Zhao says he previously worked on ALOHA and ACT at Stanford before dropping out to build Sunday. SO014
CO015 Tony Zhao’s public profile lists Stanford CS PhD work plus prior experience at Google DeepMind, Tesla Autopilot, Google X Intrinsic, and Berkeley EECS. SO014
CO016 Cheng Chi is publicly identified as Sunday’s co-founder and CTO. SO013, SO011
CO017 Cheng Chi is listed as the submitting author on the Diffusion Policy paper, supporting Sunday’s public claim to recent robot-learning research pedigree. SO015
CO018 The retained public sources do not disclose a full board roster or ownership structure for Sunday. SO002, SO007, SO013
CO019 Thomas Laffont joined Sunday’s board as part of Coatue’s March 2026 investment. SO002, SO010, SO009
CO020 Sunday’s public company story remains highly dependent on Tony Zhao and Cheng Chi for research credibility, product vision, and fundraising narrative. SO014, SO009, SO011
CO021 Sunday announced an oversubscribed $165 million Series B at a $1.15 billion valuation on 2026-03-12. SO002, SO008, SO010
CO022 Public Series B materials name Coatue as lead investor and also name Bain Capital Ventures, Fidelity Management & Research Company, Tiger Global, Benchmark, Conviction, and Xtal Ventures as participants. SO002, SO009, SO010
CO023 Tracxn’s profile implies Sunday has raised $200 million across two rounds, combining the $35 million Series A and $165 million Series B. SO013, SO011
CO024 The retained funding chronology consists of a November 2025 launch financing followed by a March 2026 Series B, with no public evidence of an intervening disclosed round. SO013, SO011, SO002
CO025 Sunday’s valuation and total-raised narrative are supported more by secondary and market-data sources than by a detailed company cap-table disclosure. SO013, SO008, SO002
CO026 The retained public sources do not disclose revenue, ARR, named customers, pricing, debt facilities, or secondary sales. SO001, SO002, SO013
CO027 TechCrunch reported that Bloomberg said Sunday already had 1,000 people on its waitlist by March 2026. SO008
CO028 Tech Funding News reported that thousands of people had applied for Sunday’s beta program after the stealth exit. SO009
CO029 Tracxn reported Sunday had 127 employees as of late March 2026. SO013
CO030 Tech Funding News reported Sunday had grown to more than 70 engineers and researchers. SO009
CO031 Sunday’s March 2026 post says engineering grew 3x, research 4x, and data operations 5x as it prepared for deployment. SO002
CO032 Sunday’s careers page shows Redwood City and remote roles across software and data operations, implying a Bay Area operating footprint that extends beyond the Mountain View label in press materials. SO006, SO011
CO033 Sunday argues robotics is bottlenecked more by data than by hardware, compute, or funding. SO003, SO009
CO034 Sunday says Skill Transform converts glove data into equivalent robot data with a 90 percent success rate. SO003
CO035 Sunday says it spent more than a year engineering the core infrastructure behind Skill Capture Glove, Skill Transform, and Memo before its current acceleration. SO003
CO036 Sunday says ACT-1 can autonomously execute table-to-dishwasher work and generalize to unseen homes using map-conditioned navigation. SO003
CO037 Sunday claims ACT-1 is the first foundation model to combine long-horizon manipulation with map-conditioned navigation in a single end-to-end model. SO003
CO038 By March 2026 Sunday was explicitly reframing itself from a demo company to a deployment company. SO002, SO009
CO039 Independent robotics commentary argues that home-humanoid timelines remain constrained by the robotics data gap, dexterity limits, and over-optimistic investor narratives. SO016, SO017
CO040 Robot Report quoted Ken Goldberg saying in-home butler timelines are hype and unlikely within the next two, five, or even ten years. SO016
CO041 IEEE Spectrum reported survey participants generally preferred special-purpose home robots over humanoids and noted that optimistic home-humanoid scenarios assume safety, regulatory approval, and insurance conditions that may still be far away. SO017
CO042 No company-specific lawsuit, sanction, or regulatory enforcement event appeared in the retained sources, but Sunday’s public disclosure package is still thin on governance, economics, and commercialization evidence. SO007, SO013, SO016
CO043 As of the run date, Sunday appears to be a late-stage private Series B company rather than a publicly launched consumer robot business with broad market deployment. SO013, SO002
CO044 The best evidence-based one-line description is that Sunday is a full-stack home robotics company building Memo and using proprietary glove-captured household data to train robots ahead of beta deployment. SO001, SO002, SO003
CO045 Sunday’s owned communications footprint remains thin in the retained sources, with the public journal feed showing only a small number of posts around the March 2026 financing and research narrative. SO007
CM001 The narrowest defensible market boundary for Sunday is household service and personal-care robots used in homes rather than the full humanoid-robot category. SM001, SM002, SM020
CM002 Sunday publicly positions Memo around household chores in kitchens, dining rooms, laundry spaces, and other lived-in home environments. SM002, SM006
CM003 Broad humanoid-robot forecasts include healthcare, hospitality, manufacturing, defense, education, retail, and residential assistance in one market frame. SM015
CM004 Unpaid household labor is a core status-quo substitute because BLS says 80% of people do household activities on an average day and spend about two hours on them. SM016
CM005 Sunday claims Memo is built to help families with dishes, laundry, tidying, table clearing, dishwasher loading, shoe organization, and espresso preparation. SM002, SM006
CM006 Mordor Intelligence estimates the domestic service robots market at $14.62 billion in 2025 and $34.93 billion by 2031. SM012
CM007 Global Market Insights estimates the household robots market at $14.7 billion in 2025 and $85 billion by 2035. SM013
CM008 Global Market Insights reports the U.S. household robots market at $3.6 billion in 2025. SM013
CM009 Mordor Intelligence says North America held 38.3% of domestic service robot revenue in 2025. SM012
CM010 Applying Mordor's 38.3% North America share to its $14.62 billion 2025 global market implies an approximately $5.6 billion North American domestic service robot market. SM012
CM011 Mordor reports that floor-cleaning units held 64.70% of domestic service robot revenue in 2025. SM012
CM012 Mordor also says vacuuming and mopping captured 65.60% of domestic service robot applications in 2025, showing how cleaning-heavy today's category still is. SM012
CM013 Future Market Insights values the humanoid robot market at $7.8 billion in 2025 and projects $248.9 billion by 2036. SM015
CM014 The best public TAM proxy for Sunday is the household or domestic service robot category, not the broad humanoid robot category. SM012, SM013, SM015
CM015 Sunday's public deployment and training evidence is U.S.-centered, making North America and the United States the most defensible near-term geography lens. SM003, SM005, SM012, SM013
CM016 No retained public source isolates a Sunday-specific SAM or SOM for premium multi-step home manipulation. SM012, SM013, SM015
CM017 Sunday's current public go-to-market picture looks like a consumer self-pay purchase rather than an insurer, employer, or government-funded one. SM003, SM006, SM024
CM018 Sunday's public adoption path begins with a Founding Family Beta for 50 households launching in late 2026. SM003, SM006
CM019 Independent coverage says Sunday plans to use early home deployments to collect feedback, improve the system, and prepare for wider adoption. SM010, SM011
CM020 Time-poor dual-income households are a plausible early buyer segment because home chores compete directly with paid work and other daily responsibilities. SM016, SM024
CM021 Older-adult and caregiver households are an adjacent segment because aging demographics are strong, but Sunday has not publicly positioned Memo as an eldercare-specific product. SM017, SM018, SM012, SM013
CM022 BLS reports that women who did household activities spent 2.7 hours on those activities versus 2.3 hours for men. SM016
CM023 BLS reports that adults in households with children under age 6 spent an average of 2.5 hours per day providing primary childcare in 2024. SM016
CM024 Altman Solon found that 65% of surveyed U.S. consumers were interested in owning an advanced home robot, but 85% reported only moderate familiarity or less. SM024
CM025 Altman Solon found that the top desired home-robot tasks were cleaning, security, cooking, laundry, and deliveries. SM024
CM026 Altman Solon found that about 50% of respondents were concerned about the safety of having a human-sized advanced robot in their home. SM024
CM027 IEEE Spectrum reports that surveyed users generally prefer special-purpose robots over humanoids because they seem safer, more private, and more comfortable in the home. SM021
CM028 IEEE Spectrum says optimistic home-humanoid scenarios in its survey assumed safety testing, regulatory approval, and insurance coverage that may be decades away from reality. SM021
CM029 Ken Goldberg told The Robot Report that useful humanoid butlers are not likely in the next two, five, or even ten years because dexterity and data remain limiting factors. SM023
CM030 UNSW's analysis says current home humanoid products still raise privacy risks, depend on teleoperation for tricky tasks, and may remain 20 years away from wide acceptance. SM025
CM031 ISO 13482 covers mobile servant robots and physical assistant robots and addresses hazards involving people, domestic animals, and property. SM020
CM032 IEEE Spectrum's standards coverage argues that current domestic-humanoid standards still lack binding compliance criteria, test methods, and enforcement mechanisms for domestic autonomy. SM022
CM033 NIST says robotics adoption in unstructured environments depends on rigorous validation, safe human teaming, and measurable performance across dexterity, mobility, perception, and interaction. SM019
CM034 Both Mordor and GMI identify smart-home integration, convenience, and aging-related demand as key growth drivers for household or domestic service robots. SM012, SM013
CM035 Both Mordor and GMI also preserve meaningful constraints around privacy, safety, awareness, and high upfront cost. SM012, SM013
CM036 Sunday claims its competitive edge is a full-stack data flywheel built from Skill Capture Glove capture, hardware control, and in-home data from more than 500 homes and about 10 million episodes. SM005, SM006, SM011
CM037 Public evidence still does not answer the key commercialization questions of price, conversion, support burden, or repeat usage strongly enough to support a precise SOM forecast. SM003, SM023, SM024
