初创公司尽调
尽调报告 healthcare / biotechnology Seed extension / early platform build 2026-07-03

Astromech

Colossal 剥离公司,战略估值支撑罕见,但公开商业验证很少。

Astromech 可能正在搭建差异化的生物 AI 平台,且与 Colossal 的毗邻关系很强,但公开记录仍缺少客户、收入和验证证据,难以确信承销据称 $2 billion 估值。

封面要素

总部 01
Austin, Texas [CO004]
阶段 02
Seed extension / early platform build [CO023]
融资总额 03
40.5 USDm [CO022]
最新估值 04
2000 USDm [CO024]
母公司背景 05
Spun out of Colossal Biosciences [CO013]

公司概况

Astromech 是一家位于 Austin 的私营生物 AI 初创公司,从 Colossal Biosciences 剥离出来,由 Ben Lamm 和 George Church 创立。公司称其平台为 Large Life Model,结合进化建模、多物种数据、泛基因组学,以及未来的多组学整合,服务医学、农业和可编程生物学应用。公开披露确认其 2025 年融资 $30 million、2026 年种子轮延伸融资约 $10.5 million;多家 2026 年媒体报道其估值约 $2 billion,但收入、客户和单位经济披露基本缺位。

官网
astromech.com
创始人
Ben Lamm, George Church
创立地点
Austin, Texas
总部
Austin, Texas
产品
一个生物 AI 平台,定位为用于多物种进化、基因组与多组学预测的 Large Life Model,而不是单一狭窄的药物发现工作流工具。
客户
目标用户可能是大型药企、生物科技公司、农业生物技术公司和研究机构,它们需要预测生物学工具;但公司尚未公开任何付费客户。
商业模式
可能是以合作伙伴为主的平台授权、里程碑或研究服务模式,但公司未公开定价或合同结构。
阶段
Seed extension / early platform build
融资情况
公开披露融资 $40.5 million,来自 2025 年 Form D 和 2026 年种子轮延伸融资;多家 2026 年媒体报道估值约 $2 billion。
[CO001, CO002, CO004, CO011, CO012, CO013, CO022, CO024]

执行摘要

主要优势

  • 创始人履历很强,且公司从 Colossal Biosciences 孵化而来,这让一家极年轻的平台公司拥有少见的战略可信度。
  • 公开技术叙事围绕进化生物学、泛基因组学和多组学展开,让 Astromech 在 AI 生物学领域有一个概念上差异化的切入角度。
  • 公司披露融资额不高,却已经拿到远高于融资额的战略估值信号,说明投资人相信其期权价值和平台杠杆。

主要风险

  • 公开资料没有收入、客户、留存或员工数披露,无法验证据称估值。
  • 公司的差异化几乎仍停留在公司自述,缺少独立技术或商业证据做基准验证。
  • 与 Colossal 的治理、股权结构和公司间依赖条款均未公开披露。
  • 生物 AI 面临的生物安全、监管和数据治理预期升温,速度快于 Astromech 公开披露控制措施的进度。

未决问题

  • 当前收入、ARR、客户数量,以及任何具名付费或生产阶段合作。
  • Large Life Model 相关主张的独立技术基准、模型验证证据或案例证明。
  • 股权结构、优先股堆叠,以及 Astromech 与 Colossal Biosciences 之间任何 IP / 数据共享或服务协议。
  • 当前员工数、烧钱速度、现金跑道,以及下一轮融资触发条件。

目录

Chapter 01

01公司概览

1.1 身份与运营模式

Astromech 将自己定位为一家生物 AI 公司,打造所谓「Large Life Model」,而不是传统药物发现单点工具。官网把系统描述为一层自主生物智能,用来建模多物种转录组、进化和结构数据,帮助研究人员理解「生命代码」,不只是做摘要。第三方报道与这一定位一致,称 Astromech 是一个预测生物学引擎,试图在进化、疾病风险和生物脆弱性变得临床可见之前做出预测。尽调上,实际结论是:Astromech 更适合被理解为一家早期平台公司,站在 AI 基础设施、计算生物学和合成生物设计的交叉点。这带来异常宽的可选性,但也意味着截至 2026-07-03,公司叙事仍明显更像平台愿景,而非商业验证。[CO001, CO002, CO003, CO004, CO005, CO006]

快照 KPI 表
指标数值 / 状态日期置信度缺口
成立2025 年从 Colossal 拆分2025-08-12具体成立月份由首份 Form D 和媒体上线时点推断。
总部德克萨斯州奥斯汀2026-03-26
当前阶段种子延伸轮 / 产品前平台2026-07-03公开资料未记录种子延伸轮之外的定价机构轮次标签。
已披露融资额(USDm)40.52026-03-26
最新公开估值(USDm)20002026-04-07头条估值来自媒体报道和创始人表述,不是已备案的定价轮文件。
当前收入 / ARR未披露2026-07-03要求提供董事会口径指标或收入桥表。
当前客户数未披露2026-07-03要求提供活跃合作和付费项目名单。
当前员工数未披露2026-07-03要求提供组织架构图或薪资名册摘要。
母公司背景从 Colossal Biosciences 拆分2026-04-29
核心技术框架Large Life Model / 生物操作系统2026-07-03

仅使用公开披露;未披露经营指标明确留空,不作推断。

[CO001, CO004, CO013, CO019, CO020, CO022]
FO002: 公司快照逻辑

Astromech 的故事把 Colossal 来源的数据背景、创始人信誉、生物 AI 雄心和远超公开经营证据的估值串在一起。

[CO002, CO003, CO005, CO006, CO013, CO021]

1.2 创始人、剥离背景与治理

Astromech 的公开叙事与 Ben Lamm 和 George Church 深度绑定。Lamm 是 Colossal Biosciences 背后的运营型创业者;Church 则凭合成生物学、基因组学及更广泛的 Harvard 相关研究提供科学可信度。Colossal 自身文章和外部报道都把 Astromech 描述为有意剥离的公司,而不是偶然的副项目;Colossal 的去灭绝项目提供了技术和数据富集的母体背景。重要之处在于,Astromech 可能同时继承 Colossal 的声誉杠杆和关键人物集中风险。公开来源确认了 Delaware 注册、Austin 办公地址,以及最初 Form D 中披露的投资人数量很少,但没有披露完整董事会、治理权、股权拆分或独立监督结构。尽调含义是,目前治理可信度更多押在创始人质量和母公司声誉上,而不是独立可见的公司控制。公开招聘信号显示职能覆盖较宽,但仍弱于具名高管名单或清晰披露的董事会结构。[CO011, CO012, CO013, CO014, CO015, CO016]

领导层与创始人表
人物角色公开适配证据外部锚点依赖判断
Ben Lamm联合创始人 / CEOColossal 出身的运营型创始人,也是 Astromech 主要对外发言人Colossal、D Magazine 与 SEC Form D
George Church联合创始人 / 顾问在基因组学和合成生物学上具备科学可信度Colossal 与 D Magazine
独立董事会披露未公开公开来源未列明独立董事会结构公开来源缺口
运营梯队除招聘信号外未公开官网显示 AI、基因组学、平台岗位都在招聘Astromech 招聘内容
母公司支持隐含但重要拆分叙事暗示可接入 Colossal 平台和网络Colossal 拆分文章

创始人在公开层面可见度高,但治理深度和当前运营梯队披露不足。

[CO011, CO012, CO014, CO015, CO016, CO017]
利益相关方 / 投资者图谱
利益相关方角色公开资料可支持什么为什么重要尽调要求
Colossal Biosciences母公司 / 拆分来源Astromech 被描述为 Colossal 拆分公司,建立在同一套生物基础设施之上可能提供数据、人才和战略信号厘清数据、IP 和服务协议。
Ben Lamm创始人兼运营者Colossal 与 Astromech 的公开门面同时带来杠杆和关键人物风险要求说明时间分配和治理权利。
George Church科学创始人 / 顾问具名联合创始人和科学架构师增强科学可信度要求说明持续参与范围。
初始投资者未完全披露2025 年原始 Form D 显示一名投资者投资团构成透明度有限要求提供股权结构表和投资者权利摘要。
种子延伸轮投资者仅由媒体部分披露媒体称 $10.5M 种子延伸轮使总额达到 $40.5M估值支撑取决于谁出资、条款是什么要求提供签署版投资条款清单和优先权层级。

投资者披露不完整;本表映射公开可见度最影响投资判断的利益相关方。

[CO013, CO018, CO019, CO020, CO021, CO024]

1.3 融资、阶段与规模信号

融资历史是外部最容易核验的规模信号。Astromech 首份公开 Form D 显示 2025 年 8 月完成 $30 million 股权融资;2026 年 3 月 Form D 又增加约 $10.5 million,披露资本合计 $40.5 million。多家 2026 年媒体把这笔不大的资本基础与约 $2 billion 估值放在一起,称这家很年轻的剥离公司在九个月内达到独角兽状态。融资额与隐含估值之间的差距,是投资判断的核心:它表明投资人承销的是战略期权价值、专有科学背景和 Colossal 邻近性,而不是当前商业指标。公开来源没有披露收入、ARR、当前客户数或当前员工数。因此,当前可支持的阶段标签不是「已规模化商业生物科技公司」,而是「披露很少的种子轮延伸 / 早期平台建设」,但估值已经计入相当多的未来成功。[CO019, CO020, CO021, CO022, CO023, CO024]

FO003: 快照 KPI

公开 KPI 显示,公司仍极早期,但对外估值很大,已披露经营分母很薄。

编辑部 0–5 分评分衡量披露质量,不衡量公司质量。

[CO011, CO019, CO020, CO022, CO023, CO024]

1.4 里程碑与负面筛查

公开且带日期的时间线很短,但有意义。到 2025 年 8 月,Astromech 已提交首份 Form D,相关媒体报道将公司描述为一家追求预测建模的 AI 生物技术平台。到 2026 年 3 月和 4 月,公开报道收敛到同一组头部事实:公司追加 $10.5 million 种子轮延伸融资,估值达到 $2 billion,并把自己定位为服务医学、农业和可编程生物学的平台。主要负面信号不是诉讼或产品事故,而是证据质量本身:公司仍处在无收入、低披露状态;外部报道常重复创始人的叙事,而不是给出独立商业验证。公开法律来源浮现了公司注册和网站政策,但截至 runDate 没有披露诉讼或制裁。更广义地看,关于合成生物学规模化的外部战略研究提醒,平台公司可能先积累叙事价值,再证明持久商业执行。这里应把它当作轻度筛查信号,而不是法律干净或运营成熟的证明。[CO029, CO030, CO031, CO032, CO033, CO034]

里程碑表
日期事件类型金额 / 状态参与方影响
2025-08-12首份 Form D 提交融资发行规模 $30.0M;已售 $29,999,998Astromech AI Corp.标记首个清晰公开融资记录。
2025-08-12披露奥斯汀主要办公地址规模1401 Lavaca Street, Unit 155Astromech AI Corp.为当前总部证据提供锚点。
2025-08-12文件中列明 Ben Lamm 为 CEO、George Church 为联合创始人 / 顾问治理领导层已披露Astromech AI Corp.确认创始人结构。
2025-08-12媒体报道将 Astromech 定位为 AI 生物技术平台产品预测生物学定位Yahoo Finance / FirstWord Pharma释放早期外部叙事。
2026-03-26第二份 Form D 提交融资追加股权 $10.5MAstromech AI Corp.已披露资本增至 $40.5M。
2026-03-31D Magazine 特写详述尚未产生收入状态和应用规模尚未产生收入,报道估值 $2BD Magazine / Ben Lamm显示估值跑在商业披露之前。
2026-04-07Axios 报道种子延伸轮和 $2B 估值融资总额 $40.5M;估值 $2BAxios / Ben Lamm强化价格与披露之间的极端落差。
2026-04-29Colossal 发布拆分定位文章合作Astromech 被描述为 Colossal 最新的独角兽拆分公司Colossal / Ben Lamm 与 George Church将 Astromech 绑定到更广的平台商业化模式。
2026-04-29GamesBeat 和 USA Today 报道拓宽公众叙事规模结束隐身 / 独角兽叙事GamesBeat / USA Today可见度扩展到小众生物技术圈之外。
2026-07-03公开来源法律筛查仍较浅不利保留来源中未发现公开诉讼或制裁尽调审查未见证据,不等于事实不存在。

由于未找到单独的成立新闻稿,成立时间通过首次公开融资和发布报道跟踪。

[CO013, CO019, CO020, CO021, CO029, CO030]
FO001: 公司里程碑时间线

Astromech 的公开记录集中在 2025 年 8 月的融资披露,以及 2026 年 3–4 月的估值集中曝光。

日期采用留存来源中可见的申报日期或发布日期。

[CO019, CO020, CO021, CO022, CO029, CO030]
Chapter 02

02市场分析

2.1 市场边界与现状替代方案

Astromech 的可服务支出落在三个相邻预算池内,而不是一个清晰品类:(1)药企和生物科技研发团队购买的 AI 药物发现和生物制剂设计工具;(2)药企及农业 / 工业生物技术买家购买的合成生物学与计算生物学基础设施;(3)更广义的预测生物学或生物安全相邻分析,可能服务政府和国防相邻买家。纳入范围包括 R&D 软件授权、计算服务和多组学数据平台;排除范围包括湿实验 CRO 执行、制造和临床试验运营,Astromech 似乎并不直接提供这些。多数买家的现状替代方案,是内部生物信息学团队、单点 AI 供应商(Recursion、Insilico、Schrödinger、Isomorphic Labs),以及传统高通量筛选或基于结构的筛选。公开来源没有显示 Astromech 指名某一目标买家预算科目,所以这里的边界来自平台定位和可比公司归类,而不是公司披露的 TAM 声明。[CM001, CM002, CM003, CM004, CM005, CM006]

市场定义表
细分 / 品类纳入支出排除支出买方 / 付款方与 Astromech 的匹配度
AI 驱动药物发现工具R&D 软件许可、模型访问、计算服务湿实验 CRO 执行、临床试验成本药企 / 生物技术 R&D 负责人直接——最贴近 Astromech 声称的平台雄心
合成生物学 AI 基础设施设计 / 仿真工具、菌株和通路工程软件发酵制造资本开支工业生物技术、农业生物 R&D相邻——契合可编程生物学叙事
多组学数据平台数据集成、模型训练基础设施、分析许可原始测序 / 湿实验生成成本药企生物信息学、产学联盟相邻——契合 Large Life Model 叙事
预测 / 生物安全分析风险建模软件、面向政府的分析合同实体生物监测基础设施政府、生物防御邻近项目推测——治理研究者提出,公司未确认
现状替代方案内部生物信息学团队、传统筛选预算N/A — 内部成本中心药企 R&D争夺同一笔预算

边界由平台定位和可比公司分类推断;Astromech 未公开披露具体目标预算线。

[CM001, CM002, CM003, CM004, CM005, CM006]

2.2 用多重视角测算市场

这里没有单一 TAM 数字可靠,因此本章用多组视角交叉校准。仅 AI 药物发现细分市场,分析师对 2024-2025 年的估算集中在 $2-3B 区间,并以约 28-46% CAGR 增长;到 2030 年代早中期,视方法和相邻服务纳入情况不同,规模可达 $10-20B 以上。MarketsandMarkets 使用的 AI 生物技术框架更大,覆盖诊断和制造等超出 Astromech 表观范围的用例。另一个 AI 合成生物学子细分市场目前估计为数亿美元,到 2030 年代初扩大数倍;同时 OECD 和 BCG 都认为,合成生物学的工业转化真实存在,但速度慢于标题 TAM 暗示。因而,对 Astromech 这样的商业化前平台而言,可辩护的 SAM 最好被视为一条受证据约束的窄楔子——近期可服务的药企 / 农业生物 R&D 工具支出可能只有低个位数十亿美元——而不是完整的数百亿美元级标题 TAM;在公司披露首批付费项目之前,SOM 实际上无法定义。[CM007, CM008, CM009, CM010, CM011, CM012]

TAM/SAM/SOM 或规模测算视角表
发布方年份地域价值(USD)CAGR方法置信度局限
MarketsandMarkets2025全球生物技术中的 AI,数十亿美元级,范围宽~25-30%自上而下,包含诊断 / 制造宽于 Astromech 表面范围
The Business Research Company2026全球药物发现 AI,基准年约 $2-3B~30-40%自上而下的分析师模型基准年数字未独立验证
Global Market Insights 研究机构2025全球药物发现 AI,基准年为低个位数十亿美元~28-35%,至 2035 年自上而下的分析师模型预测期长,提高不确定性
Research and Markets 研究机构2025全球药物发现 AI,量级相近~30-46%自上而下的分析师模型CAGR 区间宽,说明方法分歧
Custom Market Insights 研究机构2025全球合成生物学 AI,基准年数亿美元级~30%+,至 2034 年自上而下的分析师模型比药物发现小得多,也更新
证据约束 SAM(本报告)2026全球近期可服务切口为低个位数十亿美元未独立估算自下而上把分析师 TAM 收窄到 Astromech 可能服务的药企 / 农业生物 R&D 工具支出SOM 未定义——Astromech 未披露任何管线

分析师数字按区间 / 量级呈现,因为保留来源的取整口径不同;应视为方向信号,不是精确值。

[CM007, CM008, CM009, CM010, CM011, CM012]
FM001: 市场规模测算视角

AI 药物发现及相邻合成生物学工具的分析师 TAM 估算,一旦收窄到 Astromech 可能服务的制药 / 农业生物技术 R&D 工具支出,就会大幅缩小。

分层数值是分析师区间的方向性中点,不是精确点估计。

[CM007, CM008, CM009, CM012, CM013, CM006]
FM002: 市场估算区间

AI 药物发现工具的分析师 CAGR 估算约在 28% 到 46% 之间,说明方法分歧显著,而不是只有一个共识增长率。

上下限取自留存分析师对同一广义类别 CAGR 估算的低端和高端;单位统一为 CAGR 百分比。

[CM009, CM010, CM011]

2.3 买家、用户与付款方分层及采用路径

潜在买家横跨大型药企 R&D 和商务拓展团队(平台授权和联合发现交易的预算持有人)、生物科技和农业生物运营团队(用湿实验结果评估模型输出的用户),以及考虑到治理研究者提出的双重用途框架,可能还有政府 / 生物安全相邻项目。该领域可比公司呈现可重复的采用路径:先做平台展示和数据合作,再进入付费试点或共同开发里程碑交易,随后转向里程碑 / 权利金经济或更宽的多年平台授权。公开证据显示,多个可比公司(Recursion、Schrödinger、Isomorphic Labs、LabGenius)正是以这种方式与具名药企伙伴转化;在尚无 Astromech 已披露客户的情况下,这也是 Astromech 交易结构最清晰的类比。预算归 R&D / BD 领导层,而不是 IT 采购;这通常拉长销售周期,但一旦签约,合同耐久度更高。[CM014, CM015, CM016, CM017, CM018, CM019]