CM038 The U.S. and North American figures narrow Sunday's more relevant public market lens well below the broad global household-robot headline. SM012, SM013
CM039 Altman Solon's survey ceiling of mostly sub-$5,000 willingness to pay is far below the US$20,000 price point discussed around 1X's Neo by UNSW, underscoring a likely affordability gap for early home humanoids. SM024, SM025
CM040 IEEE's evidence implies that Roomba-like and other special-purpose devices remain strong substitutes because users often prefer a toolbox of smaller machines over one humanoid generalist. SM021
CM041 Sunday says Memo uses a rolling base and soft silicone-clad design, which reduces some of the fall and approachability concerns associated with bipedal home humanoids. SM006, SM011
CM042 The retained market estimates are not additive or interchangeable because domestic service robots, household robots, and humanoid robots use different category boundaries and forecast horizons. SM012, SM013, SM015
CM043 Scaling beyond affluent early adopters into older-adult or caregiver households requires stronger safety and trust evidence than is public today. SM017, SM018, SM022
CM044 Altman Solon reports that independent reviews and recommendations from friends and family are trusted purchase inputs, implying that social proof matters for broader home-robot adoption after beta. SM024
CP001 Sunday markets Memo as a helpful home robot for busy households and household chores. SP001
CP002 Sunday says Memo should work autonomously out of the box in real homes rather than only in staged demos. SP001
CP003 Sunday says it owns the full robotics stack and trains Memo without teleoperation by collecting human movement data from real homes. SP002
CP004 1X markets NEO as a home robot for household chores and everyday home assistance. SP004
CP005 1X says NEO is designed for gentle and safe interactions in the home. SP004
CP006 1X announced a $100 million Series B and said it had raised more than $125 million in less than 12 months. SP005, SP007
CP007 1X said its next milestone after the Series B was offering NEO to consumers. SP005
CP008 1X told The Robot Report that NEO will work autonomously for most tasks but will initially rely on external cloud models for advanced long-term reasoning and planning. SP007
CP009 1X's world-model post says it is gathering large amounts of robot data from homes and offices and still treats evaluation as an unresolved challenge for general-purpose robots. SP006
CP010 Figure says Helix can pick up small household objects and explicitly frames the home as robotics' hardest environment. SP008
CP011 Figure said its Series C financing exceeded $1 billion in committed capital at a $39 billion post-money valuation. SP009
CP012 Figure said it is scaling humanoid robots into homes and commercial operations at the same time. SP009
CP013 Figure said its Brookfield partnership provides access to more than 100,000 residential units for humanoid data collection and deployment work. SP010, SP011
CP014 TechCrunch reported that Figure's home effort was still alpha testing in 2025 while industrial deployments remained the more immediate commercial path. SP012
CP015 Agility says Digit is the first humanoid robot in production deployment on facility floors. SP013
CP016 Agility's retained solutions page centers warehouse automation integration, live metrics, and workflow orchestration rather than home use. SP014
CP017 IEEE described Digit's current near-term commercial task as moving warehouse totes rather than performing broad home chores. SP015
CP018 IEEE said Agility still had a large gap between doing the warehouse task most of the time and doing it robustly enough that customers would pay for it. SP015
CP019 Boston Dynamics markets Atlas for enterprise material handling and industrial automation rather than for home deployment. SP016
CP020 Boston Dynamics says Atlas is being built for factories, warehouses, or construction sites that require strength, endurance, and dexterity. SP017
CP021 Boston Dynamics told IEEE it intentionally targeted Atlas at industrial applications after learning from Spot that product-market fit must be defined before scale. SP018
CP022 Samsung says Ballie is an AI home companion that autonomously drives around the home and manages appliances for users. SP019
CP023 The retained Samsung Ballie source describes home-assistant features but does not disclose a public retail price or launch date. SP019
CP024 The retained public LG robotics evidence is about business applications such as disinfection and hospitality rather than a shipping home robot. SP020
CP025 IEEE said iRobot's Roomba j7 is a camera-equipped home robot that maps homes and avoids common floor obstacles. SP021
CP026 IEEE reported that iRobot shifted from pure autonomy toward human-robot collaboration because users wanted more control over home-robot behavior. SP022
CP027 IEEE reported that richer iRobot home sensing raises privacy risk even when more processing stays on the robot and shared images are encrypted. SP021, SP022
CP028 CBS reported that after the Amazon acquisition collapsed, iRobot cut 31% of its workforce and changed leadership. SP023
CP029 Tailos says Rosie is a commercial robot vacuum for hospitality use and generally cannot be shipped to residential addresses. SP024
CP030 Tailos says Rosie has deployed more than 1,000 robots across 12 countries. SP024
CP031 The retained Everyday Robots sources say Alphabet shut down the project as a standalone business after it had deployed more than 100 cafeteria-cleaning and sorting robots internally. SP025, SP026
CP032 WIRED reported that some Everyday Robots technology and part of the team were consolidated into Google Research after the shutdown. SP026
CP033 In the retained evidence, Sunday's closest direct home-first general-purpose peers are 1X and Figure, while Ballie and iRobot are narrower household alternatives and Agility, Boston Dynamics, Tailos, and LG are adjacent or industrial paths. SP001, SP004, SP008, SP012, SP015, SP018, SP019, SP020, SP021, SP024
CP034 Distribution and commercialization leverage is stronger today for ecosystem players and capital-rich partners than for standalone home-robot startups. SP009, SP011, SP019, SP021, SP023
CP035 Figure and 1X both pursue the home, but Figure currently shows the stronger capital and residential-data access while 1X is the more explicit consumer-home-first pitch. SP005, SP007, SP009, SP011, SP012
CP036 Households can multi-home by keeping human labor, smart-home devices, or robot vacuums alongside a new robot instead of making an immediate full switch. SP019, SP021, SP024
CP037 The main barrier to switching is trust, support, and affordability rather than hard data lock-in. SP021, SP022, SP023
CP038 Sunday's most credible public moat claim is a real-home data and training loop optimized specifically for household chores. SP001, SP002
CP039 Public sources do not yet prove that Sunday's home-data loop is stronger than the learning and distribution advantages accumulating around 1X and Figure. SP002, SP006, SP009, SP010, SP011
CP040 Figure's retained independent coverage still frames home deployment as early enough that industrial work remains the nearer-term commercialization path. SP012
CP041 Everyday Robots shows that impressive general-purpose assistive-robot demos can still fail commercially when reliability, mission clarity, and customer economics do not solidify. SP025, SP026
CP042 The retained home-robot sources discuss safety, privacy, and user trust, but they do not disclose enough certification or field-reliability detail to underwrite home deployment confidence. SP007, SP019, SP021, SP022
CP043 Across the cited general-purpose home-robot pages in this chapter, public pricing remains undisclosed. SP004, SP008, SP019
CI001 Sunday announced an oversubscribed $165 million Series B at a $1.15 billion valuation in March 2026. SI005, SI007, SI008, SI009
CI002 Public sources support a $35 million November 2025 launch round and about $200 million of total disclosed funding across two rounds. SI010, SI011, SI012
CI003 Sunday’s Founding Family Beta opened applications in November 2025 for 50 households that would receive individually numbered robots and direct support in late 2026. SI001, SI003, SI010, SI011
CI004 Sunday says Memo was trained on data from more than 500 real homes and about 10 million household episodes. SI010, SI011
CI005 Sunday says it has shipped more than 2,000 Skill Capture Gloves to Memory Developers. SI001
CI006 Independent coverage reported at least a 1,000-person waitlist and separately described thousands of beta applicants. SI007, SI008
CI007 The visible commercial motion is direct-to-household rather than enterprise-led, because Sunday markets Memo directly to families and routes early access through its own beta application flow. SI001, SI003, SI010
CI008 Sunday does not publicly disclose a Memo list price, beta fee, subscription price, or financing plan in the retained sources. SI001, SI003, SI005, SI007
CI009 No retained public source discloses Sunday revenue, ARR, paying-customer count, or named customer revenue. SI001, SI005, SI007, SI012
CI010 Sunday says it owns the full stack across software, hardware, data, training, model evaluation, and manufacturing. SI002, SI005, SI006
CI011 Sunday said public deployment starts with the beta and that engineering grew 3x, research 4x, and data operations 5x to support that shift. SI005
CI012 Tech Funding News reported that Sunday had grown to more than 70 engineers and researchers by March 2026. SI008
CI013 Sunday’s careers page still lists Redwood City and remote openings across fleet software, data-operations tooling, memory development, data annotation, and evaluation operations. SI004
CI014 March 2026 sources describe beta deliveries beginning within months or later that year as Sunday moved from demos to deployment. SI005, SI008, SI009
CI015 Sunday says its data flywheel combines glove data and robot-fleet data and now operates at tens of millions of movement episodes. SI005, SI013
CI016 Sunday argues that teleoperation creates a scaling deadlock because useful robot data is too slow and costly to gather one remote-operated hour at a time. SI006
CI017 The U.S. Bureau of Labor Statistics reported that people spent about two hours per day on household activities in 2024. SI015
CI018 Altman Solon found that 69% of surveyed consumers were unwilling to pay more than $5,000 for a home robot. SI014
CI019 Altman Solon found that 25% of surveyed consumers were unwilling to purchase a home robot at any price. SI014
CI020 Altman Solon reported that purchase preferences favored buying outright at 37% and subscription models at 24%. SI014
CI021 Altman Solon reported that 54% of consumers expected a home robot to save six hours per week or less and that 61% would pay at most $14 per hour of saved time. SI014
CI022 IEEE Spectrum reported that surveyed users generally preferred special-purpose robots over humanoids because they seem safer, more private, and more comfortable in the home. SI016
CI023 IEEE Spectrum noted that optimistic home-humanoid scenarios assumed extensive safety testing, regulatory approval, and insurance coverage that may be decades away from reality. SI016
CI024 IEEE Spectrum’s standards coverage says the current ISO 13482 revision still lacks binding compliance criteria, test methods, and enforcement mechanisms for domestic autonomy. SI017
CI025 UNSW’s analysis says current home humanoids still raise privacy questions, may rely on remote workers for tricky tasks, and could remain far from wide household acceptance. SI018
CI026 Ken Goldberg told The Robot Report that useful personal butler robots are not likely in the next two, five, or even ten years because dexterity and data remain limiting factors. SI019
CI027 Tracxn reports Sunday as a 2024-founded Series B company that had raised about $200 million and employed 127 people as of March 26, 2026. SI012
CI028 Tracxn’s funding table lists no public revenue multiple for Sunday’s March 2026 Series B. SI012
CI029 No retained Sunday source discloses cash on hand, monthly burn, runway months, debt facilities, or covenant limits. SI005, SI007, SI008, SI012
CI030 A 50-home beta with direct support implies nontrivial service-delivery cost before any mass-market scale is visible. SI003, SI010, SI011
CI031 Full-stack ownership plus heavy engineering, research, data-operations, and fleet-support hiring imply a cost base spanning R&D, manufacturing, and field support rather than pure software delivery. SI002, SI004, SI005, SI013
CI032 iRobot’s FY2024 results showed $681.8 million of revenue and a 20.9% GAAP gross margin. SI020, SI023
CI033 iRobot said a $26.6 million inventory and non-cancelable purchase-commitment charge reduced FY2024 gross margin by 3.9 percentage points. SI020, SI023
CI034 iRobot reported $134.3 million of cash and cash equivalents and $76.0 million of inventory at December 28, 2024, versus $152.5 million of inventory a year earlier. SI020, SI023
CI035 iRobot said it had reduced total headcount by about 51% to 541 and had started a strategic review that included debt refinancing or a potential sale. SI020, SI021
CI036 iRobot said Q4 2024 revenue was hurt by promotional spending, timing of orders, and competition, and that 83% of robot sales came from $300-plus MSRP bands. SI020, SI023
CI037 CBS reported that iRobot cut 31% of its workforce, or about 350 employees, after Amazon’s proposed acquisition was terminated. SI022
CI038 Sunday’s public financial picture is capital-rich but commercialization-poor: funding and activity proxies are public, while price, revenue, margin, and cash metrics are not. SI001, SI005, SI008, SI012
CI039 The public record does not reveal whether beta households pay, receive subsidized units, or participate under non-revenue test agreements, so revenue recognition remains unresolved. SI003, SI010, SI011
CI040 Public evidence leaves Sunday’s eventual revenue mix unreadable because any combination of hardware, service, support, or software could exist, but no stream-level split is disclosed. SI001, SI003, SI005, SI009
CI041 Capital adequacy depends more on undisclosed cash and burn than on the headline size of the Series B alone. SI005, SI008, SI012, SI020
CI042 Because the 2026 capital is explicitly tied to deployment and larger operating teams, Sunday now needs commercialization proof rather than just another demo milestone. SI005, SI008, SI009, SI013
CI043 iRobot's Q3 2025 results said revenue fell below internal expectations, increasing cash usage and pressuring fixed-cost leverage. SI026
CI044 As of September 27, 2025, iRobot reported $24.8 million of cash, $140.9 million of inventory, and said it had no sources upon which it could draw for additional capital. SI026
CI045 iRobot's official Roomba shop page listed products from about $249.99 to $1,399.99 and offered consumer financing through Affirm. SI030
CI046 NBER finds that robot investment is small and highly concentrated, accounting for less than 0.30% of aggregate equipment expenditures. SI027
CI047 A PLOS One literature review says smart-home adoption depends on both perceived utility and perceived risk rather than utility alone. SI028
CI048 A Frontiers study says privacy and security concerns can be barriers to domestic social robot adoption among older adults. SI029
CE001 Sunday markets Memo as a home robot for busy households. SE001, SE010, SE011
CE002 Sunday says Memo works 24/7 to lighten repetitive household chores. SE001
CE003 Sunday publicly shows Memo clearing tables, scraping food waste, and loading dishwashers. SE001, SE010, SE011
CE004 Sunday says Memo should work autonomously out of the box across different homes. SE001
CE005 Sunday says hundreds of people in unique homes show Memo how chores are done each day. SE001
CE006 Sunday says it has shipped more than 2,000 Skill Capture Gloves to Memory Developers. SE001
CE007 Sunday says Memo progressed from one-arm shoe arranging in December 2024 to multi-skill behavior by October 2025. SE001
CE008 Sunday says Memo beta deployment is planned for late 2026. SE001, SE003, SE010, SE011
CE009 Sunday says 50 households in the Founding Family Beta will receive individually numbered robots with direct support. SE003, SE010, SE011
CE010 Sunday says beta households can propose new skills for hundreds of Memory Developers to collect in the wild. SE003
CE011 Sunday says it owns the full stack across software, hardware, data, training, and model evaluation. SE002, SE007
CE012 Sunday’s March 2026 post extends its full-stack description to hardware design and manufacturing. SE007
CE013 Sunday says the main robotics bottleneck is data rather than hardware, compute, or funding alone. SE006
CE014 Sunday says its key technical insight is to solve embodiment mismatch between human demonstrations and robot execution. SE006
CE015 Sunday says Skill Capture Glove and Memo’s hand share the same geometry and sensor layout. SE006
CE016 Sunday says Skill Transform converts glove data into robot-equivalent data with a 90 percent success rate. SE006
CE017 Sunday says it spent over a year building Skill Capture Glove, Skill Transform, Memo, and ACT-1. SE006
CE018 Sunday says ACT-1 is its first foundation model. SE006
CE019 Sunday says ACT-1 completed a table-to-dishwasher routine with 33 unique and 68 total dexterous interactions across 21 objects while navigating more than 130 feet. SE006
CE020 Sunday says ACT-1 generalizes to unseen homes using 3D maps and map-conditioned navigation. SE006
CE021 Sunday highlights folding socks and pulling espresso as dexterity-heavy tasks for ACT-1. SE002, SE006
CE022 Public launch coverage says Memo training draws from about 10 million episodes sourced from more than 500 homes. SE006, SE010, SE011
CE023 Launch coverage says Memo uses a rolling base to improve balance and lower weight in the home. SE010, SE011
CE024 Launch coverage says Memo has a soft silicone-clad exterior intended to fit family living spaces. SE010, SE011
CE025 Sunday says Engineering grew 3x, Research 4x, and Data Operations 5x to support deployment. SE007
CE026 Sunday’s careers page publicly shows roles in robot fleet management, ML data and evaluations, data-operations tooling, voice interaction, memory development, annotation, and eval ops. SE005
CE027 Sunday says its proprietary data flywheel draws from both gloves and robot fleet operations. SE007
CE028 Sunday’s company page links the founders’ prior work to ALOHA and Diffusion Policy. SE004
CE029 Tony Zhao’s profile says he worked on ALOHA and ACT at Stanford before founding Sunday. SE018
CE030 Sunday’s company page says Cheng Chi invented Diffusion Policy and leads Sunday’s stack from data-collection infrastructure through machine-learning pipelines. SE004
CE031 Diffusion Policy’s public paper and project surfaces report a 46.9 percent average success-rate improvement and expose code, logs, checkpoints, and Colabs. SE012, SE013, SE014
CE032 ALOHA’s public project surfaces describe a low-cost open-source bimanual teleoperation system and ACT results with 80 to 90 percent success on selected fine-manipulation tasks. SE017, SE018
CE033 Mobile ALOHA publicly adds a mobile base and whole-body teleoperation to the ALOHA lineage and reports up to 90 percent gains from co-training on mobile tasks. SE016, SE018
CE034 ALOHA Unleashed publicly combines large-scale ALOHA 2 data collection with Diffusion Policies for dexterous bimanual tasks. SE019, SE020
CE035 Sunday’s published privacy policy covers beta or newsletter contact details, communications, device or network data, and website activity data. SE008
CE036 Sunday’s published privacy materials do not describe robot-sensor retention, in-home video handling, or fleet-data governance specifics. SE008, SE003
CE037 ISO 13482 applies to personal care robots including mobile servant robots and addresses hazards from human-robot physical contact. SE024
CE038 IEEE reports that current domestic-robot standards still lack binding compliance criteria, test methods, or enforcement mechanisms for relational home safety. SE022
CE039 NIST says trustworthy robotics adoption depends on measurement science, performance metrics, validated datasets, safe-operation testing, health monitoring, and easier integration. SE023
CE040 TechCrunch reported that Sunday had about 1,000 people on its waitlist by March 2026. SE009
CE041 The Robot Report quotes Ken Goldberg arguing that home-butler timelines are hype because dexterity and the robotics data gap remain unresolved. SE025
CE042 Sunday’s company page says it is taking ALOHA and Diffusion Policy style innovations from the lab into real homes by collecting data from hundreds of real people in real homes. SE004
CE043 The accessed public materials do not publish Memo-specific field reliability metrics, incident logs, or safety certification results. SE001, SE003, SE007, SE008
CE044 The accessed public materials do not disclose a robot-specific compliance program beyond general safety-oriented design language and category-level external standards. SE007, SE022, SE023, SE024
CE045 The accessed public materials do not name Memo manufacturing partners, component suppliers, or integration partners. SE004, SE005, SE007
CE046 The most material remaining product-technology blockers are field reliability disclosure, robot-specific data governance, safety validation detail, and supplier transparency. SE008, SE022, SE023, SE024, SE025
CU001 Sunday publicly positions Memo as a helper for busy households and families rather than as an enterprise or industrial robot. SU001, SU004, SU013
CU002 The visible buyer, user, and payer are the household or family unit, with no separate corporate payer or facility operator disclosed on customer-facing surfaces. SU001, SU004, SU018