细分市场 / 买方图谱
细分市场买方用户付款方工作流预算负责人采用触发因素
大型药企 R&D商务拓展 / R&D 负责人实验台科学家、计算生物学家药企 R&D 预算平台评估 → 付费试点 → 里程碑交易R&D/BD 负责人内部工具无法满足的具名靶点或模态缺口
中型生物技术公司CSO / 发现负责人药物发现科学家生物技术公司运营预算合作或平台许可谈判CSO 办公室需要压缩先导项目发现周期
农业生物 / 工业生物技术R&D 或创新负责人性状工程师、育种人员农业生物 R&D 预算性状设计试点 → 田间验证创新 / R&D 负责人需要加速性状发现周期
政府 / 生物安全邻近项目办公室(推测)分析师、风险建模人员政府合同载体未见证据——仅从治理文献推断项目办公室(推测)未见证据——仅从治理文献推断
学术 / 联盟资助项目 PI博士后、研究人员课题 / 联盟资金数据合作 → 共同发表资助办公室获取新的多组学 / 模型能力

大型药企之外的行,是依据可比公司的商业化落地模式外推,不来自 Astromech 的特定披露。

[CM014, CM015, CM016, CM017, CM018, CM019]
FM003: 买方 / 细分市场地图

各买方细分市场里,AI 生物工具的预算归 R&D/BD 负责人管,但使用者和付款方会随细分市场成熟度变化。

[CM014, CM015, CM016, CM017, CM018, CM020]

2.4 增长驱动与采用约束

增长驱动因素包括:单次模型训练的计算成本下降,多组学和结构生物学数据集扩大,药企成本压力使 AI 增强发现比蛮力筛选更有吸引力,以及农业和生物防御对可编程生物学的跨行业兴趣上升。约束同样实质:FDA、EMA 等监管机构仍在为监管决策中的 AI 使用建立共同原则,这会拖慢最高价值治疗用例的采用;药企合作中的切换成本和数据权利谈判很重;ROI 证明需要多年临床或田间验证,多数 AI 生物学供应商尚未跨过这一关;即便是软件优先的平台,湿实验验证循环的资本强度仍然高。治理研究者也把生物安全和双重用途审查标记为新出现的采用门槛,而不是已经稳定的经营成本;这可能给 Astromech 自身定位触及的预测进化和病原体相关应用额外增加摩擦。[CM020, CM021, CM022, CM023, CM024, CM025]

增长驱动与约束表
驱动因素 / 约束方向时间含义尽调问题
AI 训练 / 推理成本下降驱动因素持续降低生物模型迭代成本确认 Astromech 的算力成本结构
药企成本压力利好 AI 辅助发现驱动因素近期提升买方试点新平台的意愿索取正在推进的药企沟通清单
多组学 / 结构数据集增长驱动因素持续提升模型质量和防御性确认 Astromech 持有的数据访问协议
跨行业对可编程生物学的兴趣驱动因素中期将 TAM 从药企扩展到农业生物 / 工业领域索取具名非药企试点,如有
AI 用于药物开发的监管不确定性约束近期至中期放慢最高价值治疗用例的落地跟踪 FDA/EMA 指引定稿时间线
高切换成本 / 数据权利谈判约束持续拉长企业销售周期索取可比交易的平均成交周期
ROI 证明需要多年验证约束持续推迟收入确认和标杆客户形成索取任何湿实验验证项目的状态
生物安全 / 双重用途治理审查约束新兴可能专门卡住预测性进化和病原体相关用例索取 Astromech 的生物安全审查政策,如有

约束来自监管机构和治理研究者来源,而非公司披露。

[CM020, CM021, CM022, CM023, CM024, CM025]
FM004: 采用漏斗或价值链地图

可比公司模式显示,典型 AI 生物平台的采用路径会从宽泛的平台认知,收窄到少数带里程碑的合作关系。

百分比仅作示意,来自可比公司披露模式,而非 Astromech 专属漏斗数据。

[CM015, CM019, CM022, CM023]

2.5 市场规模与采用尽调缺口

几个市场规模和采用问题无法用公开证据关闭。没有来源像 Astromech 定义自身使命那样单独划出「AI 预测进化生物学」品类,因此这里所有市场数字都来自相邻品类代理,而不是直接匹配。分析师估算在当前年度市场规模和未来 CAGR 上也相差超过 2 倍,取决于是否纳入诊断、制造和服务;留存的分析师报告中,没有一份披露足够大的一手调研样本,可独立验证基年数字。采用路径证据完全来自可比公司,而不是 Astromech 自身披露的销售管线,因此本章的买家分层应被读作行业典型模式,而不是已确认的 Astromech 商业化动作。这些缺口重要,因为它们意味着本报告后面的估值工作不能依赖单一、经来源验证的 TAM;估值必须从区间推理,并明确标出规模不确定性。[CM027, CM028, CM029, CM030, CM031, CM032]

Chapter 03

03竞争格局

3.1 直接与相邻竞争者图谱

竞争集合覆盖几种不同姿态。直接同行是建设通用 AI 生物学或药物设计平台的公司,包括 Recursion Pharmaceuticals、Insilico Medicine、Isomorphic Labs 和 Schrödinger。相邻玩家聚焦蛋白 / 抗体设计和生成式生物学,包括 Generate:Biomedicines、Absci 和 LabGenius。BenevolentAI 代表上一代知识图谱路线,如今转向更窄的药企合作。Relay Therapeutics 则代表基于动力学的结构生物学替代技术路线,同样瞄准药物发现问题。现状替代仍是药企内部生物信息学团队,以及传统高通量或基于结构的筛选;这在多数在位药企中仍是默认选择。内部自建是现实威胁,因为大型药企 R&D 组织越来越多招聘内部计算生物学团队,而不是直接购买平台访问。潜在进入者包括转向生物学的资金充足 AI 实验室(沿 Isomorphic Labs 和 NVIDIA BioNeMo 模式),以及其他 Colossal 相邻剥离公司。[CP001, CP002, CP003, CP004, CP005, CP006]

竞争对手画像表
竞争对手类别规模 / 融资目标细分市场差异化局限
Recursion Pharmaceuticals直接竞争——AI 药物发现平台上市公司(Nasdaq: RXRX);已披露合作伙伴名单大型药企研发大规模表型筛选加 AI 模型临床转化要拿出证据仍需多年,路径还在形成
Insilico Medicine直接竞争——AI 药物发现平台通过香港上市;已披露财务数据大型药企、生物技术公司端到端生成式 Pharma.AI 平台相比平台主张,临床阶段验证仍偏早
Isomorphic Labs直接竞争——AI 药物设计平台大额 Series B;Alphabet 支持大型药企(J&J 等)源自 AlphaFold 的物理 / 结构建模谱系分拆阶段;商业交易成熟度仍在形成
Schrödinger直接竞争——基于物理的计算平台上市公司(Nasdaq: SDGR);已披露收入药企、生物技术、材料科学长期积累的基于物理的模拟可信度软件加药物管线的双重模式,削弱纯平台可比性
Generate:Biomedicines相邻竞争——生成式蛋白设计大额私募轮;已披露投资者关系页面生物技术公司、药企合作伙伴用于 de novo 蛋白设计的生成式生物学模态聚焦更窄,不及 Astromech 的宽平台主张
Absci相邻竞争——生成式蛋白 / 抗体设计上市公司;整合湿实验 AI药企合作伙伴湿实验 + 生成式 AI 设计闭环规模小于 Recursion/Schrödinger 同业组
LabGenius相邻竞争——抗体发现平台£35M Series B 轮;具名药企合作伙伴(Sanofi)生物制药抗体项目自动化进化抗体工程(EVA 平台)范围更窄,仅限抗体
BenevolentAI相邻 / 较早一代——知识图谱平台公开上市;战略转向药企合作伙伴(如 AstraZeneca)深厚的知识图谱和文献挖掘积累公开报告显示,其早期受挫后已战略转向
Relay Therapeutics相邻竞争——基于动态的结构生物学上市公司;已披露管线肿瘤及其他治疗领域药企蛋白运动 / 动态的计算路径技术切入点不同于进化 / 多组学叙事
Astromech直接竞争——生物 AI / Large Life Model 平台已披露融资 $40.5M;据报道估值约 $2B未披露——广泛覆盖医药、农业、生物学多物种进化和多组学建模主张未披露药企合作伙伴、收入,或相对同业的基准测试

Astromech 行用于直接对比;所有竞争对手行只使用公开披露。

[CP001, CP002, CP003, CP008, CP009, CP010]

3.2 竞争者画像:规模、融资与战略方向

几家竞争者在规模和披露上显著高于 Astromech。Recursion 和 Schrödinger 都是上市公司,披露收入、市值和多年药企合作历史。Insilico Medicine 完成香港上市,并披露财务数据。Generate:Biomedicines 和 Isomorphic Labs 完成了大型后期私募或战略轮融资,披露了具名投资人和药企合作者(Isomorphic Labs 的 Series B 轮融资报道同时提到 Alphabet 持续支持)。LabGenius 和 Absci 处在更窄的抗体 / 蛋白设计利基,披露融资额较小,但有具体药企合作验证点(LabGenius-Sanofi;Absci 的上市公司披露)。相对这一组,Astromech 是披露最少的竞争者:没有具名药企伙伴,没有披露收入;相对于已披露融资额,其估值比例也高于本表任何同行。[CP008, CP009, CP010, CP011, CP012, CP013]

FP001: 竞争定位图

相较已上市和资金充足的私营同行,Astromech 的战略叙事 / 估值很强,但公开商业证据很弱。

坐标轴评分是有证据支撑的 0-10 序数判断,不是独立基准测试的绩效指标。

[CP008, CP009, CP010, CP011, CP012, CP013]

3.3 能力、定价与 GTM 对比

能力层面,竞争者按技术路线分化:Recursion 和 Insilico 强调大规模表型与生成式筛选;Isomorphic Labs 和 Schrödinger 强调物理和基于结构的建模;Generate:Biomedicines、Absci 和 LabGenius 强调带湿实验反馈循环的生成式蛋白 / 抗体设计;Astromech 强调多物种进化和多组学建模,基于公开证据,这在同侪中是相对更少验证的角度。整个品类很少公开定价;多数竞争者靠平台授权加里程碑 / 权利金交易结构变现,而不是透明标价,Astromech 则完全没有披露定价或交易结构信息。GTM 和分销方面,公开披露的竞争者通过具名商务拓展合作触达药企买家;其中几家公司还借助公开市场投资者可见度,形成 Astromech 当前缺少的分销和可信度通道。[CP015, CP016, CP017, CP018, CP019, CP020]

功能 / 能力矩阵
购买标准AstromechRecursionIsomorphic LabsSchrödingerGenerate:Biomedicines / Absci / LabGenius
多物种进化建模已主张(未基准测试)未声明为重点未声明为重点未声明为重点未声明为重点
大规模表型 / 图像筛选未知 / 未披露核心强项(已披露)非主要重点非主要重点非主要重点
基于物理 / 结构的建模未知 / 未披露次要核心强项(已披露)核心强项(已披露)非主要重点
生成式蛋白 / 抗体设计未知 / 未披露次要次要次要核心强项(已披露)
具名药企合作证明未披露已披露(多个)已披露(J&J 及其他)已披露(多个)已披露(Sanofi 及其他)
公开财务披露无(私有公司)上市公司私有公司,已披露大额融资轮上市公司混合——Absci 上市,其他为私有公司

标注“未知 / 未披露”的单元格,表示没有公开证据,并非确认能力缺口。

[CP015, CP016, CP017, CP018]
定价 / 打包比较
竞争对手价格 / 单位 / 合同模式包含能力折扣 / 未知项含义
Recursion平台许可 + 里程碑 / 特许权使用费(由合作结构推断)模型访问、联合发现项目具体定价不公开该类别的标准结构
Isomorphic Labs战略合作 + 里程碑经济共同开发访问、模型输出交易条款不公开高接触合作模式,而非自助式
Schrödinger软件许可(作为上市公司收入线披露)+ 内部管线经济模拟软件加内部药物项目已披露分部收入,未披露单笔交易定价软件 / 管线双重模式让逐项定价比较更复杂
LabGenius / Absci / Generate:Biomedicines合作 + 里程碑 / 特许权使用费(推断)平台访问加设计服务交易经济不公开符合更广泛类别惯例
Astromech未披露未披露不存在公开定价、打包或交易结构信息相对同业组,定价完全不透明

该组没有任何竞争对手发布透明标价;数字依据已披露合作结构推断。

[CP016, CP019, CP020]
FP002: 功能广度 / 能力地图

Astromech 声称的能力(跨物种进化建模)与任何竞争对手披露的核心强项都无法完全重叠;差异化仍未基准验证,而非已经证明。

[CP015, CP017, CP018, CP019]

3.4 切换成本、锁定与分销权力

该品类的切换成本来自数据集成深度和多年合作结构,而不是简单的软件锁定。一旦药企伙伴把某个竞争者的模型输出接入正在推进的发现项目,即使新进入者提出更强技术主张,替换供应商也会变贵;这在结构上有利于 Recursion、Schrödinger 和 Isomorphic Labs 等已建立合作的在位者,也不利于包括 Astromech 在内的新进入者。多供应商并用(买家在不同项目中使用多个 AI 生物学供应商)在大型药企中似乎很常见,从竞争者披露的具名合作多样性即可看出;这降低了 Astromech 赢得首个试点的门槛,但提高了赢得排他性的门槛。分销权力不对称:上市公司竞争者可以持续利用投资者关系、已发布财务结果和分析师覆盖来提供可信度与漏斗顶部信号,而 Astromech 作为低披露私营公司,目前无法复制。[CP021, CP022, CP023, CP024]

3.5 护城河耐久性与竞争者反向证据

这一组竞争者没有哪一家拥有已经稳固且能抵御反证的护城河:BCG 关于工业规模化的研究明确警告,合成生物学和 AI 生物学商业化主张整体跑在已证明工业结果之前;这一提醒适用于整个竞争集合,不只针对 Astromech。公开证据没有显示任何 AI 原生新进入者正面对面替代在位者;相反,多数具名交易看起来是对现有药企 R&D 预算的增量,而不是替代。商品化风险正在上升,因为更多资本充足的进入者采用相似的生成式和多组学技术;这可能比任何单家公司路线图更快压缩整个品类的差异化,包括 Astromech。Astromech 的具体护城河主张——多物种进化建模——没有公开的、相对任何同行的正面对比基准;这本身就是本章最重要的反向发现:差异化被提出了,但还没有被证明。[CP025, CP026, CP027, CP028, CP029]

护城河耐久度 / 竞争风险登记表
护城河主张威胁严重程度缓释 / 尽调问题
多物种进化 / 多组学建模宽度没有任何公开基准测试可对比具名同业索取头对头基准数据或独立技术验证
Colossal 相邻性和创始人可信度声誉光环不能保证技术或商业差异化尽调中把创始人可信度与已展示的产品优越性拆开
“预测性进化生物学”的先发叙事分析机构尚未命名该类别,独立来源也未验证其足够独特确认是否有竞争对手用不同名称追求等价能力
整体类别商品化风险BCG 和 OECD 研究显示,宽泛的 AI 生物学商业化主张跑在产业结果前面跟踪全类别独立验证研究,而不只看 Astromech
数据 / 合作锁定对抗在位者Recursion、Schrödinger 和 Isomorphic Labs 已拥有多年药企整合评估 Astromech 能否现实地替换在位者,或必须转向尚未被绑定的空白买家

严重程度反映按可能性加权后的影响:Astromech 能否把技术主张转成可防守的商业位置。

[CP025, CP026, CP027, CP028, CP029]
FP003: 护城河 / 就绪度 KPI

尽管公开估值相当或更高,Astromech 在商业证明 KPI 上落后于所有披露充分的同行。

[CP013, CP014, CP025, CP029]
Chapter 04

04财务

4.1 收入流、定价与收入确认

截至 2026-07-03,Astromech 未披露任何具名收入流、定价表或收入确认政策。参考可比 AI 生物学平台的品类常规,可能的收入机制包括向药企和农业生物买家直接平台授权、围绕联合发现项目的共同开发及里程碑 / 权利金安排,以及与 Colossal 更广泛的去灭绝和生物安全叙事一致的潜在政府或生物安全合同。Schrödinger 等可比上市公司披露了软件授权加内部管线的双重收入组合;Absci 则披露了湿实验加生成式 AI 一体化服务收入。两种模式都可能类比 Astromech 未来的变现方式,但 Astromech 自身均未确认。没有披露合同结构,就无法评估收入确认处理;迄今 Astromech 未公开任何里程碑、权利金或授权费安排。[CI001, CI002, CI003, CI004, CI005]

收入流表
收入流机制单位当前价值 / 状态质量尽调问题
直接平台许可企业访问生物 AI / Large Life Model 平台$/合同(推断)未披露推测性——无公开定价索取当前价格表或许可模板
共同开发 / 里程碑-特许权使用费项目与具名合作伙伴开展联合发现项目$/交易(推断)未披露——无具名合作伙伴推测性——未披露合作伙伴索取合作管线和任何已签 term sheet
政府 / 生物安全合同面向公共部门客户的应用型生物建模$/合同(推断)未披露推测性索取任何政府或公共部门合同披露
农业 / 牲畜性状优化许可向农业生物买方许可平台输出$/合同(推断)未披露推测性索取农业生物试点清单或客户证明
组合内 / Colossal 关联方许可在 Colossal 关联实体之间共享技术未披露未披露推测性——关联方风险索取关联方许可或转让定价条款

所有行都是基于类别惯例推断的收入机制;没有一行得到 Astromech 特定披露确认。

[CI001, CI002, CI003]
定价 / 变现表
可比竞争对手价格 / 单位 / 合同模式标价 / 实现价格折扣 / 未知项来源
Astromech未披露未披露不存在公开定价、打包或交易结构信息现有来源均未披露 Astromech 定价
Schrödinger软件许可 + 内部管线经济已披露分部收入,未披露单笔交易定价双重模式让逐项比较更复杂Schrödinger 投资者关系
Absci整合湿实验 + 生成式 AI 设计费用上市公司收入按分部披露单项目经济不公开Absci 投资者关系
Recursion平台许可 + 里程碑 / 特许权使用费结构合作层面的经济以汇总口径披露单笔交易精确定价不公开Recursion 投资者关系

Astromech 行用于直接对照;可比行只使用公开投资者关系披露。

[CI004, CI005, CI006]
FI001: 收入模型桥

Astromech 可行的收入路径从平台访问开始,经由合作转化,走向里程碑 / 版税收入;但这座桥的任何阶段都没有披露且确认的美元金额。

节点代表按赛道信息推断的定性收入机制,不是已经确认的 Astromech 专属管线。

[CI001, CI002, CI003]