CU003 Sunday’s public use cases center on dishes, laundry, tidying, table clearing, shoe arranging, and similar domestic chores. SU001, SU003, SU013
CU004 Sunday has not publicly segmented Memo customers by geography, income, home type, or revenue band on the surfaces reviewed. SU001, SU002, SU004, SU023
CU005 The beta program page says Memo performs tasks autonomously in new homes and can be scheduled through direct voice interaction or a companion Sunday app. SU002
CU006 The beta program page says Founding Families will propose new skills and help Sunday discover problems Memo faces in new homes. SU002
CU007 Sunday says its training pipeline draws on data from more than 500 real homes and hundreds of families or Memory Developers. SU003, SU013, SU016, SU024, SU025, SU026
CU008 Sunday says Memo training has used about 10 million genuine household episodes. SU003, SU013, SU024, SU025
CU009 Sunday’s Founding Family Beta is planned for late 2026 with 50 early-adopter households receiving individually numbered robots and direct support. SU002, SU013, SU016, SU024, SU025, SU026
CU010 Sunday’s March 2026 deployment messaging says public deployment starts with the beta program rolling out to users in fall 2026. SU007, SU014, SU017
CU011 GlobeNewswire and Tech Funding News say Sunday has received thousands of applications for its upcoming beta program. SU012, SU017
CU012 Tech Funding News says early beta deployments are intended to collect customer feedback, improve the system, and prepare for wider adoption. SU012
CU013 TechCrunch and Yahoo report that Sunday already had about 1,000 people on its waitlist in March 2026, citing Bloomberg. SU011, SU015
CU014 The public evidence separates three different customer-adjacent cohorts: training-data homes, beta applicants or waitlist members, and the future 50-household beta cohort. SU002, SU011, SU012, SU013, SU017
CU015 Training-data homes are evidence of data collection and household exposure, not proof of paid Memo deployments. SU003, SU013, SU024
CU016 Waitlist members and beta applicants are demand signals, not active or retained customer accounts. SU011, SU012, SU017
CU017 No named paying household or production Memo customer appears in the public sources reviewed for this run. SU001, SU002, SU006, SU007, SU011, SU018
CU018 Sunday’s visible customer acquisition path is direct-to-household beta recruitment rather than a reseller, distributor, or enterprise procurement channel. SU001, SU002, SU004, SU007
CU019 The careers and Series B surfaces emphasize engineering, research, data operations, and deployment capacity rather than a scaled public sales organization. SU005, SU007, SU017
CU020 Sunday published customer-facing privacy, terms, and cookies pages effective September 3, 2025, and its sitemap still listed those pages in June 2026. SU008, SU009, SU010, SU023
CU021 The public privacy and terms surfaces are website-level legal documents and do not publicly describe robot-specific sensor retention, escalation SLAs, or hardware replacement commitments for beta households. SU002, SU008, SU009, SU010
CU022 The public record reviewed here does not disclose Memo pricing, contract length, renewal terms, churn, GRR, NRR, or customer satisfaction scores. SU001, SU002, SU007, SU011, SU018
CU023 Because no public live-customer cohort metrics are disclosed, there is no public evidence for repeat usage, renewal behavior, or land-and-expand performance. SU002, SU007, SU011, SU018
CU024 Sunday’s current customer proof is therefore pilot-stage and top-of-funnel rather than production-scale or revenue-backed. SU009, SU011, SU012, SU017, SU018
CU025 Tech Funding News says customers have already applied for the beta in the thousands, but it still frames those deployments as an early feedback phase before wider adoption. SU012
CU026 Sunday’s customer evidence stack depends heavily on company-originated pages and launch syndications rather than on independent customer reviews or named case studies. SU002, SU006, SU013, SU016, SU024, SU025, SU026
CU027 IEEE Spectrum characterizes home-humanoid demand as still uncertain and frames the category as moving from hype-filled funding to an unresolved homes-next question. SU020
CU028 UNSW says household humanoids remain clumsy in uncontrolled homes and that privacy risk rises because the best data comes from robots operating in actual homes. SU021
CU029 UNSW also cites the International Federation of Robotics view that useful and widely accepted home androids may still be 20 years away. SU021
CU030 The Robot Report quotes Ken Goldberg describing teleoperation as a still-common way to make robots perform tasks, underscoring how far many systems remain from fully autonomous household use. SU019
CU031 Altman Solon found that 65 percent of surveyed U.S. consumers expressed interest in owning an advanced home robot, showing real category curiosity. SU022
CU032 Altman Solon also found that 85 percent of respondents had only moderate familiarity or less with advanced home robots, limiting easy mainstream conversion. SU022
CU033 Nearly 70 percent of Altman Solon respondents were unwilling to pay more than $5,000 for a home robot and 25 percent were unwilling to buy at any price. SU022
CU034 Altman Solon says about half of respondents worry about the safety of a human-sized advanced robot in the home, while privacy invasion is among the top moral concerns. SU022
CU035 Altman Solon says trusted purchase inputs include independent reviews and recommendations from friends and family, which Sunday has not yet matched with public Memo case studies. SU022, SU017
CU036 Tracxn describes Sunday as a home-robot company for busy households with a Skill Library and monthly skill improvement, but it does not add customer counts or deployment depth. SU018
CU037 Sunday’s sitemap shows the company page was last modified on 2026-06-02 while the journal and series-b pages were the only dated customer-facing narrative updates found in the retained official surface. SU006, SU007, SU023
CU038 Sunday’s public customer journey currently moves from broad household positioning to beta recruitment, then to a tightly managed first-cohort deployment, with wider scale still prospective. SU001, SU002, SU007, SU012, SU017
CU039 Because no named customer concentration, account mix, or cohort revenue data is public, top-customer risk cannot yet be quantified from the public record. SU002, SU007, SU018
CU040 The strongest public adoption proof today is interest and onboarding readiness, not demonstrated renewal or scaled in-home utilization. SU007, SU011, SU012, SU017
CR001 Sunday’s homepage says Memo’s beta is launching in late 2026. SR001
CR002 Sunday’s homepage positions Memo as a household robot built for busy households and intended to work continuously on repetitive chores. SR001
CR003 Sunday’s homepage specifically markets tasks such as clearing tables, throwing out food scraps, loading the dishwasher, and running it. SR001
CR004 Sunday says hundreds of people in unique homes show Memo how chores are done each day. SR001
CR005 Sunday says it has shipped more than 2,000 Skill Capture Gloves to Memory Developers. SR001
CR006 Sunday says the Skill Capture Glove distills millions of human movements into Memo’s onboard AI. SR001
CR007 Sunday’s beta page says Memo performs tasks autonomously in new homes. SR003
CR008 Sunday’s beta page says households will schedule tasks through direct voice interaction or the companion Sunday app. SR003
CR009 Sunday’s beta page says founding households will propose new skills and help surface unique problems Memo faces in new homes. SR003
CR010 Sunday’s beta page says Founding Families are first in line to secure limited-edition numbered Memo robots. SR003
CR011 Launch coverage says Sunday’s Founding Family Beta begins in late 2026 with 50 households receiving individually numbered robots and direct support. SR012, SR016, SR017
CR012 Sunday’s launch materials say Memo was developed from routines captured in more than 500 real homes. SR012, SR016, SR017
CR013 Sunday’s launch materials say Memo training data includes approximately 10 million episodes of genuine household routines. SR012
CR014 Sunday announced an oversubscribed $165 million Series B at a $1.15 billion valuation. SR013, SR014, SR015, SR018
CR015 Sunday’s Series B messaging says the company is moving beyond demos to focus on real-world deployment of Memo. SR006, SR013, SR014
CR016 Sunday said in March 2026 that it had received thousands of applications for the beta program. SR013
CR017 TechCrunch reported that Sunday had about 1,000 people on its waitlist after emerging from stealth. SR011, SR018
CR018 Sunday’s Series B post says engineering grew 3x, research 4x, and data operations 5x to support deployment. SR006, SR013
CR019 Sunday’s careers page shows active hiring in robot fleet management, voice interaction, ML data and evaluations, and full-stack software. SR005
CR020 Sunday’s careers page lists remote and in-house Memory Developer roles plus multiple data-operations roles. SR005
CR021 Sunday’s privacy policy is effective September 3, 2025. SR008
CR022 Sunday’s privacy policy says it collects information that allows it to provide, maintain, and improve its services. SR008
CR023 Sunday’s privacy policy says users may provide first and last name, email address, and phone number when signing up for the newsletter or beta. SR008
CR024 Sunday’s terms of service are effective September 3, 2025. SR009
CR025 Sunday’s terms say disputes are governed by California law and subject to exclusive jurisdiction in Mountain View, California courts. SR009
CR026 Sunday’s cookies policy is also effective September 3, 2025. SR010
CR027 FTC business guidance says companies that make privacy promises must honor them and must maintain security appropriate to the data they possess. SR027
CR028 FTC consumer-privacy guidance tells businesses to reread privacy policies and make sure they honor the promises they have pledged. SR028
CR029 FTC COPPA guidance covers children’s privacy, including photos, videos, audio recordings, geolocation, and requirements to limit information collection. SR029
CR030 Cyber Law Monitor says AI-powered robotics providers may be subject to state privacy laws, biometric laws, and FTC requirements. SR030
CR031 Cyber Law Monitor says robotics companies may handle video, audio, geolocation, user-profile, facial-geometry, voiceprint, and gait data, and should implement minimization, retention, access-control, encryption, and vendor-management safeguards. SR030
CR032 Transformative AI Legal Leaps says home robots often use cameras, microphones, and other sensors, raising significant privacy issues in intimate home spaces. SR033
CR033 Brookings argues that notice-and-choice consent is becoming meaningless for many AI applications. SR031
CR034 Hill Dickinson says a central liability question for humanoid robots is whether responsibility for harm sits with the manufacturer, operator, or software provider. SR032
CR035 Hill Dickinson says visual, facial, and cloud-handled robot data can create compliance tension under regimes such as GDPR. SR032
CR036 ISO 13482:2014 sets safety requirements for personal care robots, including mobile servant robots, physical assistant robots, and person carrier robots. SR026
CR037 IEEE Spectrum reports that the proposed ISO update for domestic robots addresses hazard identification and risk assessment but still does not set limits, testing methods, or enforcement mechanisms for human-robot collaboration. SR022, SR026
CR038 IEEE Spectrum says no current standard fully converts domestic-autonomy knowledge into enforceable rules for home robot safety. SR022
CR039 NIST says robotics adoption still needs rigorous validation, metrics, test methods, and protocols to assure autonomy, safety, collaboration, and performance in intended environments. SR025
CR040 UNSW reported that 1X’s home robot can require remote takeover by an employee wearing a virtual-reality helmet, and that the process is recorded for future learning. SR023
CR041 UNSW warned that home robots can arrive with concealed privacy risks and invisible remote workers behind the scenes. SR023
CR042 The Robot Report quotes Ken Goldberg saying teleoperation warehouses with humans operating robots like puppets still exist in the U.S. and China. SR024
CR043 Altman Solon found that 65% of surveyed U.S. consumers expressed interest in home robots, but 69% were unwilling to pay more than $5,000 and about 50% were concerned about safety. SR019
CR044 BLS reported that 80% of people engaged in household activities on an average day, spending about two hours on them. SR020
CR045 Sunday’s own ACT-1 article says teleoperation creates a scaling deadlock for general-purpose robots and that gathering enough data the standard way could take decades. SR007
CR046 Sunday says Skill Transform converts glove data into equivalent robot data with a 90% success rate. SR007
CR047 Sunday says ACT-1 was deployed in Airbnb locations and can generalize to unseen homes with zero environment-specific training. SR007
CR048 Sunday says ACT-1 uses 3D maps and map-conditioned navigation to find key locations in new houses. SR007
CR049 iRobot’s 2024 Form 10-K says the auditor expressed substantial doubt about the company’s ability to continue as a going concern and that the company depended on lender waivers to avoid default. SR036
CR050 iRobot’s 2024 Form 10-K says the company depends nearly entirely on consumer robots, relies primarily on a single contract manufacturer and a small number of suppliers, and faces privacy and cybersecurity obligations tied to customer and robot-specific data. SR036
CR051 iRobot’s March 2025 results said the board had initiated a strategic review and reported Q4 2024 GAAP gross margin of 9.5%. SR035, SR038
CR052 CBS reported that after Amazon’s acquisition attempt was terminated amid antitrust scrutiny, iRobot cut 31% of its workforce. SR037
CR053 iRobot’s November 2025 results said cash and cash equivalents had fallen to $24.8 million, inventory remained $140.9 million, and the company had no remaining sources for additional capital. SR039
CR054 Sunday’s risk profile is elevated because public deployment claims, privacy obligations, safety standards, and consumer price ceilings are advancing faster than public proof on Memo’s field reliability, support economics, and compliance maturity. SR003, SR019, SR022, SR030
CV001 Sunday officially says it raised an oversubscribed $165 million Series B at a $1.15 billion valuation in March 2026. SV001, SV006
CV002 TechCrunch, Yahoo Finance, and Tracxn independently corroborate Sunday's March 2026 $1.15 billion valuation. SV005, SV009, SV012
CV003 Tracxn says Sunday has raised about $200 million across two rounds and had 127 employees as of March 2026. SV012
CV004 Sunday's first public deployment proof is a late-2026 50-household Founding Family Beta with direct support. SV002, SV007, SV010
CV005 Independent reporting says Sunday already has about 1,000 people on its waitlist before broad household deployment. SV005, SV009
CV006 Sunday says Memo training uses about 10 million household episodes collected from more than 500 homes. SV003, SV007, SV010
CV007 Sunday says public deployment starts with the beta and that engineering, research, and data operations have scaled 3x, 4x, and 5x respectively. SV001
CV008 Sunday's beta page shows Memo is a connected, service-heavy product with voice/app scheduling and direct company support. SV002
CV009 No retained public source discloses Memo list price, paid customer count, revenue, gross margin, or liquidation-preference details for Sunday's financing rounds. SV001, SV005, SV006, SV007, SV012
CV010 Altman Solon found that 65% of surveyed U.S. consumers express interest in an advanced home robot, but familiarity remains low. SV013
CV011 Altman Solon found that 69% of consumers are unwilling to pay more than $5,000 for a home robot and 25% are unwilling to buy at any price. SV013
CV012 Altman Solon found that 54% of consumers expect six hours per week of time savings or less and 61% would pay no more than $14 per hour saved. SV013
CV013 BLS says 80% of people do household activities on an average day and spend about two hours on them, confirming the core chore problem is real. SV032
CV014 IEEE Spectrum reported that surveyed users generally prefer special-purpose robots over humanoids in the home. SV014
CV015 That same IEEE article says home-humanoid acceptance was tested under optimistic assumptions about safety testing, regulatory approval, and insurance that may be far from current reality. SV014
CV016 UNSW warns that home robots raise privacy questions, may hide remote workers behind the scenes, and may still be around 20 years away from broad usefulness. SV015
CV017 The Robot Report says some humanoid deployments still rely on teleoperation warehouses, underscoring hidden-human-labor risk for the category. SV017
CV018 IEEE's domestic-robot standards coverage says the proposed ISO update still lacks enforceable testing methods, limits, and governance mechanisms for domestic autonomy. SV016
CV019 iRobot's 2024 SEC filing says FY2024 revenue was $681.8 million, down 23.4% year over year. SV018
CV020 iRobot's FY2024 results showed only 20.9% GAAP gross margin. SV019
CV021 iRobot's FY2024 results were hit by a $26.6 million manufacturing-transition and excess-inventory charge. SV019
CV022 In March 2025 iRobot initiated a strategic review and disclosed substantial doubt about its ability to continue as a going concern. SV019
CV023 By Q3 2025 iRobot said it had $24.8 million of cash, $140.9 million of inventory, and no additional capital sources available. SV020
CV024 Public Roomba pricing spans roughly $250 to $1,400, showing how far today's mainstream home-robot price architecture sits below premium humanoid aspirations. SV022
CV025 1X officially disclosed a $100 million Series B to bring the NEO home-assistance android to market. SV023
CV026 1X markets NEO around chores, scheduling, and gentle home interactions, making it one of the closest home-first peers to Sunday. SV024, SV031
CV027 The Robot Report says 1X had raised over $125 million in less than 12 months, reinforcing that serious home-humanoid peers still sit much earlier in disclosed financing than Sunday's $1.15 billion mark. SV025
CV028 Figure says its 2026 Series C exceeded $1 billion of committed capital at a $39 billion post-money valuation. SV026
CV029 Figure's Brookfield partnership adds access to more than 100,000 residential units and large commercial footprints for data collection and future deployment. SV027
CV030 TechCrunch says Figure's home effort is still only alpha testing and that the home path remains difficult because homes are messy and pricing questions remain open. SV029
CV031 Figure's Helix post says the home is robotics' greatest challenge and that current approaches will not scale there without a step change. SV028
CV032 Figure says Helix was trained on roughly 500 hours of teleoperated behavior data, highlighting how much manual data work still sits inside leading humanoid programs. SV028
CV033 Samsung's Ballie demonstrates that large incumbents can attack the household-assistance problem through smart-home orchestration rather than a general-purpose humanoid form factor. SV030
CV034 Because Sunday has no public price, revenue, margin, or term disclosure, the March 2026 round is best treated as sponsor-backed reference pricing rather than proof the stock is cheap. SV001, SV005, SV006, SV012
CV035 The public record supports only milestone-based or downside-protected entry discipline, not a price-insensitive buy call. SV009, SV012, SV013, SV019, SV020
CV036 A successful 50-home beta, disclosed pricing that clears the affordability hurdle or uses credible financing, and declining support intensity are the main catalysts that could justify paying above the current mark. SV002, SV007, SV013, SV024, SV029
CV037 Beta slippage, a price-to-value mismatch, hidden human labor, or investor-unfriendly financing terms would make Sunday's current valuation look full to expensive. SV001, SV013, SV015, SV017, SV019
CV038 Sunday's best public valuation references are milestone-based rather than software-multiple-based: an under-earning consumer incumbent in iRobot, a much earlier home peer in 1X, and a hyper-capitalized humanoid platform in Figure. SV019, SV020, SV023, SV025, SV026, SV027
CV039 Public evidence does not support a numeric target return today because entry price, dilution, exit timing, and unit economics remain under-specified. SV001, SV005, SV006, SV012
CV040 Exit readiness is still low-to-medium: Sunday has capital, real-home data, and a named beta cohort, but not public proof on pricing, safety governance, supportability, or commercial conversion. SV001, SV002, SV003, SV016, SV018
CV041 CNBC says Apptronik raised $520 million at a $5 billion valuation in 2026 and is already testing Apollo in factories and warehouses with partners including Mercedes-Benz and GXO Logistics. SV034
CV042 Apptronik's Apollo page says the robot is designed for warehouses and manufacturing first, with later ambition for areas including home delivery and elder care. SV033
CV043 Unitree advertises G1 at US $13.5K and warns that the humanoid industry is still in an early stage of exploration, highlighting the possibility of lower-cost hardware pressure from outside the U.S. startup set. SV035, SV036
CV044 Physical Intelligence and Skild AI are both positioning themselves as general-purpose robot-model layers, implying that Sunday may face competition not only from integrated robot OEMs but also from embodied-AI platform providers. SV038, SV039
来源
编号出版方标题引文
SO001 Sunday The helpful home robot Beta launching late 2026
SO002 Sunday Sunday’s Series B: No More Demos We just raised an oversubscribed $165M Series B at a $1.15B valuation to stop demos and bring the first autonomous home robot to households this year.