4.2 GTM 动作与销售效率

没有已披露来源描述 Astromech 的销售周期长度、获客成本、回本期或渠道经济。该品类可比公司依赖高接触、长周期的企业药企销售动作,而不是自助式采用;Isomorphic Labs 和 Relay Therapeutics 都围绕直接战略合作来组织 GTM,而非交易型销售漏斗;Recursion 和 Schrödinger 也报告了多年、关系驱动、带具名药企伙伴的交易周期。如果 Astromech 遵循同样的品类模式,其 GTM 可能需要专门的商务拓展职能和跨季度销售周期,而不是可规模化的低接触渠道;考虑到公司目前没有披露商业团队人数或合作伙伴,这会影响资本效率。留存的公开来源中,没有任何 Astromech 专属 CAC、回本或渠道经济代理。[CI006, CI007, CI008]

4.3 成本结构、利润率与资本强度

Astromech 未披露成本结构、毛利率、营运资本状况或资本开支计划。作为计算生物学平台,而非纯湿实验公司,其成本基础可能更偏向计算基础设施、模型训练费用和专门研究人员,而不是制造 capex;这与 Recursion 和 Schrödinger 的模式一致,两家公司都披露云 / 计算成本线和 R&D 人员是最大成本驱动项。Absci 的湿实验加 AI 一体化模式显示出显著更高的资本强度,因为实体实验会增加设施、试剂和仪器成本;如果 Astromech 所称的多组学和进化建模雄心最终需要湿实验验证循环,其资本强度可能走向 Absci 模式,而不是停留在较轻的 Recursion / Schrödinger 软件优先成本画像。没有 Astromech 专属毛利率、营运资本或 capex 数字可直接检验任一情境。[CI009, CI010, CI011, CI012]

单位经济表
指标值/null置信度为什么重要尽调问题
年度经常性收入(ARR)未披露n/a任何估值对收入合理性检查的基线索取当前 ARR 或收入年化水平
毛利率未披露n/a判断平台扩张更像软件,还是更像服务业务索取收入成本拆分
客户获取成本(替代指标)未披露n/a检验商业化落地动作是否资本高效索取销售 / BD 人员规模和周期时间数据
回本周期(代理指标)未披露n/a一旦产生收入,可用它判断商业动作的资本效率合同签署后,索取交易层面的经济性数据
人均收入(代理指标)未披露n/a研发密集型平台常用的横向效率核查同步索取当前员工人数和收入数据

Astromech 目前所有指标都为空值;保留这些行,是为了把披露缺口说清楚,而不是把比较删掉。

[CI009, CI010, CI011, CI012]
FI002: 单位经济模型桥

Astromech 假设性单位经济模型桥的每个阶段,目前都只是定性占位,而不是有来源的数字。

仅为定性节点;Astromech 没有数值输入,也不应推断为精确值。

[CI009, CI010, CI011, CI016]

4.4 公开商业进展与私有指标缺口

Astromech 的公开商业进展仅限融资和估值数字;没有来源披露收入、ARR、GMV、单位量、活跃用户、使用率或客户数。这与上市同行形成鲜明对比:Schrödinger、Recursion、Absci 和 Insilico Medicine 都披露季度收入,几家公司还通过投资者关系渠道披露客户或伙伴数量。Datanyze 式估算规模数据为 Astromech 提供了一个近似员工人数代理,但这是第三方估算,不是公司披露的商业进展,也没有留存来源用官方员工数进行交叉验证。标题估值与完全缺失任何公开商业进展指标之间的落差,是本章最关键的财务尽调缺口。[CI013, CI014, CI015]

公开财务缺口表
缺失的私有指标影响精准尽调路径
收入 / ARR无法用任何收入倍数核验约 $2B 估值是否合理直接向公司索取当前收入或 ARR 数据
具名客户数量无法评估商业牵引力或销售落地进展索取客户或试点名单,并附推荐联系人
毛利率无法评估单位经济性或平台模式扩展性索取收入成本和货品成本拆分
现金余额与烧钱速度无法脱离类别代理指标评估真实现金可支撑期索取最新现金余额和月度烧钱
员工人数(官方)无法验证第三方员工人数估计,也无法评估 R&D 能力按职能索取当前官方员工人数

每一行都对应一项截至 2026-07-03 找不到公开来源的私有指标。

[CI013, CI014, CI015]

4.5 资本充足性与融资依赖

Astromech 的融资时间线——截至 2026 年 Form D 文件,披露资本最终达到 $40.5M——已在公司概览中确立,这里不再重复;本节评估前瞻资本充足性。没有来源披露 Astromech 当前现金余额、月度现金消耗率或可支撑运营期。以已披露 $40.5M 融资作为上限,并套用可比早期计算生物学平台中观察到的品类典型现金消耗 区间(参考 Absci 和 Schrödinger 在规模化前的早期成本披露),合理的可支撑运营期估计落在最近一次披露融资后的 12 到 24 个月区间;但这只是代理,不是确认数字。Astromech 没有披露任何债务融资、项目融资义务或信用额度。下一轮触发因素尚未确认,但合理情境是:要么在宣布具名商业合作后融资,要么在估算可支撑运营期接近下限时融资。[CI016, CI017, CI018, CI019]

资本充足性表
账面现金月度烧钱可支撑月数计划资金用途下一轮融资触发因素债务 / 项目融资义务
未披露(已融资上限为 $40.5M)未披露(仅为类别代理估计)约 12-24 个月(代理指标,未确认)按 Form D 和媒体报道,用于 R&D、模型训练 / 算力和招聘未确认 — 可能与具名商业合作或现金可支撑期逼近下限有关未披露

账面现金上限沿用「公司概览」中的融资时间线,但不复制其主张 ID;这里所有前瞻性数字均为「财务」本章本地估算。

[CI016, CI017, CI018, CI019]
FI004: 资本强度 / 现金流地图

定性现金流桥从已披露 $40.5M 融资上限推向估计剩余现金余额;使用同类公司烧钱代理值,而不是已披露烧钱速度。

数值只是按同类公司推算的示意估计,仅锚定已披露 $40.5M 总融资;Astromech 没有确认现金余额。

[CI017, CI018, CI019]

4.6 关于收入质量、利润率路径与尽调阻塞项的财务结论

Astromech 当前财务画像还无法支撑独立承销判断:公司没有披露收入、利润率、单位经济或资本可支撑运营期数字;本章所有量化估计都是品类代理,而非公司专属事实。BCG 关于工业规模化的研究强化了更广泛的警示:整个合成生物学和 AI 生物学品类的商业化主张,往往跑在已证明的财务结果之前;这直接适用于 Astromech 以 $40.5M 披露资本支撑 $2B 叙事的情况。最关键的尽调阻塞项,是没有任何披露的收入或销售管线数字、没有披露现金和消耗率状况、没有具名付费客户;投资前优先级最高的请求,应是关闭这三个缺口。[CI020, CI021, CI022]

FI003: 财务估算区间

按同类赛道代理估算,Astromech 的资金续航和估值 / 资本比率区间很宽,凸显公司确认的信息很少。

上下限是按同类公司推算的代理值,锚定已披露 $40.5M 融资和约 $2B 报道估值,并非公司确认数字。

[CI020, CI021, CI022]
Chapter 05

05产品与技术

5.1 以客户工作流定义产品

Astromech 将其 Large Life Model 定位为通用生物 AI 系统,用来跨物种建模进化和多组学过程,所称应用覆盖医学、农业和生物安全。按工作流理解,平台被描述为「输入到预测」系统:用户或伙伴提交一个生物学问题或数据集,收到模型生成的关于进化轨迹、性状行为或分子结果的预测。留存公开来源没有一步步描述具体用户工作流、API、用户界面或具体交付格式;这使 Astromech 区别于 Schrödinger、Absci 等技术文档丰富的同行,后者会随平台发布详细工作流说明。缺乏有文档支撑的工作流,本身就是一个值得注意的尽调缺口,尤其在平台提出如此宽泛应用主张的情况下。[CE001, CE002, CE003]

工作流 / 用例表
用户任务当前工作流(现状)Astromech 方案(按描述)可衡量收益限制
预测生物系统的进化或性状结果人工文献综述,加上传统湿实验Large Life Model 预测(仅用概括性措辞描述)未披露 — 没有基准或案例研究没有成文工作流、API 或输出格式
大规模筛选候选分子或生物体传统高通量筛选,或内部计算筛选隐含的多组学建模支持(未展开)未披露没有具名试点或客户部署
把模型输出接入现有 R&D 流水线定制内部工具或点状解决方案未发现披露的集成入口(API / SDK)未披露任何保留来源都未找到集成文档

所有行都是推断出的工作流,因为没有保留来源从端到端记录实际 Astromech 用户工作流。

[CE001, CE002, CE003]

5.2 模块、资产与产品线地图

留存公开来源没有列出 Astromech 的离散模块、SKU 或产品线;公司公开技术叙事描述的是单一统一的 Large Life Model,而不是一组不同产品。这不同于竞争者模式:Recursion 披露了名为「Recursion OS」的产品套件,Schrödinger 披露模块化的基于物理模拟产品线,Absci 披露具名生成式设计和湿实验集成模块。没有披露模块地图,就无法判断 Astromech 平台哪些部分已经可生产使用,哪些仍是愿景;也无法判断任何具名或假设客户实际会授权哪些模块。缺少模块地图还让 Astromech 难以与同行逐项比较,因为本章每个竞争者画像至少能指向一个具名、已披露模块,而 Astromech 只给出其技术的单一伞状描述。[CE004, CE005, CE006]

产品模块 / 资产矩阵
模块 / 资产 / 产品线用户状态 / 成熟度差异化尽调缺口
Large Life Model(核心平台)内部 R&D / 未披露的外部用户早期阶段,未被独立基准测试多物种进化与多组学建模主张未披露模块边界或发布状态
进化轨迹预测(隐含能力)未披露未确认 — 仅用概括性措辞描述不同于同业切入点的预测性进化框架无公开基准或案例研究
多组学集成层(隐含能力)未披露未确认跨物种、跨组学建模雄心未披露数据集成架构
可比项:Recursion OS(竞争对手参考)制药 R&D 伙伴已披露,生产级平台大规模表型筛选n/a — 对照用参考行

Astromech 各行体现的,是公司没有披露模块地图;Recursion 行只作为类别成熟度参考点。

[CE004, CE005, CE006]

5.3 架构与运营模型

Astromech 的公开技术叙事强调多物种进化和多组学建模架构,但留存来源没有披露模型架构家族、训练数据来源、计算基础设施或基准结果。按品类类比,可比平台披露的架构细节更多:NVIDIA BioNeMo 记录了用于生物分子 AI 的基础模型加微调架构,Schrödinger 披露基于物理的模拟引擎架构,Insilico Medicine 披露生成式化学和靶点识别流水线架构(Pharma.AI)。在类似水平的技术文档、基准或独立评审公开前,Astromech 的架构应被视为未验证主张。在此类文档出现之前,运营模型最好被描述为概念层图,而不是已验证系统,因为没有留存来源把具体计算提供商、模型家族或数据集与 Astromech 所称能力连接起来。[CE007, CE008, CE009, CE010]

技术 / 运营架构表
层级 / 流程 / 组件作用依赖风险
模型架构(未披露的家族)进化 / 多组学建模的核心预测引擎算力基础设施(未披露提供商)缺少架构细节,无法评估技术风险
训练数据来源决定模型泛化能力,以及法律 / 伦理暴露未披露数据源无法评估来源与授权风险
算力基础设施支撑大规模模型训练和推理可能依赖云服务商(未确认)缺少披露,供应商集中风险未知
安全 / 合规控制保护数据和模型完整性未披露认证或框架仅凭公开渠道没有信息,无法确认没有发生过安全事件

Astromech 没有披露可比粒度的自有架构细节,因此各行按类别常见架构层拆分。

[CE007, CE008, CE009, CE010, CE017, CE018]
FE001: 产品架构图

Astromech 的架构只停留在概念层;留存来源没有披露底层模型、数据或基础设施层,细节远少于可比同行。

层级来自一般公开叙事推断,并非披露的架构图。

[CE007, CE008, CE009]

5.4 部署、集成、可靠性与路线图

没有留存公开来源披露 Astromech 的部署模型(云、本地部署或混合)、集成面(API、SDK 或数据连接器清单)、正常运行时间 / 可靠性指标,或公开产品路线图。Recursion 的公开 GitHub 组织和 Isomorphic Labs 的工程岗位发布,为可比公司的平台工程活动提供了开发者信号代理;但 Astromech 方面没有识别到等价的公开开发者表面(GitHub 活动、API 文档,或超出通用招聘语言的工程招聘信号)。鉴于缺少公开开发者表面,Astromech 自身招聘页面在这里被视为最接近的可用开发者信号代理;这符合对没有公开开发者社区公司的工作流指引。[CE011, CE012, CE013]

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
2026-04(走出隐身)Astromech 平台叙事公开发布已宣布建立公开技术叙事,但不是发布里程碑GeekWire / D Magazine 报道
未披露具名试点或客户部署未披露尚无生产部署证据无保留来源披露试点
未披露公开 API、SDK 或开发者文档发布未披露开发者信号代理指标仅限通用招聘表述Astromech 招聘页面(代理指标)
未披露独立基准或同行评审验证未披露核心差异化主张仍未经过基准测试无保留来源

Astromech 没有披露公开产品路线图;各行反映的是里程碑类别,而不是确认日期。

[CE011, CE012, CE013]
FE002: 客户工作流 / 运营流程

推断中的 Astromech 用户工作流,从生物问题或数据集输入走到模型生成预测;但没有任何阶段披露了具体界面、API 或输出格式。

流程来自一般平台叙事推断;留存来源没有记录端到端用户工作流。

[CE001, CE002, CE011]

5.5 差异化、知识产权与竞争技术位置

Astromech 所称差异化是多物种进化建模的广度;按竞争者章节,没有具名竞争者把这一点作为自身核心重点。留存公开来源没有披露专利申请、专有数据集描述或独立验证基准来支撑这一差异化主张。竞争者技术文档显示,各同行把差异化锚定在不同技术楔子上——Recursion 是大规模表型筛选,Schrödinger 和 Isomorphic Labs 是基于物理 / 结构的建模,Generate:Biomedicines / Absci / LabGenius 是生成式蛋白和抗体设计——这些都不与 Astromech 所称进化建模楔子直接重叠,因此 Astromech 的主张不是被竞争或证伪,而是尚未有基准验证。对尽调而言,这种缺乏直接重叠有两面性:没有竞争者公开证伪 Astromech 的主张,但也没有独立方验证它,差异化论点完全压在公司自身叙事上。[CE014, CE015, CE016]

信任 / 质量 / 合规表
控制 / 认证 / 质量指标状态范围缺口
安全认证(如 SOC 2、ISO 27001)未披露n/a没有任何正式认证证据
数据治理 / 隐私框架有隐私政策(通用,不针对平台)仅公司层面法律条款没有平台级数据治理细节
生物安全 / 生物安保合规声明未披露n/a鉴于平台覆盖生物建模,这一缺口值得注意
独立技术 / 安全审计未披露n/a任何保留来源都未找到第三方评审

缺少披露应视为缺口,而不是不合规证据。

[CE017, CE018, CE019]
FE004: 产品成熟度 / 能力地图

在所有已披露的成熟度维度上,Astromech 都落后于技术文档丰富的同行,显示平台仍处叙事阶段,而非经过基准验证的产品阶段。

[CE014, CE015, CE016]

5.6 信任、安全、安保与合规

除公司概览中识别到的一般隐私和服务条款文件外,Astromech 没有发布专门的信任、安全或合规页面;留存来源也没有描述安全认证(如 SOC 2)、数据治理框架,或特定于平台的生物安全 / 生物安保合规声明。考虑到平台所称生物建模范围,这一缺口很重要,因为生物安全相邻担忧是合理的。可比公司披露深度不一:Schrödinger 和 Absci 作为上市公司,通过投资者和技术材料披露了比 Astromech 迄今更正式的合规和质量控制语言。未识别到 Astromech 发生事故、泄露或合规失败的证据,但没有任何公开控制框架,也限制了能从这种缺席中得到的保证。[CE017, CE018, CE019]

5.7 附录图表

FE003: 关键依赖地图

Astromech 平台可能依赖未披露的算力基础设施、训练数据来源,以及 Colossal Biosciences 血缘;这些都没有在公开来源中得到独立确认。

依赖项来自赛道常态和 Colossal 分拆关系推断;留存来源没有披露实际供应商或基础设施清单。

[CE009, CE010, CE020]
Chapter 06

06客户

6.1 客户基础分层

Astromech 未披露客户基础,因此公司自身不存在买家 / 用户 / 付款方分层。按品类类比,可比 AI 生物学平台把买家分为大型药企 R&D 组织(Bayer 与 Recursion、Johnson & Johnson 与 Isomorphic Labs)、中型生物制药公司(Sanofi 与 LabGenius、AstraZeneca 与 BenevolentAI、Merck 与 Absci),以及长期战略合作者(Eli Lilly 与 Schrödinger、Novartis 与 Relay Therapeutics)。如果 Astromech 追求类似买家画像,R&D 或商务拓展领导层可能是买家,实验科学家或计算科学家是用户,R&D 预算是付款方;这与市场分析中已建立的买家地图一致。鉴于已披露客户完全缺席,Astromech 无法做收入带、地理或渠道分层。[CU001, CU002, CU003]

客户分层表
分层买方 / 用户 / 付款方用例规模收入 / 战略价值缺口
大型制药 R&D(类别参考)BD / R&D 领导层买方;实验台 / 计算科学家用户;R&D 预算付款方肿瘤学及更广泛发现研究协作多年、多项目(Bayer-Recursion 模式)战略价值高;已披露但未按财务拆项未披露等同的 Astromech 关系
中型市值生物制药(类别参考)CSO 办公室买方;发现科学家用户;生物技术公司运营预算付款方抗体或小分子发现协作单项目到多项目(Sanofi-LabGenius、Merck-Absci 模式)中高战略价值未披露等同的 Astromech 关系
长期战略协作者(类别参考)高管发起人买方;联合研究团队用户;战略 R&D 预算付款方跨模态、多年药物开发协作大规模、多年(J&J-Isomorphic、Lilly-Schrödinger 模式)战略价值很高未披露等同的 Astromech 关系
Astromech(实际)未披露未披露已披露账户为零未披露客户分层数据完全缺失

类别参考行描述可比竞争对手已披露的买方模式;Astromech 行则显示公司自身完全没有这类披露。

[CU001, CU002, CU003]
FU001: 客户旅程图

Astromech 没有披露客户,因此这张旅程图展示同类公司典型路径:从最初药企接触,到扩展为多项目协作;具名竞争对手案例是唯一可用证据。

节点描述从竞争对手披露推断出的同类公司典型旅程;没有任何阶段有 Astromech 专属数据点。

[CU001, CU004]