SO003 Sunday ACT-1: A Robot Foundation Model Trained on Zero Robot Data We convert glove data into equivalent robot data with a 90% success rate.
SO004 Sunday Mundane made magic
SO005 Sunday Mundane made magic
SO006 Sunday Shape the future of AI robotics Redwood City, CA
SO007 Sunday Journal | Sunday Robotics | The helpful robotics company
SO008 TechCrunch Humanoid robotics maker Sunday reaches $1.15B valuation to build household robots Sunday says it has raised $165 million at a $1.15 billion valuation in a Series B round led by Coatue Management.
SO009 Tech Funding News Sunday raises $165M Series B at $1.15B valuation to launch home robots The company has grown to more than 70 engineers and researchers.
SO010 Yahoo Finance / GlobeNewswire Sunday raises $165M to launch first autonomous home robot Sunday, the company building helpful home robots, announced it has raised an oversubscribed $165 million Series B funding round at a $1.15 billion valuation.
SO011 IT News Online / GlobeNewswire Sunday Launches Memo, the Robot That Actually Learns Your Home Sunday is a Mountain View–based robotics startup founded by Tony Zhao and Cheng Chi.
SO012 Yahoo Finance / GlobeNewswire Sunday launches Memo, the robot that actually learns your home Its development centers on learning from authentic daily routines collected in more than 500 real homes.
SO013 Tracxn Sunday - 2026 Company Profile, Team, Funding & Competitors Sunday is a series B company based in Mountain View (United States), founded in 2024.
SO014 Tony Z. Zhao GitHub Pages - Tony Z. Zhao Hi, I’m the co-founder and CEO of Sunday Robotics (sunday.ai). I previously worked on ALOHA and ACT at Stanford before dropping out.
SO015 arXiv Diffusion Policy From: Cheng Chi
SO016 The Robot Report Why humanoid robots aren’t advancing as fast as AI chatbots I’m not saying it’s not going to happen, but I’m saying it’s not going to happen in the next two years, or five years or even 10 years.
SO017 IEEE Spectrum Home Humanoid Robots Survey Reveals Surprising Preferences Our survey showed that people generally prefer special-purpose robots over humanoids.
SO018 The AI Insider Household robot maker Sunday raises $165M in Series B funding with $1.15B valuation Sunday has raised $165 million in an oversubscribed Series B round led by Coatue, valuing the company at about $1.15 billion.
SO019 Sunday The helpful robotics company
SO020 Sunday Sunday Robotics | The helpful robotics company
SO021 Benchmark 404 File not found
SO022 Conviction 404 File not found
SO023 Crunchbase Target URL blocked or unavailable during fetch
SO024 WIRED WIRED promotional landing page returned instead of article text
SO025 Coatue Portfolio
SM001 Sunday The helpful home robot Beta launching late 2026
SM002 Sunday Technology | Sunday Robotics | The helpful robotics company Watch Memo autonomously clear the dining table, fold socks, and pull a real espresso.
SM003 Sunday Beta Program Fifty households will become early adopters, receiving individually numbered robots along with direct support and the opportunity to guide future capabilities.
SM004 Sunday Sunday's Series B: No More Demos Public deployment starts with our beta program, rolling out to users this Fall.
SM005 Sunday ACT-1: A Robot Foundation Model Trained on Zero Robot Data We deployed ACT-1 in a series of Airbnb locations, tasking it with clearing a dining table and loading a dishwasher.
SM006 Yahoo Finance / GlobeNewswire Sunday Launches Memo, the Robot That Actually Learns Your Home Starting November 19th, 2025, Sunday will accept applications for Memo’s Founding Family Beta, launching in late 2026. Fifty households will become early adopters.
SM007 Yahoo Finance / GlobeNewswire Sunday Raises $165M to Launch First Autonomous Robots by Thanksgiving Sunday plans to break that cycle with their upcoming Beta taking on time-intensive household chores.
SM008 IT News Online / GlobeNewswire Sunday Launches Memo, the Robot That Actually Learns Your Home Its development centers on learning from authentic daily routines collected in more than 500 real homes.
SM009 TechCrunch Humanoid robotics maker Sunday reaches $1.15B valuation to build household robots
SM010 Tech Funding News Robots doing your chores: Sunday raises $165M at $1.15B valuation to make it happen The company plans to use these early deployments to collect feedback, improve the system, and prepare for wider adoption.
SM011 The AI Insider Household Robot Maker Sunday Raise $165M in Series B Funding With $1.15B Valuation Memo is Sunday’s home robot features a stable rolling base, a soft silicone-clad design built to handle chores such as dishes, laundry and general tidying.
SM012 Mordor Intelligence Domestic Service Robots Market Size, Competitive Landscape 2026 – 2031 The domestic service robots market size is expected to grow from USD 14.62 billion in 2025 to USD 16.9 billion in 2026 and is forecast to reach USD 34.93 billion by 2031 at 15.62% CAGR over 2026-2031.
SM013 Global Market Insights Household Robots Market Size, Share & Forecast, 2026–2035 The global household robots market was estimated at USD 14.7 billion in 2025.
SM014 Future Market Insights Household Robot Market Size, Trends & Forecast 2024-2034
SM015 Future Market Insights Humanoid Robot Market | Global Market Analysis Report - 2036 The humanoid robot market was valued at USD 7.80 billion in 2025 and is forecast to expand to USD 248.90 billion by 2036 at a 37.0% CAGR.
SM016 U.S. Bureau of Labor Statistics American Time Use Survey Summary On an average day, 80 percent of people engaged in household activities such as housework, cooking, lawn care, or household management, spending about two hours on these activities.
SM017 U.S. Census Bureau U.S. Older Population Grew From 2010 to 2020 at Fastest Rate Since 1880 to 1890 The older population reached 55.8 million or 16.8% of the population of the United States in 2020.
SM018 OECD Elderly population Elderly population is the share of the population aged 65 years and over.
SM019 NIST Robotics This promise cannot be fulfilled without rigorous validations and characterizations of these technologies to ensure that they meet the applications and environments for which they are intended.
SM020 ISO ISO 13482:2014 ISO 13482:2014 specifies requirements and guidelines for the inherently safe design, protective measures, and information for use of personal care robots.
SM021 IEEE Spectrum Home Humanoid Robots Survey Reveals Surprising Preferences Our survey showed that people generally prefer special-purpose robots over humanoids.
SM022 IEEE Spectrum Domestic Humanoid Robot Safety Standards Are Shifting No current standard fully converts that knowledge into enforceable rules for domestic autonomy.
SM023 The Robot Report Why humanoid robots aren’t advancing as fast as AI chatbots I’m saying it’s not going to happen in the next two years, or five years or even 10 years.
SM024 Altman Solon Humanoid Home Robotics Survey 69% of respondents are unwilling to pay more than $5,000 for a home robot.
SM025 UNSW Sydney / The Conversation Humanoid home robots are on the market – but do we really want them? Useful and widely accepted home androids may still be 20 years away.