6.2 采用轨迹

留存公开来源没有披露任何 Astromech 部署、试点、账户数量或使用率指标。可比竞争者展示了多年采用轨迹:Recursion 与 Bayer 的合作、以及 NVIDIA 基础设施伙伴关系,已经跨多个披露阶段延续;Isomorphic Labs 与 Johnson & Johnson 的合作从一开始就被设计为多模态、多年期研究项目。这些模式说明,该品类采用通常从单个旗舰伙伴关系开始,并在多年内扩展,而不是通过快速自助注册扩张;如果 Astromech 披露首个客户,这就是其可能采用路径。截至 2026-07-03,Astromech 在这条轨迹上仍处于采用前阶段,披露账户数为零。[CU004, CU005, CU006]

客户增长 / 采用轨迹表
指标数值日期来源置信度含义缺失分母
Astromech 已披露账户 / 试点02026-07-03无保留来源披露任何账户高(所有保留来源均未发现相关披露)无法衡量采用轨迹可触达试点总数未知
Recursion-Bayer 合作阶段(类别参考)多阶段,持续中2025-12-09 披露Bayer 新闻室展示该类别多年采用模式Astromech 特定等同项未知
Isomorphic Labs-J&J 合作结构(类别参考)多模态项目,持续中2025-01-13 披露J&J Innovation展示先落地、再扩张的采用模式Astromech 特定等同项未知

Astromech 特定行显示采用数据为零;类别参考行说明一旦合作启动,该细分通常会呈现的采用模式。

[CU004, CU005, CU006]
FU002: 采用 / 部署漏斗

采用漏斗每个阶段,Astromech 披露账户数都是零;与此相对,可比竞争对手披露了多阶段、多年的采用模式。

数值反映 Astromech 各阶段已披露账户数(全为零);同类比较对象只在注释中定性引用,未量化。

[CU005, CU006]

6.3 具名客户证明

Astromech 在任何留存公开来源中都没有披露具名客户、试点或参考账户。因此,本章列举来自可比竞争者的具名客户验证样本——Bayer、Johnson & Johnson、Sanofi、Eli Lilly、Novartis、AstraZeneca 和 Merck——作为品类参考点,明确不是 Astromech 自身客户。这些具名关系均由竞争者自身披露或独立新闻报道佐证;多数描述的是生产或持续研究合作,而不是早期试点。下方列举表中的 Astromech 行用于直接对照,显示其完全缺少任何等价验证点。这个样本有代表性但并不穷尽:每个竞争者很可能还拥有该单一旗舰样本之外未披露或较小的合作关系,因此品类真实伙伴关系总数几乎肯定高于本章列出的七个关系。[CU007, CU008, CU009, CU010]

具名客户证明表
客户分层部署 / 用例生产 / 试点结果限制
Bayer(Recursion 客户)大型制药聚焦肿瘤学的研究协作生产级 / 持续多年合作在已披露阶段之间持续推进不是 Astromech 客户 — 仅作类别参考
Johnson & Johnson(Isomorphic Labs 客户)大型制药跨模态药物开发协作生产级 / 持续宣布战略合作,范围较广不是 Astromech 客户 — 仅作类别参考
Sanofi(LabGenius 客户)生物制药抗体发现协作生产级 / 持续在平台更新同时披露具名制药伙伴关系不是 Astromech 客户 — 仅作类别参考
Eli Lilly(Schrödinger 客户)大型制药长期计算发现协作生产级 / 持续多年具名合作不是 Astromech 客户 — 仅作类别参考
Novartis(Relay Therapeutics 客户)大型制药基于结构 / 动力学的药物发现协作生产级 / 持续公司网站披露具名伙伴不是 Astromech 客户 — 仅作类别参考
AstraZeneca(BenevolentAI 客户)大型制药知识图谱驱动的发现协作生产级 / 持续重定位后披露具名伙伴关系不是 Astromech 客户 — 仅作类别参考
Merck(Absci 客户)大型制药生成式抗体设计协作生产级 / 持续公司网站披露具名伙伴关系不是 Astromech 客户 — 仅作类别参考
Astromechn/an/an/a — 未披露客户任何保留来源都不存在具名客户、试点或结果具名客户证明为零;这是本章最大单一缺口

第 1-7 行是可比竞争对手的具名客户证明点,保留在此是为了说明类别级客户证明长什么样;第 8 行(Astromech)显示公司自身完全没有这类披露。 覆盖范围是本报告覆盖的竞争对手中,具名且已披露的制药伙伴关系代表性样本,并非每个竞争对手全部伙伴关系的详尽普查。

[CU007, CU008, CU009, CU010]
FU003: 客户证明矩阵

Astromech 在所有客户证明维度上都处于最低位;具名竞争对手合作拿出了生产阶段、独立佐证的证据,每个维度都达标。

[CU007, CU008, CU009, CU010, CU011]

6.4 留存、耐久性与满意度

没有留存来源披露 Astromech 的续约率、净收入留存或毛收入留存、流失率、合同期限或客户满意度信号,因为公司没有披露任何可衡量的客户关系。品类类比再次成为唯一可用视角:Recursion-Bayer、Schrödinger-Lilly 等多年续约药企合作表明,该品类一旦建立伙伴关系,往往具有耐久性和多阶段特征,而不是单笔交易;这可能反映上线后高切换成本和深度数据集成。该模式无法针对 Astromech 具体验证;对于一家没有披露可留存客户关系的公司,也不存在满意度或留存代理。[CU011, CU012, CU013]

留存 / 重复使用 / 满意度表
指标数值 / 空值分层置信度尽调要求
Astromech 净 / 毛收入留存未披露 / 不适用n/an/a — 不存在客户基础先索取任何客户或试点关系,作为留存分析前提
Astromech 流失率未披露 / 不适用n/an/a披露首个客户前无法评估
类别多年续约模式(Recursion-Bayer、Schrödinger-Lilly)定性上持久、多阶段(示例)大型制药中 — 基于持续披露的合作,而不是已发布留存百分比确认 Astromech 未来商业化路径是否会遵循同样的持久性模式

Astromech 没有数字留存数据;多数具名竞争对手也没有。类别行只是定性展示持久性,不是已披露留存百分比。

[CU011, CU012, CU013]
FU004: 留存 / 重复合作队列

该队列视图仅是示意性的类别模型——纳入来源中没有任何 Astromech 专属留存百分比,也没有竞争对手精确披露的留存百分比。

百分比是示意性的类别模型,依据可比制药合作披露的多年期续约定性模式搭建;它们既非 Astromech 专属,也不是竞争对手精确披露的数据。

[CU011, CU012]

6.5 扩张与集中风险

Astromech 承担该品类中最极端的集中风险画像:其客户基础目前 100% 未披露或不存在,意味着没有收入集中度可衡量,也没有任何扩张动作证明。可比竞争者显示出落地后扩张模式,即初始药企合作(例如与 Bayer 或 Sanofi 的单一项目)随着时间扩展为更广泛的多年或多项目关系;如果 Astromech 最终拿到首个具名伙伴,在披露第二、第三个具名关系前,过度依赖该单一账户将成为实质集中风险。因此,Astromech 当前与尽调相关的扩张风险不是客户集中,而是客户缺席——这是本章通常评估风险的反面。Novartis 和 AstraZeneca 都与不止一家 AI 生物学平台供应商保持具名关系,强化了这一点:成熟买家在该品类中很少单一采购,也进一步凸显 Astromech 相对于其余已画像竞争集合仍处在多早期的 GTM 位置。[CU014, CU015]

扩张与集中风险表
扩张驱动因素集中风险影响尽调路径
初始制药合作内先落地、再扩张(类别模式)不适用于 Astromech——目前还没有初始合作关系高——Astromech 尚未拿下合作关系,也就无从扩展索取当前合作伙伴管线和任何具名潜在客户名单
合作伙伴上线后的多项目扩展(品类模式,例如 Bayer-Recursion)如果 Astromech 的首个合作伙伴变成唯一合作伙伴,未来会出现单一客户集中风险高——公司早期披露的第一个合作伙伴往往很快成为主要收入来源跟踪具名合作伙伴数量变化;如果任何单一账户占活动比重过大,就标记风险
借 Colossal Biosciences 网络形成的渠道 / 合作伙伴依赖介绍来源依赖 Colossal 网络,声誉和获客都受其牵引中——该渠道可能带来线索,但不是多元化商业化动作索取任何非 Colossal 来源合作伙伴或线索的证据

Astromech 尚未披露客户,集中度风险目前仍是理论问题;更紧迫的发现是,公司连可供集中分析的客户都没有。

[CU014, CU015]
Chapter 07

07风险

7.1 按严重程度排序的风险概览

按严重程度排序,Astromech 的最高风险是:(1)近乎彻底的公开披露风险——公司没有发布财务数据、具名客户、技术基准或合规认证,几乎所有主张都无法独立核验;(2)品类层面的双重用途和生物安全治理风险,适用于任何能够建模进化或基因组动态的 AI 生物学平台;(3)如果 Astromech 所称湿实验验证雄心需要越来越昂贵的基础设施,则存在资本强度风险;(4)与创始人 Ben Lamm 及 Colossal Biosciences 网络绑定的关键人物和单一生态依赖风险;(5)在隐身状态下运营且没有披露 GTM 动作带来的执行风险。上述风险目前都没有 Astromech 专属的披露缓释措施;缓释成熟度只能从品类模式定性评估,在公司披露更多信息前,几乎所有维度的剩余暴露都高。[CR001, CR002, CR003]

FR001: 风险热力图

披露风险以及行业层面的生物安全 / 监管风险,在发生可能性和严重性上得分最高;鉴于公司仍处早期,财务和人员风险严重性高,但近期发生可能性较低。

[CR001, CR002, CR003]
FR002: 风险传导图

展示 Astromech 的核心风险如何从披露和监管敞口,传导到融资、客户,最终影响估值结果。

边表示根据行业模式和 Astromech 已披露结构推断出的合理因果传导路径,并非公司确认的风险模型。

[CR026, CR027]

7.2 监管与法律风险

留存来源没有披露任何 Astromech 专属监管批准、许可、诉讼、执法行动或 IP 申请。更广泛的品类正受到监管关注:FDA 和 EMA 已联合发布医药开发中 AI 使用的共同原则,FDA 已发布监管决策中 AI 使用指南(通过 Federal Register 通知正式化),NIST 维护一个活跃的合成核酸生物安全项目,直接关系到任何建模基因组或进化序列的平台。Astromech 已披露的法律材料只有标准隐私政策、服务条款页面和基础 Texas 公司登记记录;这些都没有处理 AI 特定、生物安全特定或双重用途特定合规。考虑到公司所称生物建模范围,这一缺口需要提高关注。[CR004, CR005, CR006, CR007, CR008, CR009]

监管 / 法律风险登记表
规则 / 许可 / 案件司法辖区状态可能性严重性缓释措施剩余暴露尽调路径
药物研发 AI 监管指南(FDA)美国指南已发布,但尚未成为有约束力的法规Astromech 未披露合规路线图披露前为高索取 Astromech 明确的监管合规路线图
FDA/EMA 联合 AI 医疗原则美国 / 欧盟联合原则已发布,不具约束力Astromech 未披露监测任一司法辖区是否推进有约束力的规则制定
NIST 合成核酸序列生物安全筛查美国联邦项目已启动,按类别适用Astromech 未披露筛查立场披露前为高索取 Astromech 的序列筛查或生物安全合规立场
Astromech 公司注册(德克萨斯州)德克萨斯州,美国公开登记显示有效标准注册代理备案已到位没有具体疑虑时无需进一步行动
Astromech 隐私 / 条款披露n/a(网页政策)已发布,标准格式标准隐私政策和条款已到位低-中——没有 AI 专项或生物安全专项隐私附录如有 AI / 生物安全专项数据治理附录,要求提供

本表只收录留存来源中与严重性最相关的监管和法律事项,并非穷尽每一条适用规则或许可;各行按严重性排序。

[CR004, CR005, CR006, CR007, CR008]

7.3 运营、质量与安全风险

Astromech 未披露制造、供应链或设施信息,这与当前阶段的计算平台(而非湿实验平台)一致;不过,公司自身公开材料描述了走向湿实验验证的雄心,届时会引入目前尚未出现的设施、试剂供应和生物安全运营风险。Astromech 平台没有安全认证、事故历史或正常运行时间 / 可靠性披露。关于双重用途 AI 生物学系统的同行评议治理文献,正是把这类运营缺口——缺少上游风险收益审查流程——标记为品类层面的实质担忧;这不专属于 Astromech,但因其品类身份适用于它。Center for Health Security 的独立治理研究也强化了这一发现,指出正式运营安全框架在整个 AI 生物学品类中仍不成熟,并非只有 Astromech 如此。[CR010, CR011, CR012, CR013, CR014, CR015]

运营 / 质量 / 安全风险登记表
失效模式可能性严重性缓释成熟度剩余暴露未解决缺口
未披露安全认证或事件响应立场未披露索取当前安全立场、认证和事件历史
湿实验室验证目标会带来未来生物安全 / 设施风险,但目前尚未出现中-高不适用——尚未披露湿实验室运营中——取决于未来范围扩张询问是否以及何时计划开展湿实验室运营
双用途 AI 生物学系统缺少品类层面的上游风险收益评估品类层面的治理文献已存在;未披露 Astromech 专属流程询问 Astromech 是否采用了任何上游风险评估框架
未披露平台正常运行时间、可靠性或宕机历史低-中未披露——没有披露处于生产使用中的产品要等到出现客户或试点、生成可靠性数据后才能评估

各行按严重性从高到低排序;几乎每一种失效模式都反映披露缺口,而非已确认事件。

[CR010, CR011, CR012, CR013]

7.4 伙伴与依赖风险

Astromech 是 Colossal Biosciences 直接剥离公司,与母公司共享网络、声誉,以及可能的推荐 / 获客依赖,类似 Colossal 早前剥离的 Form Bio。留存来源没有披露 Astromech 的云、计算或基础设施提供商,也没有披露 Colossal 生态本身之外的任何具名供应商、分销商或渠道伙伴。这个单一生态依赖构成集中风险:如果 Colossal 的声誉、融资或网络访问受损,鉴于共享品牌和创始人,Astromech 的线索来源和可信度可能受到影响。除种子轮和延伸融资外,没有披露资本提供方;由于没有识别到监管申报,也没有披露监管特定依赖。[CR016, CR017, CR018, CR019, CR020, CR021]

合作伙伴 / 依赖风险登记表
依赖项交易对方角色集中度失效情景严重性缓释措施剩余暴露
项目来源与声誉生态Colossal Biosciences母公司 / 分拆赞助方高——唯一披露的生态关联Colossal 出现声誉或融资压力,会外溢到 Astromech共同创始人和品牌今天能提供可信度,但未披露独立缓冲独立合作伙伴或客户网络披露前为高
过往分拆先例Form Bio(2022 年 Colossal 分拆公司)可比先例,不是当前依赖中——仅作示例不直接适用于 Astromechn/a——仅作历史参考
算力 / 基础设施提供商未披露Unknown未知——未点名提供商未知宕机或成本上升风险未披露提供商披露前为中
资本提供方集中度未披露的种子轮 / 延展轮投资者融资交易对方高——只披露两轮融资,投资者未完全点名如果后续融资延迟或失败,会出现融资缺口未披露融资计划和下一轮时间表披露前为高

各行按严重性从高到低排序。

[CR016, CR017, CR018, CR019]
FR003: 依赖关系图

Astromech 最关键的依赖,是其与 Colossal Biosciences 共享的生态;算力 / 基础设施和资本提供方依赖披露得很浅。

节点和边只反映纳入公开来源中可识别的依赖;未披露依赖(例如算力提供方)标为未解决,而非省略。

[CR016, CR017, CR020]

7.5 人才与执行风险

Astromech 与单一创始人 Ben Lamm 高度绑定;Lamm 同时领导 Colossal Biosciences,并创立过多家公司。除这一创始人关联外,公司没有披露高管团队、技术领导名单或继任计划。没有披露招聘速度、流失率或组织结构。公开身份集中在一个人身上,是实质关键人物风险:截至目前,Astromech 的可信度、投资人关系和媒体报道,很大程度锚定在创始人的个人声誉和履历,而不是独立披露的领导梯队。公司也没有披露商业或商务拓展职能,在当前隐身阶段,执行风险进一步集中在一支规模小且基本未披露的团队中。[CR022, CR023, CR024, CR025]

人员 / 执行风险登记表
角色 / 职能依赖或缺口可能性严重性缓释措施尽调路径
创始人 / CEO(Ben Lamm)公开身份和可信度集中在一人身上,而他同时领导 Colossal Biosciences低-中未披露——未发现继任计划索取组织架构图、高管团队名单和继任计划
技术 / 科学领导层未披露具名 CTO、首席科学家或技术领导团队中-高未披露索取具名技术领导层及其过往记录
招聘与组织规模未披露员工人数、招聘节奏或流失数据未披露索取当前员工人数和招聘计划
商业化执行团队未披露销售、BD 或合作伙伴职能中-高中-高未披露询问在披露首个客户前是否已有商业团队

各行按严重性从高到低排序。

[CR022, CR023, CR024, CR025]

7.6 财务与模型风险

如财务章节所述,Astromech 披露了 $30 million 初始 Form D 发行,随后约 $10.5 million 延伸融资,披露资本合计 $40.5 million;对应所称 $2 billion 估值,但没有披露收入、现金消耗率或可支撑运营期。由此形成实质财务模型风险:在没有披露收入的情况下给出高估值,意味着公司很可能在获得任何商业验证前还需要大量后续资本;下一轮融资的任何延迟都可能带来资金缺口风险。如果 Astromech 的湿实验验证雄心落地,资本强度很可能进一步上升,放大这一风险。未识别到欺诈、信用或营运资本信号,但这反映的是缺少任何已披露财务报表,而不是一张健康证明。[CR026, CR027, CR028, CR029]

7.7 缓释、监测与打破论点的触发因素

由于 Astromech 几乎没有披露自身风险状态,投资人今天可用的实际缓释手段就是尽调本身:在任何进一步资本投入前,要求审计或审阅财务报表、具名客户参考、合规 / 认证路线图,以及技术架构说明。可监测的打破论点触发因素包括:在合理的退出隐身窗口内未披露首个具名客户;出现影响该品类的不利生物安全或监管行动;创始人离职且没有披露继任计划;或出现下行轮 / 后续融资失败。尽管截至 2026-07-03 这些因素都无法评估,但往后每一个都能通过公开来源衡量。投资人应在下一次披露融资事件或首个具名客户公告时完整重看本风险章节,以先发生者为准。[CR030, CR031, CR032, CR033]