SP001 Sunday Sunday Robotics | The helpful robotics company
SP002 Sunday Technology | Sunday Robotics | The helpful robotics company
SP003 1X 1X | Home Robots
SP004 1X NEO | 1X Home Robot
SP005 1X 1X secures $100M in Series B funding
SP006 1X 1X World Model
SP007 The Robot Report 1X Technologies raises $100M Series B to advance NEO humanoid robot
SP008 Figure AI Helix
SP009 Figure AI Series C
SP010 Figure AI Project Go-Big
SP011 Figure AI Figure Announces Strategic Partnership with Brookfield
SP012 TechCrunch Figure will start alpha testing its humanoid robot in the home in 2025
SP013 Agility Robotics Agility Robotics
SP014 Agility Robotics Solutions | Agility Robotics
SP015 IEEE Spectrum Agility Robotics’ Digit Is Ready for Work—Maybe
SP016 Boston Dynamics Atlas | Boston Dynamics
SP017 Boston Dynamics Training a Humanoid Robot for Hard Work
SP018 IEEE Spectrum Boston Dynamics' New Electric Atlas Robot Is a Humanoid, and a Product
SP019 Samsung Newsroom A Day in the Life With Ballie: An AI Companion Robot for the Home
SP020 The Robot Report LG Electronics autonomous disinfection robot targets business applications
SP021 IEEE Spectrum With New Roomba j7, iRobot Wants to Understand Our Homes
SP022 IEEE Spectrum iRobot Announces Major Software Update, Shift From Pure Autonomy to Human-Robot Collaboration
SP023 CBS News Amazon calls off bid to buy iRobot. The Roomba vacuum maker will now cut 31% of its staff.
SP024 Tailos Rosie - AI-Powered Commercial Robot Vacuum | Tailos
SP025 The Robot Report Alphabet closes Everyday Robots among layoffs
SP026 WIRED Alphabet Layoffs Hit Trash-Sorting Robots
SI001 Sunday The helpful home robot
SI002 Sunday Mundane made magic
SI003 Sunday Mundane made magic
SI004 Sunday Shape the future of AI robotics
SI005 Sunday Sunday’s Series B: No More Demos
SI006 Sunday ACT-1: A Robot Foundation Model Trained on Zero Robot Data
SI007 TechCrunch Humanoid robotics maker Sunday reaches $1.15B valuation to build household robots
SI008 Tech Funding News Sunday raises $165M Series B at $1.15B valuation to launch home robots
SI009 Yahoo Finance / GlobeNewswire Sunday raises $165M to launch first autonomous home robots by Thanksgiving
SI010 Yahoo Finance / GlobeNewswire Sunday launches Memo, the robot that actually learns your home
SI011 IT News Online / GlobeNewswire Sunday Launches Memo, the Robot That Actually Learns Your Home
SI012 Tracxn Sunday - 2026 Company Profile, Team, Funding & Competitors
SI013 The AI Insider Household robot maker Sunday raises $165M in Series B funding with $1.15B valuation
SI014 Altman Solon Humanoid Home Robotics Survey
SI015 U.S. Bureau of Labor Statistics American Time Use Survey -- 2024 Results
SI016 IEEE Spectrum Home Humanoid Robots Survey Reveals Surprising Preferences
SI017 IEEE Spectrum Home Robot Safety Is All About Relationships
SI018 UNSW Newsroom Humanoid home robots are on the market. Do we want them?
SI019 The Robot Report Why humanoid robots aren’t advancing as fast as AI chatbots
SI020 iRobot / PR Newswire iRobot Reports Fourth-Quarter and Full-Year 2024 Financial Results
SI021 SEC irbt-20241228
SI022 CBS News Amazon calls off bid to buy iRobot. The Roomba vacuum maker will now cut 31% of its staff.
SI023 Nasdaq iRobot Reports Fourth-Quarter and Full-Year 2024 Financial Results
SI024 Sunday The helpful robotics company
SI025 Sunday Journal | Sunday Robotics | The helpful robotics company
SI026 iRobot iRobot Reports Third-Quarter 2025 Financial Results
SI027 NBER Robots and Firm Investment
SI028 PLOS One Smart home adoption factors: A systematic literature review and research agenda
SI029 Frontiers in Psychology Qualitative study on domestic social robot adoption and associated security concerns among older adults in Slovenia
SI030 iRobot Roomba® Robot Vacuum Cleaners | iRobot®
SE001 Sunday Sunday Robotics | The helpful robotics company
SE002 Sunday Technology | Sunday Robotics | The helpful robotics company
SE003 Sunday Beta Program
SE004 Sunday Company | Sunday Robotics | The helpful robotics company
SE005 Sunday Careers | Sunday Robotics | The helpful robotics company
SE006 Sunday ACT-1: A Robot Foundation Model Trained on Zero Robot Data | Sunday Robotics | The helpful robotics company
SE007 Sunday Sunday's Series B: No More Demos | Sunday Robotics | The helpful robotics company
SE008 Sunday Privacy Policy and Cookies | Sunday Robotics | The helpful robotics company
SE009 TechCrunch Humanoid robotics maker Sunday reaches $1.15B valuation to build household robots
SE010 Yahoo Finance Sunday Launches Memo, the Robot That Actually Learns Your Home
SE011 IT News Online Sunday Launches Memo, the Robot That Actually Learns Your Home
SE012 arXiv Diffusion Policy: Visuomotor Policy Learning via Action Diffusion
SE013 GitHub real-stanford/diffusion_policy
SE014 Columbia University project page Diffusion Policy: Visuomotor Policy Learning via Action Diffusion
SE015 GitHub tonyzhaozh/aloha
SE016 Mobile ALOHA project page Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation
SE017 ALOHA project page Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware
SE018 Tony Z. Zhao Tony Z. Zhao
SE019 arXiv ALOHA Unleashed: A Simple Recipe for Robot Dexterity
SE020 ALOHA Unleashed project page ALOHA Unleashed
SE021 IEEE Spectrum Home Humanoid Robots Survey Reveals Surprising Preferences
SE022 IEEE Spectrum Domestic Humanoid Robot Safety Standards Are Shifting
SE023 National Institute of Standards and Technology Robotics
SE024 International Organization for Standardization ISO 13482:2014
SE025 The Robot Report Why humanoid robots aren’t advancing as fast as AI chatbots
SU001 Sunday Sunday Robotics | The helpful robotics company
SU002 Sunday Sunday Beta Program Memo performs tasks autonomously in new homes. You will be able to schedule Memo to do tasks either through direct voice interaction or the companion Sunday app.
SU003 Sunday Sunday Technology
SU004 Sunday Sunday Company
SU005 Sunday Sunday Careers
SU006 Sunday Sunday Journal
SU007 Sunday Sunday's Series B: No More Demos Public deployment starts with our beta program, rolling out to users this Fall.
SU008 Sunday Privacy Policy and Cookies
SU009 Sunday Terms of Service
SU010 Sunday Cookies Policy
SU011 TechCrunch Humanoid robotics maker Sunday reaches $1.15B valuation to build household robots The company emerged from stealth late last year and already has 1,000 people on its waitlist, Bloomberg reports.
SU012 Tech Funding News Robots doing your chores: Sunday raises $165M at $1.15B valuation to deploy home robots
SU013 Yahoo Finance Sunday Launches Memo, the Robot That Actually Learns Your Home Starting November 19th, 2025, Sunday will accept applications for Memo’s Founding Family Beta, launching in late 2026. Fifty households will become early adopters, receiving individually numbered robots along with direct support and the opportunity to guide future capabilities.
SU014 Yahoo Finance Sunday Raises $165M to Launch First Autonomous Robots by Thanksgiving
SU015 Yahoo Finance Humanoid maker Sunday reaches $1.15B valuation to build household robots
SU016 IT News Online Sunday Launches Memo, the Robot That Actually Learns Your Home
SU017 GlobeNewswire Sunday Raises $165M to Launch First Autonomous Robots by Thanksgiving Since emerging from stealth last November, Sunday has seen strong demand, receiving thousands of applications for its upcoming Beta program.
SU018 Tracxn Sunday - Company Profile
SU019 The Robot Report Why humanoid robots aren't advancing as fast as AI chatbots Currently people have been doing this thing called teleoperation, where humans operate a robot like a puppet so it can perform tasks.
SU020 IEEE Spectrum Do People Really Want Humanoid Robots in Their Homes?
SU021 UNSW Newsroom / The Conversation Humanoid home robots are on the market – but do we really want them? According to the International Federation of Robotics, useful and widely accepted home androids may still be 20 years away.
SU022 Altman Solon Humanoid home robotics survey
SU023 Sunday Sunday sitemap.xml
SU024 Robotics & Automation News Sunday unveils new humanoid robot for the home
SU025 Robotics 24/7 VIDEO: Sunday launches Memo personal robot that actually learns your home
SU026 The Manila Times Sunday Launches Memo, the Robot That Actually Learns Your Home
SR001 Sunday Sunday Robotics | The helpful robotics company
SR002 Sunday Sunday Technology
SR003 Sunday Sunday Beta Program Memo performs tasks autonomously in new homes. You will be able to schedule Memo to do tasks either through direct voice interaction or the companion Sunday app.
SR004 Sunday Sunday Company
SR005 Sunday Sunday Careers
SR006 Sunday Sunday's Series B: No More Demos Public deployment starts with our beta program, rolling out to users this Fall.
SR007 Sunday ACT-1: A Robot Foundation Model Trained on Zero Robot Data | Sunday Robotics | The helpful robotics company
SR008 Sunday Privacy Policy and Cookies
SR009 Sunday Terms of Service
SR010 Sunday Cookies Policy
SR011 TechCrunch Humanoid robotics maker Sunday reaches $1.15B valuation to build household robots The company emerged from stealth late last year and already has 1,000 people on its waitlist, Bloomberg reports.