缓释措施与否决标准表
风险可监测触发项阈值 / 事件行动含义
披露风险(无客户、财务数据或认证)披露首个具名客户、经审计财务报表或认证退出隐身状态后 12 个月内披露上述任一项正向——降低本章最大的单一风险类别
资本强度 / 融资缺口风险后续融资轮未能在合理跑道窗口内完成后续融资负向——表明融资承压;提高尽调紧迫性
关键人物风险创始人离任或公开领导层变动Ben Lamm 离任且未公布具名继任者负向——触发论点破裂,需要立即重做投资论证
品类监管 / 生物安全风险有约束力的药物研发 AI 规则或生物安全执法行动任何影响该品类的有约束力规则制定或执法行动若为不利事项则为负向——需要重新评估整个品类论点,而不只是 Astromech

尽管截至 2026-07-03 尚未观察到这些触发项,但未来可通过公开来源持续监测。

[CR030, CR031, CR032, CR033]
Chapter 08

08估值

8.1 投资正反论点

多头论点押在品类动能上:Recursion、Isomorphic Labs、Schrödinger、Insilico Medicine 等 AI 生物学平台已经拿到大型药企合作,有几家还完成上市;Astromech 由 Colossal Biosciences 创始人主导拆分出来,在这一品类里有可信源起和品牌光环。反面论点在于,要支撑这套论点,每个公司级证据——收入、具名客户、技术基准、合规状态、独立领导层梯队——截至 2026-07-03 都未披露;同时,已披露融资只有 $40.5M,并不能明显支撑 $2 billion 估值。什么会改变多头观点:披露具名药企合作、技术基准,或独立财务报表。什么会改变空头观点:隐身期结束后的合理窗口过去后,上述三项仍沉默,或出现降估值融资。[CV001, CV002, CV003, CV004]

论点 / 反论点表
论点什么会改变判断
正方:品类势能(Recursion、Isomorphic Labs、Schrödinger)加上 Colossal 关联创始人的可信度,支撑 Astromech 的定位披露具名制药合作伙伴或独立技术基准
反方:收入、客户、技术基准或合规立场均未披露,使 $2B 估值无法验证退出隐身状态后的合理窗口期过后仍继续不披露,或出现下调估值融资
正方:创始人履历(Colossal Biosciences、过往创业)显示其具备执行能力披露独立技术领导团队,降低关键人物集中度
反方:估值 / 资本比例($2B vs $40.5M 已融资)相对公开可比公司异常偏高一轮定价后续融资或公开上市,独立佐证 $2B 数字

每行将一个论点与会改变评估的具体披露配对。

[CV001, CV002, CV003, CV004]

8.2 建议、置信度与估值立场

本章建议为观察 / 不领投;此前各章记录的披露缺口很多,因此置信度为低到中。风险评级:高。估值立场:缺乏公开证据支持——媒体所称 $2 billion 是私募轮叙事,不是任何独立备案、分析师报告或已披露收入倍数佐证的数字。在 Astromech 至少披露首个具名客户和基础财务指标前,无法负责任地设定目标回报、持有期或退出时间表;今天投入的资本,应按品类期权下注来定规模,而不是当成已经被证据验证的成长投资。只要出现任何重大新披露,就应立即重审这一立场,因为当前建议反映的是信息缺口,并非已确认 Astromech 底层技术或团队存在负面信号。[CV005, CV006, CV007, CV008]

建议摘要表
建议置信度风险评级估值立场决策含义
观望 / 不领投低-中缺少独立公开证据支撑至少披露一个具名客户或一份独立财务报表前,不投入新增资本

单行摘要,反映截至 2026-07-03 本章的整体建议。

[CV005, CV006, CV007, CV008]
FV001: 推荐逻辑

推荐逻辑先看规模和证明尚未跑通,再看普遍披露缺口把风险评级推高,最终得出在独立验证前应“观察 / 不领投”的立场。

[CV005, CV006]
FV004: 投资 KPI

Astromech 在证明和证据质量上得分最低,在类别 / 市场动能上得分最高,与本章“观察 / 不领投”的建议一致。

[CV007, CV008]

8.3 融资与估值背景

Astromech 已披露融资包括一笔 $30 million 的初始 Form D 发行(2025-08-12)和约 $10.5 million 的延伸轮(2026-03-26),已披露资本合计 $40.5 million,对应报道中的 ~$2 billion 估值——相较 Absci、Schrödinger 等可比上市 AI 生物学公司,这一估值 / 资本比异常高;后者的市值有已披露收入和临床 / 商业里程碑托底。公司未披露稀释表、清算优先权层级或股权结构细节,因此无法评估优先权悬顶。今天的入场纪律应假设完整 $2 billion 数字尚未验证,直到独立数据点出现:定价后续轮、已披露收入,或公开上市。[CV009, CV010, CV011, CV012]

8.4 多头、基准与空头情景

多头情景下,Astromech 在 12 个月内披露具名药企合作和技术基准,支撑按当前 $2 billion 或更高估值进行后续融资,符合 Recursion、Isomorphic Labs 曾出现的先落地再扩张模式。基准情景下,Astromech 未来 12-18 个月仍基本不披露,先按持平到下调估值做一笔小型过桥轮或延伸轮,同时推进首个客户。空头情景下,Astromech 拿不到具名客户,也无法及时完成后续融资;估值被下调到更贴近已披露资本基础的水平,呼应更广泛合成生物学品类面临的估值压缩风险。考虑到当前未解披露缺口很多,概率信号更偏向基准情景。[CV013, CV014, CV015, CV016]

乐观 / 基准 / 悲观情景表
情景假设估值 / 回报逻辑关键风险概率信号
乐观12 个月内披露具名制药合作伙伴和技术基准后续融资定价达到或高于当前 $2B 标记如果披露的基准落后于品类同业,会产生执行风险低-中——目前没有即将披露的证据
基准12-18 个月内继续有限披露;小规模桥接 / 延展融资下一轮估值持平到下调,等待证明点披露缺口拖得越久,投资者信心越被侵蚀中-高——与前面各章呈现的当前轨迹一致
悲观未拿下具名客户,也未及时完成后续融资估值下调到与已披露资本基础相匹配的水平融资缺口风险,以及潜在困境出售或关停情景低-中——需要具体负面催化剂,目前未披露

概率信号是基于本次尽调各章记录的披露轨迹作出的定性判断,不是建模得出的概率分布。

[CV013, CV014, CV015, CV016]
FV002: 估值敏感性

Astromech 的隐含估值,对未来 12-18 个月内是否披露首个具名客户和独立财务报表最敏感。

百分比是示意性敏感性信号,反映各披露事件的相对重要性,并非模型化估值输出。

[CV014, CV015]

8.5 可比公司组

最相关可比组覆盖不同阶段的上市和私营 AI 生物学公司:Absci 与 Schrödinger(上市,披露收入和市值)、Recursion(上市,披露现金头寸和药企合作)、Relay Therapeutics(上市,SEC 备案,具名药企伙伴)、BenevolentAI(Euronext 上市,披露投资者材料)、Insilico Medicine(通过 HKEX 文件披露上市信息,披露 IPO 估值),以及 Generate:Biomedicines(私营,仅有股票分析式估算)。Astromech 自身 $2 billion 数字是私募轮叙事,没有这些参考点里同等独立佐证;这正是今天用可比法给公司估值的核心限制。各可比公司的商业模式范围也与 Astromech 单一平台叙事有实质差异,因此这组公司推导出的任何倍数都只能作为方向信号,不能当成精确估值锚。[CV017, CV018, CV019, CV020]

可比估值表
可比对象指标倍数 / 估值 / 状态相关性局限
Astromech(实际)私募轮估值报道约 $2B,已披露融资 $40.5M本次尽调的直接对象完全来自叙事性来源;没有独立佐证
Absci Corporation公开市值见于股票分析统计直接可比的公开 AI 生物学同业市值反映公开交易动态,不是私募轮可比口径
Schrödinger公开市值见于股票分析统计直接可比的公开 AI 生物学同业,兼有软件 / 管线模式业务组合比 Astromech 的单一平台叙事更宽
Recursion Pharmaceuticals公开市值 / 现金头寸见于投资者关系材料具名可比公司,已披露制药合作伙伴(Bayer)公司规模更大、阶段比 Astromech 更成熟
Relay Therapeutics公开市值见于 SEC EDGAR 文件和股票统计具名可比公司,已披露制药合作伙伴(Novartis)临床阶段生物技术模式不同于 Astromech 的纯平台叙事
BenevolentAIEuronext 上市估值见于投资者材料AI 生物学平台的可比公开上市结果重定位后的公司画像与 Astromech 所处阶段差异明显
Insilico MedicineIPO 估值见于 HKEX 文件近期 AI 生物学 IPO 可比对象地域不同(香港上市),业务组合也不同
Generate:Biomedicines私营公司估值估算按类似股票分析的统计估算融资阶段相近的私营 AI 生物学平台可比对象估值是第三方近似估计,不是已披露轮次价格

本表覆盖留存来源中最相关的公开和私营 AI 生物学可比对象代表样本,并非穷尽所有可能的可比公司。

[CV017, CV018, CV019, CV020]

8.6 退出准备度与最终尽调要求

按任何标准衡量,Astromech 都还不具备退出条件:没有收入、没有具名客户、没有独立财务报表,也没有已披露技术基准,无法支撑未来 IPO 或战略收购论证。最关键的破坏论点触发项是:后续融资失败或估值显著下调;隐身期结束后的合理窗口过去后,仍没有任何具名客户;或出现对整个品类不利的监管 / 生物安全行动。最能提升投资判断信心的最终尽调要求是:经审计或审阅的财务报表、具名客户或试点参考、独立技术基准或第三方验证研究,以及已披露合规 / 认证路线图。[CV021, CV022, CV023, CV024]

论点破裂与否决触发项表
触发项阈值对论点的传导行动含义
后续融资失败或估值下调任何定价融资轮的估值等值不高于已披露的 $40.5M 资本基础直接削弱 $2B 估值叙事按新估值重做整套投资论证
持续没有具名客户退出隐身状态后 12-18 个月内仍未披露具名客户证实客户与风险章节中标记的客户获取风险将建议从观望下调为放弃
品类层面出现不利监管或生物安全行动任何影响 AI 生物学序列建模的有约束力执法行动推高整个品类的监管风险,不只是 Astromech广泛重估品类论点,而不是只重估 Astromech
创始人离任且无接班计划任何披露显示 Ben Lamm 离任且未指定继任者直接抬高「风险」中标出的关键人风险立即重新承销判断;按打破投资假设事件处理

这些触发项未来都可通过公开来源监测;截至 2026-07-03,尚未观察到任何一项。

[CV022, CV023]
最终尽调追问表
主题缺失证据重要性负责人 / 尽调路径
财务报表覆盖收入、烧钱速度和资金续航期的审计或审阅财务报表没有这项证据,无法验证 $2B 估值,也无法评估融资缺口风险把它列为融资轮前置条件,直接向公司管理层索取
客户证明具名客户、试点或意向书缺少它,本次尽调的客户假设只能靠同类公司类比支撑要求客户背调电话或披露具名试点
技术验证独立技术基准测试或第三方验证研究没有它,无法从技术上评估核心 “Large Life Model” 主张要求技术简报,或委托独立基准测试
合规与治理公开的合规 / 认证路线图和风险监督流程整个赛道受到监管关注,这项风险在上升,不是静态风险要求合规路线图,以及任何董事会层面的风险监督文件

按底层披露缺口的严重程度排序。

[CV021, CV024]

8.7 附录图表

FV003: 估值 / 回报区间

鉴于几乎完全缺少独立验证,本章将 Astromech 的估值框定为宽幅、低置信度区间,而非精确估算。

该区间基于类别可比倍数和已披露的 $40.5M 资本基础,而非任何 Astromech 专属财务指标。

[CV009, CV010, CV011]