SR012 Yahoo Finance Sunday Launches Memo, the Robot That Actually Learns Your Home Starting November 19th, 2025, Sunday will accept applications for Memo’s Founding Family Beta, launching in late 2026. Fifty households will become early adopters, receiving individually numbered robots along with direct support and the opportunity to guide future capabilities.
SR013 GlobeNewswire Sunday Raises $165M to Launch First Autonomous Robots by Thanksgiving Since emerging from stealth last November, Sunday has seen strong demand, receiving thousands of applications for its upcoming Beta program.
SR014 Yahoo Finance Sunday Raises $165M to Launch First Autonomous Robots by Thanksgiving
SR015 Tech Funding News Robots doing your chores: Sunday raises $165M at $1.15B valuation to deploy home robots
SR016 IT News Online Sunday Launches Memo, the Robot That Actually Learns Your Home
SR017 Robotics & Automation News Sunday unveils new humanoid robot for the home
SR018 Yahoo Finance Humanoid maker Sunday reaches $1.15B valuation to build household robots
SR019 Altman Solon Humanoid home robotics survey
SR020 U.S. Bureau of Labor Statistics American Time Use Survey -- 2024 Results
SR021 IEEE Spectrum Do People Really Want Humanoid Robots in Their Homes?
SR022 IEEE Spectrum Domestic Humanoid Robot Safety Standards Are Shifting
SR023 UNSW Newsroom / The Conversation Humanoid home robots are on the market – but do we really want them? According to the International Federation of Robotics, useful and widely accepted home androids may still be 20 years away.
SR024 The Robot Report Why humanoid robots aren't advancing as fast as AI chatbots Currently people have been doing this thing called teleoperation, where humans operate a robot like a puppet so it can perform tasks.
SR025 National Institute of Standards and Technology Robotics
SR026 International Organization for Standardization ISO 13482:2014
SR027 Federal Trade Commission Privacy, Security, and Data Transfers If your company makes privacy promises ... the FTC Act requires you to live up to those claims.
SR028 Federal Trade Commission Consumer Privacy Think your company doesn't make any privacy claims? Think again — and reread your privacy policy to make sure you're honoring the promises you've pledged.
SR029 Federal Trade Commission Complying with COPPA: Frequently Asked Questions COPPA required the Federal Trade Commission to issue and enforce regulations concerning children's online privacy.
SR030 Cyber Law Monitor Cybersecurity Best Practices for AI-Powered Robotics Under State and Federal Privacy Laws Many providers of AI-powered robotics will be subject to U.S. state comprehensive privacy laws, U.S. state biometric privacy laws, and Federal Trade Commission (FTC) requirements.
SR031 Brookings Institution Protecting privacy in an AI-driven world This charade of consent has made it obvious that notice-and-choice has become meaningless.
SR032 Hill Dickinson Humanoid robots and the law: preparing for a new era of risk Who is responsible if a robot causes harm - the manufacturer, the operator, or the software provider?
SR033 Transformative AI Legal Leaps The Robots Are Coming: Navigating Privacy Challenges in AI-Powered Robotics in Public Settings and Homes (Part 1) These robots often come equipped with cameras, microphones and other sensors that collect data to perform their tasks.
SR034 JD Supra / DLA Piper Comprehensive Federal Privacy Legislation Introduced The SECURE Data Act proposes comprehensive federal privacy legislation based upon five broad themes.
SR035 iRobot / PR Newswire iRobot Reports Fourth-Quarter and Full-Year 2024 Financial Results
SR036 SEC irbt-20241228
SR037 CBS News Amazon calls off bid to buy iRobot. The Roomba vacuum maker will now cut 31% of its staff.
SR038 Nasdaq iRobot Reports Fourth-Quarter and Full-Year 2024 Financial Results
SR039 iRobot iRobot Reports Third-Quarter 2025 Financial Results
SV001 Sunday Sunday's Series B: No More Demos We just raised an oversubscribed $165M Series B at a $1.15B valuation to stop demos and bring the first autonomous home robot to households this year.
SV002 Sunday Sunday Beta Program You will be able to schedule Memo to do tasks either through direct voice interaction or the companion Sunday app.
SV003 Sunday Sunday Technology
SV004 Sunday Sunday Company
SV005 TechCrunch Humanoid robotics maker Sunday reaches $1.15B valuation to build household robots Sunday says it has raised $165 million at a $1.15 billion valuation in a Series B round led by Coatue Management.
SV006 Yahoo Finance / GlobeNewswire Sunday Raises $165M to Launch First Autonomous Home Robots by Thanksgiving Sunday announced it has raised an oversubscribed $165 million Series B funding round at a $1.15 billion valuation.
SV007 Yahoo Finance Sunday Launches Memo, the Robot That Actually Learns Your Home Fifty households will become early adopters, receiving individually numbered robots along with direct support and the opportunity to guide future capabilities.
SV008 Tech Funding News Sunday raises $165M Series B at $1.15B valuation to launch home robots
SV009 Yahoo Finance Humanoid maker Sunday reaches $1.15B valuation to build household robots The company emerged from stealth late last year and already has 1,000 people on its waitlist, Bloomberg reports.
SV010 IT News Online / GlobeNewswire Sunday Launches Memo, the Robot That Actually Learns Your Home
SV011 The AI Insider Household robot maker Sunday raises $165M in Series B funding with $1.15B valuation
SV012 Tracxn Sunday - 2026 Company Profile, Team, Funding & Competitors Sunday has raised a total funding of $200M over 2 rounds ... Its latest funding round was a Series B round on Mar 12, 2026 for $165M.
SV013 Altman Solon Humanoid Home Robotics Survey 69% of respondents are unwilling to pay more than $5,000 for a home robot.
SV014 IEEE Spectrum Do People Really Want Humanoid Robots in Their Homes? People generally prefer special-purpose robots over humanoids.
SV015 UNSW Newsroom / The Conversation Humanoid home robots are on the market – but do we really want them? According to the International Federation of Robotics, useful and widely accepted home androids may still be 20 years away.
SV016 IEEE Spectrum Home Robot Safety Is All About Relationships The proposed ISO update addresses hazard identification, risk assessment, and different use scenarios. It does not, however, set limits, propose testing methods, or have enforcement mechanisms.
SV017 The Robot Report Why humanoid robots aren't advancing as fast as AI chatbots There are warehouses in China and the U.S. where humans are being paid to do this.
SV018 SEC iRobot Annual Report on Form 10-K for fiscal 2024 Our total revenue for fiscal 2024 was $681.8 million, declining 23.4% from revenue of $890.6 million in fiscal 2023.
SV019 iRobot / PR Newswire iRobot Reports Fourth Quarter and Full-Year 2024 Financial Results For full-year 2024, GAAP and Non-GAAP gross margins were negatively impacted by a $26.6 million non-recurring charge related to the write-off of excess component inventory.
SV020 iRobot iRobot Reports Third Quarter 2025 Financial Results At this time, Company has no sources upon which it can draw for additional capital.
SV021 CBS News Amazon-iRobot merger is canceled, Roomba vacuum-maker to lay off 31% of staff
SV022 iRobot Roomba Robots Roomba® 105 Vac Robot + AutoEmpty™ Dock ... $249.99 ... Roomba Combo® 10 Max robot + AutoWash™ dock ... $1,399.99.
SV023 1X 1X secures $100M in Series B funding 1X ... has raised $100 million in Series B funding.
SV024 1X NEO Give your NEO a list of chores, schedule a time you want them done and come back to a cleaner home everyday.
SV025 The Robot Report 1X Technologies raises $100M Series B to advance NEO humanoid robot 1X noted that it has now raised over $125 million in less than 12 months.
SV026 Figure Series C We have exceeded more than $1 billion in committed capital through our Series C financing round, at a post-money valuation of $39 billion.
SV027 Figure Figure announces strategic partnership with Brookfield Brookfield ... has over 100,000 residential units under management.
SV028 Figure Helix The home presents robotics' greatest challenge.
SV029 TechCrunch Figure will start alpha testing its humanoid robot in the home in 2025 Figure ... will begin “alpha testing” its Figure 02 robot in the home setting later in 2025.
SV030 Samsung Newsroom [Video] CES 2024: A Day in the Life with Ballie, an AI Companion Robot for the Home Ballie acts as a personal home assistant, autonomously driving around the home to complete various tasks.
SV031 1X 1X
SV032 U.S. Bureau of Labor Statistics American Time Use Survey -- 2024 Results On an average day, 80 percent of people engaged in household activities ... spending about two hours on these activities.
SV033 Apptronik Apollo Apollo will operate in warehouses and manufacturing plants in the near term eventually extending into ... home delivery, elder care and countless more areas.
SV034 CNBC Apptronik raises $520 million at $5 billion valuation for Apollo robot Humanoid robotics startup Apptronik was valued at $5 billion in a funding round that included capital from Google.
SV035 Unitree G1 Price (Tax and Shipping cost excluded) US $13.5K.
SV036 Unitree H1 / H1-2 Unitree's first universal humanoid robot.
SV037 Unitree Unitree Robotics
SV038 Skild AI Skild AI Our AI can execute low-level skills like grasping, handover, and navigation on mobile platforms.
SV039 Physical Intelligence Physical Intelligence (π) Physical Intelligence is bringing general-purpose AI into the physical world.
SV040 Robozaps Humanoid Robot Industry Report 2026