免责声明

基于截至 2026-07-03 的公开来源编制;私营公司披露不完整,本报告不构成投资、法律或会计建议。

证据索引

结论
编号陈述可信度来源
CO001 Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life. SO001, SO003
CO002 The company says it is building a Large Life Model rather than a conventional narrow workflow tool. SO003, SO001
CO003 Astromech's official technical framing includes multi-species inputs, transcriptomic data, ancestral reconstruction, graph-based pangenomics, and a future multi-omics integrator. SO003, SO021
CO004 The company is publicly associated with Austin, Texas through its SEC filing address and multiple media summaries. SO010, SO011
CO005 Astromech positions its platform as applicable to therapeutic development, trait optimization, and programmable biology. SO003, SO014
CO006 Third-party coverage consistently describes Astromech as a predictive biology or biological AI company rather than as a single-asset biotech. SO012, SO014
CO007 Astromech is best framed publicly as a platform company at the intersection of AI, computational biology, and synthetic biology. SO003, SO012
CO008 Public materials emphasize option value across medicine, agriculture, livestock, and wildlife rather than a single first paying use case. SO014, SO020
CO009 Astromech's homepage includes a wide hiring slate across AI, genomics, data, and bioinformatics roles. SO005, SO001
CO010 The public company story currently contains substantially more technical ambition than commercial proof. SO001, SO014
CO011 Ben Lamm is publicly identified as a co-founder and CEO of Astromech. SO010, SO014
CO012 George Church is publicly identified as a co-founder and advisor of Astromech. SO010, SO009
CO013 Colossal describes Astromech as its newest spinout and a product of the same biological infrastructure built for de-extinction work. SO009, SO012
CO014 Public sources tie Astromech's founder credibility directly to Colossal's broader scientific and commercial narrative. SO009, SO014
CO015 A full independent board or governance structure is not publicly disclosed in retained sources. SO014, SO024
CO016 The first public financing record shows only one investor already in the 2025 Form D filing. SO010, SO018
CO017 Public sources do not disclose an ownership split between founders, Colossal, and outside investors. SO010, SO016
CO018 The spinout context implies strategic dependence on Colossal even though explicit IP-transfer terms are not public. SO009, SO022
CO019 Astromech filed a Form D on 2025-08-12 for a $30 million equity offering. SO010, SO018
CO020 The 2025 Form D reported $29,999,998 sold and $2 remaining in the offering. SO010, SO019
CO021 Astromech filed a second Form D on 2026-03-26 for roughly $10.5 million of additional equity. SO011, SO016
CO022 The two retained Form D filings imply $40.5 million of publicly disclosed capital raised to date. SO010, SO011
CO023 Axios reported that the March 2026 financing was a seed extension rather than a fully new large round. SO016, SO011
CO024 Multiple 2026 media sources reported that Astromech was valued at about $2 billion less than a year after spinout. SO014, SO016
CO025 The public valuation signal is unusually high relative to the disclosed $40.5 million capital base. SO010, SO016
CO026 D Magazine described Astromech as pre-revenue in March 2026. SO014, SO016
CO027 No retained public source disclosed current ARR or a revenue run-rate as of 2026-07-03. SO014, SO020
CO028 No retained public source disclosed a named roster of paying customers as of 2026-07-03. SO014, SO012
CO029 No retained public source disclosed a current employee headcount as of 2026-07-03. SO005, SO014
CO030 The Austin office address appears in the Form D filing and is echoed by D Magazine. SO010, SO014
CO031 Media coverage in August 2025 already framed Astromech as an AI-biotechnology platform company. SO018, SO019
CO032 The March 2026 Form D is the cleanest dated public evidence of the seed-extension milestone. SO011, SO016
CO033 Colossal's April 2026 write-up says Astromech reached unicorn status within nine months of operation. SO009, SO012
CO034 GamesBeat described Astromech as a Colossal Biosciences spinout at a $2 billion valuation. SO012, SO009
CO035 Public legal sources confirm corporate existence and website policies, but they do not provide deeper governance rights or risk disclosures. SO006, SO008
CO036 No retained public source surfaced lawsuits or sanctions involving Astromech through runDate. SO024, SO008
CO037 The most important overview-level diligence issue is the mismatch between a strategic $2 billion narrative and missing operating denominators. SO014, SO016
CO038 Astromech's public launch narrative is inseparable from Colossal's broader commercialization strategy for synthetic-biology infrastructure. SO009, SO022
CM001 Astromech's addressable spend spans AI-enabled drug discovery tooling, synthetic-biology infrastructure, and potentially biosecurity-adjacent analytics rather than one clean market category. SM021, SM001
CM002 Excluded spend includes wet-lab CRO execution, manufacturing, and clinical-trial operations, which Astromech does not appear to offer directly. SM019, SM021
CM003 The status-quo substitute for most buyers is in-house bioinformatics teams plus point-solution AI vendors such as Recursion, Insilico, Schrödinger, and Isomorphic Labs. SM017, SM001
CM004 Comparable companies publicly categorize themselves inside the AI-in-drug-discovery or computational-biology platform category, the closest public analog to Astromech's positioning. SM001, SM003
CM005 Synthetic-biology and multi-omics infrastructure spend are adjacent markets that plausibly expand Astromech's addressable spend beyond pure drug discovery. SM009, SM013
CM006 No retained public source shows Astromech itself naming a specific target market or addressable-spend figure. SM019, SM025
CM007 The Business Research Company estimates the AI-in-drug-discovery market at roughly $2-3 billion in its base year. SM003
CM008 Global Market Insights and Research and Markets both size the AI-in-drug-discovery category in a similar low-single-digit-billion base-year range. SM005, SM007
CM009 MarketsandMarkets frames AI-in-biotechnology as a substantially broader multi-billion-dollar category that includes diagnostics and manufacturing use cases beyond Astromech's apparent scope. SM001
CM010 Analyst CAGR estimates for AI-in-drug-discovery tooling range from roughly 28% to 46% depending on methodology. SM005, SM007
CM011 Custom Market Insights sizes AI-in-synthetic-biology at a much smaller hundreds-of-millions base figure than AI-in-drug-discovery, projecting over 30% CAGR to the mid-2030s. SM009
CM012 Retained analyst reports disagree by more than 2x on both base-year market size and forward CAGR depending on category scope. SM007, SM001
CM013 A defensible evidence-constrained SAM for Astromech is a low-single-digit-billion near-term wedge of pharma/agbio R&D tooling spend, well below the full multi-tens-of-billions headline TAM. SM011, SM001
CM014 No public source discloses an Astromech-specific SOM; the company has not disclosed a pipeline, pilots, or revenue that would support one. SM025, SM023
CM015 OECD research frames synthetic biology's industrial-scale translation as real but materially slower than headline market-size projections imply. SM013, SM015
CM016 BCG similarly cautions that synthetic biology is only getting closer to industrial scale, not yet broadly there, which tempers optimistic TAM narratives. SM015, SM013
CM017 Large pharma R&D and business-development leadership are the most probable buyers and budget owners for AI-biology platform licensing. SM017, SM001
CM018 Comparable-company deal patterns show a repeatable path from platform demonstration to paid pilot to milestone or platform-license economics. SM017, SM023
CM019 Agbio and industrial-biotech buyers follow a similar but distinct path centered on trait-design pilots and field validation rather than clinical milestones. SM009, SM013
CM020 Governance researchers raise the possibility of a government or biosecurity-adjacent buyer segment, but no retained source confirms an active public-sector deal in this category. SM013, SM015
CM021 Public case evidence on pilot-to-milestone conversion rates for comparable AI-biology companies is thin and mostly qualitative rather than quantified. SM017, SM001
CM022 Falling AI compute cost and growing multi-omics datasets are frequently cited growth drivers for AI-enabled biology tooling. SM001, SM011
CM023 Pharma cost pressure is cited as a driver favoring AI-augmented discovery over brute-force screening approaches. SM003, SM011
CM024 FDA and EMA guidance on AI in drug-development decision-making remains under active development rather than settled, which slows adoption in the highest-value therapeutic use cases. SM013, SM015
CM025 Enterprise pharma deals in this category carry heavy data-rights negotiation and switching-cost friction that lengthens sales cycles. SM017, SM001
CM026 Governance researchers flag biosecurity and dual-use review as an emerging adoption gate specifically relevant to predictive-evolution and pathogen-relevant applications. SM013, SM015
CM027 Wet-lab validation loops remain capital-intensive even for software-forward AI-biology platforms, tempering pure-software margin assumptions. SM015, SM013
CM028 Retained analyst reports materially contradict each other on both base-year size and CAGR for the same nominal category. SM007, SM005
CM029 Closing the gap between broad analyst TAM figures and an Astromech-specific SAM/SOM requires the company to disclose named pilots, deal structure, or pipeline detail not currently public. SM025, SM027
CM030 No retained analyst category explicitly names "predictive evolutionary biology" the way Astromech frames its own mission, so all sizing here is a proxy from adjacent categories. SM021, SM019
CM031 Astromech's own homepage language emphasizes multi-species and evolutionary modeling rather than a single named commercial category, reinforcing the proxy-based sizing approach used in this chapter. SM021, SM019
CM032 Colossal's spinout framing positions Astromech as serving medicine, agriculture, and programmable biology broadly, which supports treating multiple analyst categories as relevant rather than picking one. SM027, SM023
CM033 D Magazine's reporting describes Astromech's applications as spanning multiple industries without naming a specific first commercial market, consistent with the multi-lens sizing approach used here. SM025, SM023
CM034 Recursion's public partner roster demonstrates the pharma buyer/user/payer pattern this chapter uses as the closest available analog for Astromech's likely go-to-market. SM017
CM035 GamesBeat's coverage of Astromech's spinout and valuation gives no indication of a signed commercial pilot as of the reporting date. SM023
CM036 Intuition Labs' 2026 analysis of AI-biologics investment trends frames pharma partnership deal structures as the dominant commercialization path for platform-stage AI-biology companies. SM011
CP001 Recursion Pharmaceuticals, Insilico Medicine, Isomorphic Labs, and Schrödinger represent the closest direct competitive set to Astromech's general AI-biology platform positioning. SP001, SP017
CP002 Generate:Biomedicines, Absci, and LabGenius compete in an adjacent generative protein/antibody design niche rather than Astromech's broader multi-species framing. SP007, SP024
CP003 BenevolentAI represents an earlier-generation knowledge-graph approach that has publicly repositioned its strategy. SP026
CP004 Relay Therapeutics competes with a dynamics-based structural-biology technical approach distinct from Astromech's evolutionary framing. SP022
CP005 In-house pharma bioinformatics teams and traditional screening remain the default status-quo substitute across the category. SP001, SP010
CP006 Large pharma R&D organizations increasingly build internal computational biology capability, creating an internal-build threat for external platform vendors. SP020, SP010
CP007 Well-funded AI labs pivoting into biology, following the Isomorphic Labs pattern, are a plausible source of new entrants into this competitive set. SP017, SP019
CP008 Recursion Pharmaceuticals is a public company with a disclosed partner roster spanning multiple pharma collaborators. SP001, SP010
CP009 Insilico Medicine completed a Hong Kong Stock Exchange listing with disclosed financial statements. SP005, SP003
CP010 Isomorphic Labs raised a large Series B round with continued strategic backing tied to its Alphabet lineage. SP017, SP019
CP011 Schrödinger is a public company (Nasdaq-listed) with disclosed software and internal-pipeline revenue. SP020, SP010
CP012 LabGenius raised a £35 million Series B with a named pharma partnership disclosed alongside the round. SP014, SP016
CP013 Absci is a public company integrating wet-lab experimentation with generative AI design. SP024
CP014 Astromech's reported ~$2 billion valuation against $40.5 million of disclosed capital produces a materially higher valuation-to-capital ratio than any named competitor in this set. SP010, SP005
CP015 Recursion emphasizes large-scale phenotypic and image-based screening as its primary technical approach. SP001
CP016 Isomorphic Labs and Schrödinger both emphasize physics- and structure-based modeling lineages in their public technical framing. SP019, SP020
CP017 Generate:Biomedicines, Absci, and LabGenius emphasize generative protein or antibody design with wet-lab feedback loops. SP007, SP024
CP018 No retained public source shows any named competitor pursuing multi-species evolutionary modeling as a stated core capability. SP001, SP019
CP019 Pricing across the AI-biology category is rarely published; most competitors monetize through platform licensing plus milestone or royalty structures. SP001, SP009
CP020 Astromech has disclosed no pricing, packaging, or deal-structure information in any retained public source. SP029, SP028
CP021 Public-company competitors gain a distribution and credibility channel through investor relations and analyst coverage that Astromech currently lacks. SP010, SP005
CP022 Switching costs in this category are driven by deep data-integration into active pharma discovery programs rather than simple software lock-in. SP001, SP020
CP023 Multi-homing across more than one AI-biology vendor appears common among large pharma buyers given the diversity of disclosed partnerships across competitors. SP007, SP014
CP024 Already-partnered incumbents such as Recursion, Schrödinger, and Isomorphic Labs are structurally favored over new entrants once a pharma integration is live. SP001, SP017
CP025 BCG's research explicitly cautions that synthetic-biology and AI-biology commercialization claims across the industry outrun demonstrated industrial-scale results. SP030
CP026 No retained public source documents durable displacement of an incumbent screening or discovery method by any named AI-native competitor. SP001, SP010
CP027 Rising numbers of well-capitalized entrants using similar generative and multi-omics techniques create commoditization risk across the whole competitive set, including Astromech. SP007, SP024
CP028 Astromech's multi-species evolutionary modeling claim has no public head-to-head benchmark against any named competitor. SP029, SP001
CP029 Astromech currently has zero disclosed named pharma partnerships, versus at least one for every other competitor profiled in this chapter. SP028, SP014
CP030 Recursion, Schrödinger, and Insilico Medicine have each publicly signaled continued platform-expansion and partnership-growth strategy rather than a pivot away from AI-biology. SP001, SP020
CP031 BenevolentAI's public disclosures indicate a strategic repositioning toward narrower, partnership-led programs after an earlier broader-platform push. SP026
CP032 No retained public source discloses independent trust, safety, or regulatory-posture certification for any named competitor beyond general corporate compliance pages. SP001, SP020
CP033 Public disclosure on competitor trust and regulatory posture is thin across the entire category, not just for Astromech, which limits how far this comparison can be pushed. SP019, SP024
CP034 No retained public source documents litigation, restructuring, or a failed partnership for any of the nine named competitors profiled in this chapter. SP001, SP007
CP035 Absence of disclosed adverse events across competitors should be read as a light screen rather than proof of a clean record, consistent with the same caveat applied to Astromech in Company Overview. SP024, SP026
CI001 No retained public source discloses a named revenue stream, pricing sheet, or recognition policy for Astromech as of 2026-07-03. SI001, SI024
CI002 Plausible revenue mechanisms for Astromech include direct platform licensing, co-development/milestone-royalty programs, and applied government or agbio contracts, based on category norms among comparable AI-biology platforms. SI008, SI006
CI003 Schrödinger discloses a dual software-license-plus-internal-pipeline revenue model, while Absci discloses integrated wet-lab-plus-generative-AI revenue, both plausible analogues for how Astromech could eventually monetize. SI008, SI006
CI004 No retained source discloses a pricing sheet, license template, or deal-structure example for Astromech. SI001, SI024
CI005 Recursion discloses partnership-level economics in aggregate through investor-relations materials without publishing per-deal pricing. SI010
CI006 Revenue-recognition treatment cannot be assessed for Astromech without a disclosed contract structure, milestone schedule, or royalty rate. SI004, SI024
CI007 Isomorphic Labs, Relay Therapeutics, Recursion, and Schrödinger all structure go-to-market around long-cycle, high-touch strategic partnerships rather than a transactional or self-serve sales funnel. SI012, SI010
CI008 No retained public source discloses a CAC, payback period, or channel-economics proxy specific to Astromech. SI001, SI014
CI009 A high-touch, multi-quarter enterprise sales motion, if it applies to Astromech by category analogy, would require dedicated business-development headcount not currently disclosed. SI016, SI014
CI010 Recursion and Schrödinger report compute infrastructure and R&D headcount as their largest disclosed cost drivers, a plausible analogue for Astromech's cost base. SI010, SI008
CI011 Absci's integrated wet-lab-plus-AI model carries materially higher capital intensity than pure computational-modeling peers because physical experimentation adds facility, reagent, and instrumentation cost. SI006, SI018
CI012 No retained source discloses Astromech's gross margin, working-capital position, or capital-expenditure plan. SI001, SI004
CI013 If Astromech's stated multi-omics and evolutionary-modeling ambitions require wet-lab validation, its capital intensity could rise toward the Absci pattern rather than remaining at a lighter software-only cost profile. SI006, SI026
CI014 No retained public source discloses Astromech revenue, ARR, GMV, unit volume, active users, or a named customer count. SI001, SI024
CI015 Schrödinger, Recursion, Absci, and Insilico Medicine all disclose quarterly revenue and, in several cases, customer or partner counts through investor-relations channels, unlike Astromech. SI008, SI027
CI016 A third-party estimated-scale data source provides an approximate Astromech headcount proxy, but no official disclosure corroborates it. SI014, SI016
CI017 No retained source discloses Astromech's current cash balance or monthly burn rate. SI004, SI024
CI018 Using the disclosed $40.5M raised as a ceiling and category-typical early-stage burn proxies, a plausible Astromech runway estimate falls in the 12-to-24-month range, though this is not a confirmed figure. SI004, SI006
CI019 No debt facility, project-finance obligation, or credit line has been disclosed for Astromech in any retained public source. SI004, SI001
CI020 A plausible next-round trigger for Astromech would follow either a named commercial-partnership announcement or approaching the lower bound of the estimated runway window. SI004, SI026
CI021 Company Overview establishes Astromech's funding chronology through its 2026 Form D filing; this chapter references that chronology without restating or duplicating its claim ids. SI004
CI022 BCG's industrial-scale-up research cautions that synthetic-biology and AI-biology commercialization claims across the category tend to outrun demonstrated financial results, which applies directly to Astromech's valuation narrative. SI023
CI023 The absence of any disclosed revenue, cash-and-burn position, or named paying customer are the three most consequential financial diligence blockers for Astromech today. SI001, SI004
CI024 Astromech's reported ~$2B valuation against $40.5M of disclosed capital produces a valuation-to-capital ratio materially higher than public-company peers such as Absci and Schrödinger. SI018, SI020
CI025 No retained public source documents burn-rate stress, layoffs, or financing difficulty at Recursion, Schrödinger, Absci, or Insilico Medicine as of 2026-07-03. SI010, SI027
CI026 A disclosed current revenue figure, cash position, or named paying customer would most materially change Astromech's investment thesis if obtained. SI004, SI024
CI027 GeekWire's stealth-launch coverage described Astromech's applications broadly across medicine, agriculture, and biosecurity without naming a specific first paying market, consistent with the revenue-stream ambiguity noted elsewhere in this chapter. SI016, SI024
CI028 Relay Therapeutics discloses its financing and cash position through SEC-filed investor-relations materials, providing a public-company benchmark for the disclosure standard Astromech has not yet reached. SI012, SI008
CI029 Insilico Medicine's Hong Kong Stock Exchange filing discloses audited financial statements, an even higher disclosure bar than the investor-relations pages used by several other named peers. SI027, SI006
CI030 Datanyze's estimated company-scale data for Astromech is a third-party modeled estimate rather than a scraped or disclosed official figure, which limits how much weight it can bear in financial analysis. SI014
CI031 Category-typical burn rates at early-stage computational-biology platforms, inferred from Absci and Schrödinger's disclosed early-stage cost structures, support treating 12-24 months as a reasonable proxy range rather than a company-specific fact for Astromech. SI006, SI008
CI032 Colossal Biosciences' own $200M Series C disclosure provides one data point on the capital intensity of Ben Lamm-affiliated biotechnology ventures, though it is not a direct Astromech financial disclosure. SI026
CI033 No retained public source discloses any customer-concentration risk for Astromech specifically, because no named customer has been disclosed at all. SI001, SI024
CI034 Public-company peers Absci and Schrödinger both trade at valuation levels directly observable through stock-analysis data, offering a transparent comparison point that Astromech's private, undisclosed valuation basis lacks. SI018, SI020
CI035 Recursion's investor-relations disclosures show no evidence of covenant breach, going-concern language, or emergency financing as of the most recent reporting period. SI010
CE001 Astromech positions its Large Life Model as a general-purpose biological AI system for multi-species evolutionary and multi-omics modeling. SE004, SE001
CE002 No retained public source documents a concrete Astromech user workflow, API, or deliverable output format. SE004, SE007
CE003 Schrödinger and Absci both publish workflow-level technical documentation describing how a user or partner engages their respective platforms, a disclosure depth Astromech has not matched. SE015, SE013
CE004 No retained source enumerates discrete Astromech modules, SKUs, or a product-line map; public narrative describes a single unified platform. SE004, SE001
CE005 Recursion discloses a named "Recursion OS" product suite and Schrödinger discloses a modular physics-based simulation product line, both more granular than Astromech's single-platform framing. SE023, SE015
CE006 Without a disclosed module map, no part of Astromech's platform can be assessed as production-ready versus aspirational. SE004, SE007
CE007 No retained public source discloses Astromech's model architecture family, training-data provenance, or compute infrastructure. SE004, SE001
CE008 NVIDIA BioNeMo documents a foundation-model-plus-fine-tuning architecture for biomolecular AI as a category reference point for what mature architecture disclosure looks like. SE011, SE023
CE009 Schrödinger discloses a physics-based simulation engine architecture and Insilico Medicine discloses a generative chemistry and target-identification pipeline architecture, both more detailed than Astromech's public technical narrative. SE015, SE017
CE010 Astromech's architecture claim should be treated as unverified until comparable technical documentation, benchmark data, or independent review is made public. SE004, SE011
CE011 No retained public source discloses an Astromech deployment model, integration surface, or reliability/uptime metric. SE004, SE007
CE012 Recursion's public GitHub organization and Isomorphic Labs' engineering job postings provide developer-signal proxies for platform engineering activity that has no direct Astromech equivalent. SE009, SE025
CE013 Astromech's careers page is the closest available developer-signal proxy given the absence of a public GitHub organization, API documentation, or engineering community. SE007, SE025
CE014 No retained public source discloses a public Astromech product roadmap or release milestone beyond its 2026 stealth-exit launch narrative. SE001, SE007
CE015 Astromech's stated differentiation — multi-species evolutionary modeling breadth — does not overlap with any named competitor's stated core technical focus. SE004, SE023
CE016 No retained public source discloses a patent filing or proprietary dataset description supporting Astromech's differentiation claim. SE001, SE004
CE017 Recursion's phenotypic-screening wedge, Schrödinger/Isomorphic Labs' physics-based wedge, and Generate/Absci/LabGenius' generative-design wedge each occupy a distinct technical niche that does not directly contest Astromech's evolutionary-modeling claim. SE023, SE015
CE018 Astromech has not published a dedicated trust, security, or compliance page beyond general privacy and terms-of-service documents. SE001, SE004
CE019 No retained public source discloses a security certification, data-governance framework, or biosafety/biosecurity compliance statement specific to Astromech's platform. SE001, SE007
CE020 Schrödinger and Absci, as public companies, disclose more formal compliance and quality-control language through investor and technical materials than Astromech has published to date. SE015, SE013
CE021 No public evidence of a security incident, data breach, or compliance failure has been identified for Astromech in any retained source. SE001, SE004
CE022 Astromech's platform plausibly depends on undisclosed compute infrastructure, undisclosed training-data sources, and its Colossal Biosciences parent/spinout relationship. SE004, SE001
CE023 The Colossal Biosciences relationship functions as both a strategic asset (founder credibility, capital access) and a dependency (reputational and narrative linkage) for Astromech's technology roadmap. SE001, SE007
CE024 An independent technical benchmark or third-party architecture review would most reduce diligence uncertainty about Astromech's core modeling claim. SE004, SE011
CE025 Isomorphic Labs' research collaboration with Johnson & Johnson demonstrates a category pattern of deep technical integration with a named pharma partner, a proof point Astromech has not yet disclosed for itself. SE027
CE026 Generate:Biomedicines and LabGenius both disclose named technology pages describing generative protein or antibody design architecture, providing additional category reference points for technical disclosure depth. SE019, SE021
CE027 Astromech's public materials describe applications spanning medicine, agriculture, and biosecurity, but no retained source ties any specific deliverable, output, or interface to any one of these application areas. SE001, SE004
CE028 No retained source discloses an example input dataset or a sample prediction output for Astromech's Large Life Model, which limits independent assessment of workflow fidelity. SE004, SE007
CE029 Insilico Medicine's Pharma.AI platform documentation separates generative chemistry, target identification, and clinical-trial-design modules, illustrating a category pattern of modular technical disclosure that Astromech has not matched. SE017
CE030 Absci's technology page documents a named integrated wet-lab-plus-generative-AI design loop as a distinct product capability, another example of module-level disclosure absent from Astromech's public materials. SE013
CE031 No retained source discloses whether Astromech's platform runs on a proprietary compute cluster, a public cloud provider, or a hybrid infrastructure model. SE004, SE001
CE032 LabGenius's EVA platform technology page documents an automated evolutionary antibody-engineering architecture, offering a directly relevant category comparison point to Astromech's stated evolutionary-modeling ambition despite the narrower antibody-only scope. SE021
CE033 No retained source discloses whether Astromech has filed for or been granted any patent related to its evolutionary or multi-omics modeling methods. SE001, SE004
CE034 Generate:Biomedicines' technology disclosures describe a generative-biology architecture focused specifically on de novo protein design, a narrower and more specific technical claim than Astromech's broad multi-species framing. SE019
CE035 No retained source discloses whether Astromech has undergone or plans to undergo an independent security or biosafety audit. SE001, SE007
CU001 Astromech has not disclosed a customer base, buyer, user, or payer segmentation in any retained public source. SU022, SU024
CU002 Comparable AI-biology competitors segment buyers into large pharma R&D organizations, mid-cap biopharma, and long-term strategic collaborators, based on their disclosed named partnerships. SU001, SU006
CU003 If Astromech pursues a similar buyer profile to its named competitors, R&D or business-development leadership would plausibly be the buyer and R&D budget the payer, consistent with the buyer map established in Market Analysis. SU009, SU010
CU004 No retained public source discloses any Astromech deployment, pilot, account count, or utilization metric. SU022, SU024
CU005 Recursion's Bayer collaboration and NVIDIA infrastructure partnership have continued across multiple disclosed phases, illustrating a multi-year category adoption pattern. SU001, SU004
CU006 Isomorphic Labs' Johnson & Johnson collaboration was structured as a multi-modality, multi-year research program from inception, illustrating a land-and-expand adoption pattern typical of this category. SU009, SU001
CU007 Bayer's named research collaboration with Recursion is independently corroborated through Bayer's own newsroom disclosure, describing an oncology-focused, production-stage partnership. SU001, SU004
CU008 Johnson & Johnson's collaboration with Isomorphic Labs, Sanofi's collaboration with LabGenius, and Merck's and AstraZeneca's collaborations with Absci and BenevolentAI are each independently disclosed named pharma partnerships at production stage. SU009, SU006
CU009 Eli Lilly's collaboration with Schrödinger and Novartis's collaboration with Relay Therapeutics are each named, ongoing, multi-year partnerships disclosed on the respective company's own site. SU010, SU012
CU010 The named customer-proof census in this chapter is a representative sample of disclosed partnerships across seven comparable competitors, not an exhaustive census of every partnership each competitor holds, and Astromech itself contributes zero rows to this census. SU018, SU020
CU011 No retained source discloses an Astromech net/gross revenue retention figure, churn rate, or renewal event, because no customer relationship has been disclosed to measure. SU022, SU024
CU012 Multi-year, continuing collaborations such as Recursion-Bayer and Schrödinger-Lilly suggest that once a partnership is established in this category, it tends to be durable and multi-phase rather than single-transaction. SU001, SU010
CU013 No customer-satisfaction proxy (review, testimonial, or conference talk referencing Astromech by name) has been identified in any retained source. SU022, SU024
CU014 Comparable competitor partnerships illustrate a land-and-expand pattern where an initial single-program collaboration can expand into a broader multi-program relationship over time. SU001, SU016
CU015 If Astromech secures only a single named partner, over-reliance on that account would represent a material concentration risk until a second and third relationship are disclosed. SU014, SU016
CU016 Astromech's current customer-risk profile is better described as customer absence than customer concentration, since there is no revenue base to concentrate around. SU022, SU024
CU017 Astromech's origination channel plausibly runs through the Colossal Biosciences network given its spinout relationship, though no retained source confirms a non-Colossal-sourced lead or partnership. SU022, SU024
CU018 Large-pharma buyers in this category typically require multi-stage procurement involving R&D leadership sign-off, data-security review, and often a pilot period before a production collaboration is signed. SU009, SU006
CU019 Seven distinct named pharma partnerships are disclosed across the comparable competitor set profiled in this chapter — Bayer, Johnson & Johnson, Sanofi, Eli Lilly, Novartis, AstraZeneca, and Merck. SU001, SU009
CU020 The absence of any named customer, pilot, or reference account is the single most consequential customer-evidence gap for underwriting Astromech today. SU022, SU024
CU021 BCG's commercialization-risk research suggests category-wide customer-adoption claims tend to outpace demonstrated commercial results, which is directly relevant to assessing any future Astromech customer-adoption narrative. SU026
CU022 A single named pilot or customer disclosure, ideally with a stated use case and outcome, would most quickly close the customer-evidence gap for Astromech. SU022, SU024
CU023 Government, academic, or non-pharma customer segments are plausible for Astromech given its stated agriculture and biosecurity applications, but no retained source discloses any specific engagement in these segments. SU024, SU022
CU024 No retained public source documents any named competitor customer relationship at risk of non-renewal, churn, or public complaint. SU020, SU018
CU025 Sanofi's collaboration with LabGenius is corroborated on LabGenius's own partner page, describing an antibody discovery collaboration disclosed alongside a Series B funding update. SU006
CU026 Merck's collaboration with Absci and AstraZeneca's collaboration with BenevolentAI are each disclosed on the respective partner pages as production-stage generative-design or knowledge-graph-driven collaborations. SU016, SU014
CU027 Insilico Medicine's partners page and Recursion's partners page each list additional named collaborations beyond the single flagship example highlighted in this chapter, supporting the sample-coverage framing used for the enumeration table. SU018, SU020
CU028 NVIDIA's infrastructure partnership with Recursion is a technology/compute dependency rather than a pharma buyer relationship, and is therefore treated as partner-proof rather than customer-proof in this chapter's classification. SU004
CU029 No retained source discloses whether Astromech has begun any formal sales, business-development, or partnership-outreach process with a prospective customer. SU022, SU024
CU030 Relay Therapeutics' partner page discloses its collaboration with Novartis alongside other named oncology-focused partnerships, consistent with the broader category pattern of multiple concurrent pharma relationships. SU012
CU031 Schrödinger's collaboration with Eli Lilly is disclosed as a long-term computational-discovery partnership on Schrödinger's own site, consistent with the durability pattern discussed in the retention section of this chapter. SU010
CU032 The complete absence of any Astromech customer segmentation, adoption, or retention data means every quantitative figure in this chapter's tables for Astromech itself is a disclosed zero rather than an estimate. SU022, SU024
CU033 Astromech's stated cross-industry applications (medicine, agriculture, biosecurity) imply a plausible customer base broader than the pharma-only buyer set used as the category comparison in this chapter, but no segment-specific evidence supports this broader thesis yet. SU022, SU024
CU034 A land-and-expand pattern observed at Bayer-Recursion and Merck-Absci suggests that, once secured, an initial pharma partner is more likely to expand its relationship than to churn within the first several years. SU001, SU016
CU035 Novartis's and AstraZeneca's named collaborations with Relay Therapeutics and BenevolentAI, respectively, further corroborate that large pharma buyers in this category typically maintain more than one concurrent AI-biology platform relationship, reinforcing the multi-homing pattern noted in Competitors. SU012, SU014
CR001 Astromech's single largest risk category is disclosure risk: no financials, named customers, technical benchmarks, or compliance certifications have been published as of 2026-07-03. SR030, SR032
CR002 No retained source discloses any risk register, formal risk-management framework, or board-level risk oversight process at Astromech. SR030, SR013
CR003 Astromech's top risks plausibly transmit sequentially from disclosure gaps and category regulatory uncertainty through slower customer acquisition to financing and valuation pressure. SR032, SR034
CR004 No retained public source discloses any Astromech-specific litigation, enforcement action, or regulatory sanction. SR013, SR011
CR005 The FDA has published guidance on considerations for using artificial intelligence to support regulatory decision-making in drug and biological product development, formalized via a Federal Register notice. SR003, SR005
CR006 The FDA and EMA jointly announced common principles for the use of AI across the medicine development lifecycle, indicating active cross-jurisdictional regulatory attention to this category. SR001, SR003
CR007 No retained source discloses whether Astromech has adopted any sequence-screening or biosecurity-compliance posture consistent with NIST's biosecurity-for-synthetic-nucleic-acid-sequences program. SR006, SR030
CR008 NIST maintains an active biosecurity screening program for synthetic nucleic-acid sequences that applies by category to any platform modeling genomic or evolutionary sequences. SR006, SR021
CR009 Astromech's published privacy policy and terms-and-conditions pages are standard-form web boilerplate with no AI-specific or biosecurity-specific data-governance provisions disclosed. SR009, SR011
CR010 Astromech's corporate registration record shows an active status with a registered agent on file as of the most recent retained registry snapshot. SR013
CR011 No retained source discloses any Astromech manufacturing, facilities, or physical supply-chain exposure, consistent with an early-stage computational platform. SR030, SR032
CR012 No retained source discloses any Astromech security certification, penetration-test result, or incident-response history. SR030, SR011
CR013 If Astromech's stated wet-lab validation ambitions materialize, the company would likely take on facilities, reagent-supply, and biosafety-review operational risk not present in its current disclosed computational-only scope. SR032, SR021
CR014 Peer-reviewed governance literature identifies the absence of upstream risk-benefit review processes as a material, category-wide operational gap for dual-use AI-biology systems. SR015, SR017
CR015 The Center for Health Security's AI-and-biosecurity governance research similarly identifies a persistent lack of formal governance frameworks across the AI-biology category as of its publication. SR019
CR016 No retained source discloses any Astromech platform uptime, reliability metric, or outage history, consistent with the absence of any disclosed production customer. SR030, SR032
CR017 Astromech is a direct spinout of Colossal Biosciences and shares founder, brand, and origination-network dependency with its parent company. SR033, SR032
CR018 Colossal Biosciences' 2022 spinout of Form Bio is a directly comparable precedent illustrating the parent company's established pattern of spinning out computational-biology ventures. SR027, SR029
CR019 No retained source names Astromech's cloud, compute, or infrastructure provider. SR030, SR033
CR020 Astromech's only disclosed capital-provider relationships are an undisclosed-investor seed round and a follow-on extension round, indicating a concentrated and only partially named financing-counterparty base. SR022, SR023
CR021 No retained source discloses any regulator-specific relationship or dependency for Astromech beyond general category-wide regulatory context. SR013, SR011
CR022 Astromech's single most material critical dependency is its shared ecosystem with Colossal Biosciences, ahead of any disclosed infrastructure or capital-provider dependency. SR033, SR032
CR023 Astromech's public identity and credibility are substantially concentrated in founder Ben Lamm, who simultaneously leads Colossal Biosciences and has founded multiple prior ventures. SR032, SR025
CR024 No retained source names a chief technology officer, chief scientist, or other technical leadership figure at Astromech beyond its founder. SR030, SR026
CR025 No retained source discloses Astromech's current headcount, hiring pace, or organizational structure. SR030, SR024
CR026 No retained source discloses any Astromech sales, business-development, or commercial go-to-market function, consistent with the company's stealth-stage posture and zero disclosed customers. SR030, SR032
CR027 Astromech has disclosed a $30 million initial Form D offering (2025-08-12) followed by a roughly $10.5 million extension (2026-03-26), bringing total disclosed capital to $40.5 million against a stated $2 billion valuation, with no disclosed revenue or burn rate. SR023, SR022
CR028 A large stated valuation with no disclosed revenue implies Astromech will likely require substantial further capital before any commercial validation, creating financing-gap risk if a follow-on round is delayed. SR022, SR025
CR029 If Astromech's stated wet-lab validation ambitions materialize, capital intensity would likely rise further, compounding existing financing-gap risk noted in Financials. SR032, SR026
CR030 No retained source discloses any fraud, credit exposure, or working-capital stress signal at Astromech, though this reflects the absence of any disclosed financial statement rather than a confirmed clean bill of health. SR013, SR023
CR031 No retained source discloses any mitigation specific to Astromech for its top-ranked risks; all mitigation assessment in this chapter is qualitative and category-level only. SR030, SR015
CR032 Monitorable thesis-break triggers for Astromech include failure to disclose a first named customer within a reasonable post-stealth window, a failed or delayed follow-on financing round, an undisclosed founder departure, or an adverse category-wide regulatory or biosecurity action. SR032, SR003
CR033 The highest-value diligence asks for Astromech are a named customer or pilot reference, an audited or reviewed financial statement, a compliance/certification roadmap, and a named technical leadership roster. SR030, SR008
CR034 Synthetic-biology commercialization research indicates that industrial-scale adoption across this broader category has historically taken longer than initial company narratives suggest, a relevant category-level caution for Astromech's own timeline. SR034
CR035 No retained source discloses whether Astromech has adopted any formal quality-management system or biosafety-review committee, distinct from the narrower absence of a security-certification disclosure. SR030, SR021
CR036 NIST's engineering and synthetic-biology program materials describe measurement and quality standards relevant to biological engineering generally, providing a category benchmark against which Astromech discloses no equivalent internal standard. SR021
CR037 Astromech's privacy policy and terms pages do not reference any AI-specific regulatory framework such as the FDA's AI guidance or the EMA's joint AI principles, reinforcing the legal-disclosure gap identified elsewhere in this chapter. SR011, SR003
CR038 Bizapedia's registry record and Astromech's own terms page are the only two retained legal-adjacent sources for the company, underscoring how thin Astromech's public legal-disclosure footprint is relative to the regulatory context surrounding its category. SR013, SR011
CR039 The Form Bio precedent shows that Colossal Biosciences spinouts have historically raised independent capital shortly after spinout, suggesting Astromech's own financing pattern (seed plus rapid extension) is consistent with, not anomalous relative to, its parent's established playbook. SR027, SR029
CR040 The joint FDA/EMA AI-in-medicine principles and the NIST biosecurity-screening program together indicate that regulators in multiple jurisdictions are actively building category-specific oversight frameworks, meaning Astromech's current lack of disclosed compliance posture is a growing rather than static risk. SR001, SR006
CV001 Astromech's bull case rests on category momentum among AI-biology platforms and the credibility conferred by its Colossal Biosciences origin, but neither factor is Astromech-specific proof. SV007, SV027
CV002 Astromech's anti-thesis is that every company-specific proof point required to underwrite a $2 billion valuation — revenue, named customers, technical benchmarks, and compliance posture — remains undisclosed as of 2026-07-03. SV001, SV003
CV003 A disclosed named pharma partnership or independently validated technical benchmark would be the single most valuation-relevant disclosure Astromech could make in the near term. SV005, SV001
CV004 Continued non-disclosure of revenue, customers, or a technical benchmark past a reasonable post-stealth window is the single strongest evidence for the bear case today. SV003, SV029
CV005 Given the volume of unresolved disclosure gaps documented across market, competitive, financial, product, customer, and risk chapters, the appropriate recommendation is Watch / Do Not Lead rather than a directional buy or sell call. SV001, SV003
CV006 Confidence in this recommendation is low-to-medium because it is built primarily from category analogy rather than Astromech-specific verified evidence. SV007, SV005
CV007 Astromech's risk rating is High given the combination of disclosure risk, category-wide regulatory exposure, key-person concentration, and unverified valuation documented in this diligence. SV003, SV030
CV008 No independent filing, analyst note, or disclosed revenue multiple corroborates Astromech's reported ~$2 billion valuation; it remains a private-round narrative figure only. SV001, SV005
CV009 Astromech's disclosed financing history consists of a $30 million initial Form D offering and a roughly $10.5 million extension, totaling $40.5 million in disclosed capital against a reported ~$2 billion valuation. SV017, SV018
CV010 No retained source discloses any Astromech dilution schedule, liquidation-preference structure, or cap-table detail. SV028, SV001
CV011 Astromech's valuation-to-capital ratio is unusually high relative to public comparables such as Absci and Schrödinger, whose market capitalizations are grounded in disclosed revenue and clinical or commercial milestones rather than private-round narrative alone. SV025, SV026
CV012 Given the fully unverified nature of the $2 billion figure, entry discipline today should assume it is unconfirmed until an independent data point (a priced round, disclosed revenue, or public listing) becomes available. SV001, SV003
CV013 In the bull scenario, a disclosed named pharma partnership and technical benchmark within 12 months would support a follow-on round priced at or above the current $2 billion mark. SV023, SV022
CV014 In the base scenario, Astromech remains largely non-disclosing for another 12-18 months, plausibly raising a modest bridge or extension round at a flat-to-down valuation while working toward a first customer. SV007, SV005
CV015 In the bear scenario, failure to secure a named customer or timely follow-on financing would plausibly force a valuation write-down toward a level more consistent with the disclosed $40.5M capital base. SV030, SV003
CV016 The base scenario currently carries the highest probability signal given the consistent pattern of non-disclosure documented across every prior chapter of this diligence. SV001, SV005
CV017 The most relevant comparable set for Astromech spans Absci, Schrödinger, Recursion, Relay Therapeutics, BenevolentAI, and Insilico Medicine, each at a different public-market or late-private stage. SV025, SV026, SV023, SV013
CV018 No comparable company's valuation independently corroborates Astromech's reported $2 billion figure; each comparable's valuation is grounded in its own distinct disclosed financial or listing evidence. SV011, SV019
CV019 Astromech is materially earlier-stage than its most mature public comparables (Absci, Schrödinger, Recursion), each of which discloses revenue, clinical milestones, or named pharma partnerships that Astromech has not yet disclosed. SV021, SV022
CV020 The central limitation of any comparable-based valuation exercise for Astromech is the absence of a disclosed revenue, clinical, or partnership metric on which to anchor a multiple. SV024, SV020
CV021 Astromech is not exit-ready by any public-market or strategic-acquisition standard given the absence of disclosed revenue, named customers, or an independent financial statement. SV001, SV003
CV022 The most consequential thesis-break triggers for Astromech are a failed or down-valued follow-on financing round, continued absence of a named customer, an adverse category-wide regulatory action, or an undisclosed founder departure. SV005, SV029
CV023 The highest-priority final diligence asks are an audited or reviewed financial statement, a named customer or pilot reference, an independent technical benchmark, and a disclosed compliance/certification roadmap. SV007, SV003
CV024 A failed or materially down-valued follow-on financing round would directly undermine the entire $2 billion valuation narrative and require full re-underwriting of the investment thesis. SV001, SV030
CV025 No retained source discloses whether Pitchbook, CB Insights, or Crunchbase hold any independent cap-table or valuation corroboration for Astromech beyond public news reporting. SV001, SV005
CV026 Pitchbook's AI-in-biotech investment-trends report indicates elevated funding and valuation multiples across the category over the past several years, providing context for why Astromech's valuation, while high, is not without category precedent. SV007
CV027 BenevolentAI's Euronext listing illustrates how public markets can materially re-rate an AI-biology company's valuation once independent financial disclosure begins, a relevant precedent for how Astromech's eventual public-market value could differ from its current private narrative. SV011
CV028 Insilico Medicine's HKEX-listed IPO valuation provides a data point for achievable public-market valuations in the AI-biology category, useful context for assessing whether Astromech's $2 billion figure is plausible at eventual public-market scale. SV019
CV029 No retained source discloses any down-round, failed raise, or other adverse valuation event among Astromech's named comparable companies as of 2026-07-03. SV015, SV020
CV030 Synthetic-biology commercialization research indicates that valuation compression is a recurring category-wide risk once initial hype-driven private valuations meet slower-than-expected commercialization timelines. SV030
CV031 Datanyze's estimated scale profile for Astromech suggests a small headcount relative to its reported $2 billion valuation, reinforcing the valuation-to-execution-capacity gap noted elsewhere in this chapter. SV028
CV032 Colossal Biosciences' own $200 million Series C disclosure provides a parent-company data point on capital intensity within this ecosystem, indicating that large private rounds are an established pattern for Ben Lamm-led ventures generally. SV027
CV033 Siliconangle and TechCrunch coverage of Astromech's April 2026 valuation update are the only two independent press outlets that reported the $2 billion figure at the time it circulated, and neither cites a priced round or independent analyst source for the number. SV009, SV010
CV034 CB Insights' company-profile methodology typically derives private-company valuations from disclosed round data rather than independent modeling, meaning its Astromech profile likely reflects the same unverified narrative figure rather than a separately corroborated estimate. SV003
CV035 Relay Therapeutics' SEC EDGAR filings and stock-analysis statistics together provide one of the more complete disclosed-financial comparables in this set, in contrast to Astromech's complete absence of any equivalent filing. SV013, SV015
CV036 Recursion's investor-relations disclosures include a named pharma partnership (Bayer) and a disclosed cash position, both of which remain entirely undisclosed for Astromech, widening the comparable-evidence gap between the two companies. SV023
CV037 No retained source discloses whether Astromech has engaged an investment bank, financial advisor, or auditor in connection with its reported valuation or any future financing round. SV001, SV018
CV038 The wide low-confidence valuation range used in this chapter reflects the fact that Astromech's disclosed capital base ($40.5M) and its reported narrative valuation ($2B) differ by roughly fiftyfold, an unusually large spread even for an early-stage private AI company. SV018, SV007
CV039 Schrödinger's dual software-license-plus-pipeline business model, as disclosed in its investor-relations materials, offers a more diversified revenue base than the single-platform narrative Astromech has described publicly, a relevant limitation when using Schrödinger as a comparable. SV022
CV040 Across all six analyst-market-data and filing sources retained for this chapter, none independently prices or corroborates a private company at Astromech's specific $40.5M-capital-base-to-$2B-valuation ratio, underscoring that this ratio is an outlier even within the broader AI-biology comparable set. SV001, SV003
来源
编号出版方标题引文
SO001 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SO002 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SO003 Astromech Astromech technical homepage At the heart of Astromech is a mission to construct the Large Life Model.
SO004 Astromech Astromech technical homepage At the heart of Astromech is a mission to construct the Large Life Model.
SO005 Astromech Astromech careers signals on homepage
SO006 Astromech Astromech privacy policy
SO007 Astromech Astromech privacy policy
SO008 Astromech Astromech terms and conditions
SO009 Colossal Biosciences How Colossal Biosciences Turned De-Extinction Into a Multibillion-Dollar Science Movement
SO010 U.S. Securities and Exchange Commission Astromech AI Corp. Form D filing (2025)
SO011 U.S. Securities and Exchange Commission Astromech AI Corp. Form D filing (2026)
SO012 GamesBeat Colossal Biosciences spins out AI startup Astromech at $2B valuation
SO013 GamesBeat Colossal Biosciences spins out AI startup Astromech at $2B valuation
SO014 D Magazine Astromech, the New AI Startup from Ben Lamm of Colossal Biosciences
SO015 D Magazine Astromech, the New AI Startup from Ben Lamm of Colossal Biosciences
SO016 Axios Colossal spinout Astromech raises $40.5M at $2B valuation
SO017 Axios Colossal spinout Astromech raises $40.5M at $2B valuation
SO018 Yahoo Finance Ben Lamm's Astromech Secures $30 Million to Advance AI-Biotechnology Innovation
SO019 FirstWord Pharma Ben Lamm's Astromech Secures $30 Million to Advance AI-Biotechnology Innovation
SO020 USA Today Unicorn Start-up, Astromech, Emerges from Stealth at Two Billion Dollar Valuation
SO021 Science Times Predictive Evolutionary Trend Modeling Now a Possibility; Astromech Launches from Stealth at $2 Billion
SO022 Business Wire Colossal Biosciences Spins Out Form Bio
SO023 Dallas Innovates Form Bio Spins Out of Colossal Bioscience
SO024 Bizapedia Astromech AI Corp. company record
SO025 Boston Consulting Group Synthetic Biology Is Getting Closer to Industrial Scale
SM001 MarketsandMarkets AI in Biotechnology Market Report 2025-2035
SM002 MarketsandMarkets AI in Biotechnology Market Report 2025-2035
SM003 The Business Research Company Artificial Intelligence in Drug Discovery Market Report 2026
SM004 The Business Research Company Artificial Intelligence in Drug Discovery Market Report 2026
SM005 Global Market Insights Artificial Intelligence in Drug Discovery Market Size, Share – 2035
SM006 Global Market Insights Artificial Intelligence in Drug Discovery Market Size, Share – 2035
SM007 Research and Markets Artificial Intelligence (AI) in Drug Discovery Market Report 2025
SM008 Research and Markets Artificial Intelligence (AI) in Drug Discovery Market Report 2025
SM009 Custom Market Insights Global AI in Synthetic Biology Market Size, Share 2025-2034
SM010 Custom Market Insights Global AI in Synthetic Biology Market Size, Share 2025-2034
SM011 IntuitionLabs AI Biologics Discovery: 2026 Pharma Investment Trends
SM012 IntuitionLabs AI Biologics Discovery: 2026 Pharma Investment Trends
SM013 OECD Synthetic Biology in Focus
SM014 OECD Synthetic Biology in Focus
SM015 Boston Consulting Group Synthetic Biology Is Getting Closer to Industrial Scale
SM016 Boston Consulting Group Synthetic Biology Is Getting Closer to Industrial Scale
SM017 Recursion Pharmaceuticals Recursion partners
SM018 Recursion Pharmaceuticals Recursion partners
SM019 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SM020 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SM021 Astromech Astromech technical homepage At the heart of Astromech is a mission to construct the Large Life Model.
SM022 Astromech Astromech technical homepage At the heart of Astromech is a mission to construct the Large Life Model.
SM023 GamesBeat Colossal Biosciences spins out AI startup Astromech at $2B valuation
SM024 GamesBeat Colossal Biosciences spins out AI startup Astromech at $2B valuation
SM025 D Magazine Astromech, the New AI Startup from Ben Lamm of Colossal Biosciences
SM026 D Magazine Astromech, the New AI Startup from Ben Lamm of Colossal Biosciences
SM027 Colossal Biosciences How Colossal Biosciences Turned De-Extinction Into a Multibillion-Dollar Science Movement
SP001 Recursion Pharmaceuticals Recursion Pharmaceuticals homepage
SP002 Recursion Pharmaceuticals Recursion Pharmaceuticals homepage
SP003 Insilico Medicine Insilico Medicine about
SP004 Insilico Medicine Insilico Medicine about
SP005 Hong Kong Exchange Insilico Medicine HKEX filing
SP006 Hong Kong Exchange Insilico Medicine HKEX filing
SP007 Generate:Biomedicines Generate:Biomedicines home
SP008 Generate:Biomedicines Generate:Biomedicines home
SP009 Generate:Biomedicines Generate:Biomedicines investor relations
SP010 Stock Analysis Generate:Biomedicines stock statistics
SP011 Stock Analysis Generate:Biomedicines stock statistics
SP012 LabGenius LabGenius home
SP013 LabGenius LabGenius home
SP014 LabGenius LabGenius raises £35 million Series B
SP015 LabGenius LabGenius raises £35 million Series B
SP016 UKTN LabGenius raises £35m for AI-powered drug discovery
SP017 Isomorphic Labs Isomorphic Labs announces Series B investment round
SP018 Isomorphic Labs Isomorphic Labs announces Series B investment round
SP019 Isomorphic Labs Isomorphic Labs home
SP020 Schrödinger Schrödinger home
SP021 Schrödinger Schrödinger home
SP022 Relay Therapeutics Relay Therapeutics homepage
SP023 Relay Therapeutics Relay Therapeutics homepage
SP024 Absci Corporation Absci Corporation homepage
SP025 Absci Corporation Absci Corporation homepage
SP026 BenevolentAI BenevolentAI homepage
SP027 BenevolentAI BenevolentAI homepage
SP028 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SP029 Astromech Astromech technical homepage At the heart of Astromech is a mission to construct the Large Life Model.
SP030 Boston Consulting Group Synthetic Biology Is Getting Closer to Industrial Scale
SI001 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SI002 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SI003 U.S. Securities and Exchange Commission Astromech AI Corp. Form D filing (2025)
SI004 U.S. Securities and Exchange Commission Astromech AI Corp. Form D filing (2026)
SI005 U.S. Securities and Exchange Commission Astromech AI Corp. Form D filing (2026)
SI006 Absci Corporation Absci Corporation investor relations
SI007 Absci Corporation Absci Corporation investor relations
SI008 Schrödinger Schrödinger investor relations
SI009 Schrödinger Schrödinger investor relations
SI010 Recursion Pharmaceuticals Recursion Pharmaceuticals investor relations
SI011 Recursion Pharmaceuticals Recursion Pharmaceuticals investor relations
SI012 Relay Therapeutics Relay Therapeutics investor relations
SI013 Relay Therapeutics Relay Therapeutics investor relations
SI014 Datanyze Astromech company profile and estimated scale
SI015 Datanyze Astromech company profile and estimated scale
SI016 GeekWire Unicorn Astromech Launches From Stealth With Predictive Biology Capacity
SI017 GeekWire Unicorn Astromech Launches From Stealth With Predictive Biology Capacity
SI018 Stock Analysis Absci Corporation stock statistics
SI019 Stock Analysis Absci Corporation stock statistics
SI020 Stock Analysis Schrödinger stock statistics
SI021 Stock Analysis Schrödinger stock statistics
SI022 Stock Analysis Generate:Biomedicines stock statistics
SI023 Boston Consulting Group Synthetic Biology Is Getting Closer to Industrial Scale
SI024 D Magazine Astromech, the New AI Startup from Ben Lamm of Colossal Biosciences
SI025 D Magazine Astromech, the New AI Startup from Ben Lamm of Colossal Biosciences
SI026 Colossal Biosciences Colossal Biosciences Raises $200M Series C Round
SI027 Hong Kong Exchange Insilico Medicine HKEX filing
SE001 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SE002 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SE003 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SE004 Astromech Astromech technical homepage At the heart of Astromech is a mission to construct the Large Life Model.
SE005 Astromech Astromech technical homepage At the heart of Astromech is a mission to construct the Large Life Model.
SE006 Astromech Astromech technical homepage At the heart of Astromech is a mission to construct the Large Life Model.
SE007 Astromech Astromech careers signals on homepage
SE008 Astromech Astromech careers signals on homepage
SE009 GitHub Recursion Pharmaceuticals public GitHub organization
SE010 GitHub Recursion Pharmaceuticals public GitHub organization
SE011 NVIDIA NVIDIA BioNeMo platform for AI drug discovery
SE012 NVIDIA NVIDIA BioNeMo platform for AI drug discovery
SE013 Absci Corporation Absci technology platform overview
SE014 Absci Corporation Absci technology platform overview
SE015 Schrödinger Schrödinger physics-based computational platform
SE016 Schrödinger Schrödinger physics-based computational platform
SE017 Insilico Medicine Insilico Medicine Pharma.AI platform overview
SE018 Insilico Medicine Insilico Medicine Pharma.AI platform overview
SE019 Generate:Biomedicines Generate:Biomedicines generative biology technology
SE020 Generate:Biomedicines Generate:Biomedicines generative biology technology
SE021 LabGenius LabGenius EVA platform technology overview
SE022 LabGenius LabGenius EVA platform technology overview
SE023 Recursion Pharmaceuticals Recursion OS product overview
SE024 Recursion Pharmaceuticals Recursion OS product overview
SE025 Isomorphic Labs Isomorphic Labs careers and engineering hiring signals
SE026 Isomorphic Labs Isomorphic Labs careers and engineering hiring signals
SE027 Johnson & Johnson Innovation Isomorphic Labs enters research collaboration with Johnson & Johnson
SE028 Johnson & Johnson Innovation Isomorphic Labs enters research collaboration with Johnson & Johnson
SU001 Bayer Bayer and Recursion focus research collaboration on oncology
SU002 Bayer Bayer and Recursion focus research collaboration on oncology
SU003 Bayer Bayer and Recursion focus research collaboration on oncology
SU004 NVIDIA NVIDIA and Recursion speed pharma R&D with AI
SU005 NVIDIA NVIDIA and Recursion speed pharma R&D with AI
SU006 LabGenius LabGenius collaboration with Sanofi
SU007 LabGenius LabGenius collaboration with Sanofi
SU008 LabGenius LabGenius collaboration with Sanofi
SU009 Johnson & Johnson Innovation Isomorphic Labs enters research collaboration with Johnson & Johnson
SU010 Schrödinger Schrödinger collaboration with Eli Lilly
SU011 Schrödinger Schrödinger collaboration with Eli Lilly
SU012 Relay Therapeutics Relay Therapeutics partner and collaboration page
SU013 Relay Therapeutics Relay Therapeutics partner and collaboration page
SU014 BenevolentAI BenevolentAI collaboration with AstraZeneca
SU015 BenevolentAI BenevolentAI collaboration with AstraZeneca
SU016 Absci Corporation Absci partner and collaboration page
SU017 Absci Corporation Absci partner and collaboration page
SU018 Insilico Medicine Insilico Medicine partners and collaborations
SU019 Insilico Medicine Insilico Medicine partners and collaborations
SU020 Recursion Pharmaceuticals Recursion partners as customer proof
SU021 Recursion Pharmaceuticals Recursion partners as customer proof
SU022 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SU023 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SU024 D Magazine Astromech, the New AI Startup from Ben Lamm of Colossal Biosciences
SU025 D Magazine Astromech, the New AI Startup from Ben Lamm of Colossal Biosciences
SU026 Boston Consulting Group Synthetic Biology Is Getting Closer to Industrial Scale
SR001 European Medicines Agency EMA and FDA set common principles for AI in medicine development
SR002 European Medicines Agency EMA and FDA set common principles for AI in medicine development
SR003 U.S. Food and Drug Administration Considerations for the Use of Artificial Intelligence to Support Regulatory Decision-Making
SR004 U.S. Food and Drug Administration Considerations for the Use of Artificial Intelligence to Support Regulatory Decision-Making
SR005 Federal Register Federal Register notice on FDA AI guidance
SR006 National Institute of Standards and Technology Biosecurity for Synthetic Nucleic Acid Sequences
SR007 National Institute of Standards and Technology Biosecurity for Synthetic Nucleic Acid Sequences
SR008 National Center for Biotechnology Information Understanding Risks Related to Future Biotechnology Products
SR009 Astromech Astromech privacy policy
SR010 Astromech Astromech privacy policy
SR011 Astromech Astromech terms and conditions
SR012 Astromech Astromech terms and conditions
SR013 Bizapedia Astromech AI Corp. company record
SR014 Bizapedia Astromech AI Corp. company record
SR015 Frontiers Improving governance in the age of synthetic biology, artificial intelligence, and diverging threats
SR016 Frontiers Improving governance in the age of synthetic biology, artificial intelligence, and diverging threats
SR017 Frontiers Dual-use artificial intelligence and biology: upstream risk-benefit reviews
SR018 Frontiers Dual-use artificial intelligence and biology: upstream risk-benefit reviews
SR019 Johns Hopkins Center for Health Security AI and biosecurity: The need for governance
SR020 Johns Hopkins Center for Health Security AI and biosecurity: The need for governance
SR021 National Institute of Standards and Technology Engineering / synthetic biology
SR022 Axios Colossal spinout Astromech raises $40.5M at $2B valuation
SR023 Yahoo Finance Ben Lamm's Astromech Secures $30 Million to Advance AI-Biotechnology Innovation
SR024 FirstWord Pharma Ben Lamm's Astromech Secures $30 Million to Advance AI-Biotechnology Innovation
SR025 USA Today Unicorn Start-up, Astromech, Emerges from Stealth at Two Billion Dollar Valuation
SR026 Science Times Predictive Evolutionary Trend Modeling Now a Possibility; Astromech Launches from Stealth at $2 Billion
SR027 Business Wire Colossal Biosciences Spins Out Form Bio
SR028 Business Wire Colossal Biosciences Spins Out Form Bio
SR029 Dallas Innovates Form Bio Spins Out of Colossal Bioscience
SR030 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SR031 Astromech Astromech homepage Astromech describes itself as an autonomous biological intelligence system designed to navigate the deep architecture of life.
SR032 D Magazine Astromech, the New AI Startup from Ben Lamm of Colossal Biosciences
SR033 Colossal Biosciences How Colossal Biosciences Turned De-Extinction Into a Multibillion-Dollar Science Movement
SR034 Boston Consulting Group Synthetic Biology Is Getting Closer to Industrial Scale
SV001 PitchBook Astromech company profile
SV002 PitchBook Astromech company profile
SV003 CB Insights Astromech company profile
SV004 CB Insights Astromech company profile
SV005 Crunchbase Astromech organization profile
SV006 Crunchbase Astromech organization profile
SV007 PitchBook AI in biotech investment trends report
SV008 PitchBook AI in biotech investment trends report
SV009 SiliconANGLE Colossal spinout Astromech reaches $2B valuation
SV010 TechCrunch Astromech raises more at $2 billion valuation
SV011 BenevolentAI BenevolentAI investor and public-listing disclosures
SV012 BenevolentAI BenevolentAI investor and public-listing disclosures
SV013 U.S. Securities and Exchange Commission Relay Therapeutics SEC EDGAR filings
SV014 U.S. Securities and Exchange Commission Relay Therapeutics SEC EDGAR filings
SV015 Stock Analysis Relay Therapeutics stock statistics
SV016 Stock Analysis Relay Therapeutics stock statistics
SV017 U.S. Securities and Exchange Commission Astromech AI Corp. Form D filing (2025)
SV018 U.S. Securities and Exchange Commission Astromech AI Corp. Form D filing (2026)
SV019 Hong Kong Exchange Insilico Medicine HKEX filing
SV020 Stock Analysis Generate:Biomedicines stock statistics
SV021 Absci Corporation Absci Corporation investor relations
SV022 Schrödinger Schrödinger investor relations
SV023 Recursion Pharmaceuticals Recursion Pharmaceuticals investor relations
SV024 Relay Therapeutics Relay Therapeutics investor relations
SV025 Stock Analysis Absci Corporation stock statistics
SV026 Stock Analysis Schrödinger stock statistics
SV027 Colossal Biosciences Colossal Biosciences Raises $200M Series C Round
SV028 Datanyze Astromech company profile and estimated scale
SV029 GeekWire Unicorn Astromech Launches From Stealth With Predictive Biology Capacity
SV030 Boston Consulting Group Synthetic Biology Is Getting Closer to Industrial Scale
SV031 Boston Consulting Group Synthetic Biology Is Getting Closer to Industrial Scale