AITOPAI热报
Industry

Google 被指实质收购初创团队，留下空壳公司

Gergely Orosz@GergelyOrosz
Post
View the post
Why we picked it

Gergely Orosz 点破大厂新套路：不买公司只挖创始团队，Inflection、Covariant、Adept 都是这么被吸收的，绕开反垄断。

Google 被指通过收购创始团队的方式吸收一家初创公司，原公司则变成空壳。Gergely Orosz 在推文中将这种做法与其他大厂案例并列：Microsoft 与 Inflection AI、Amazon 与 Covariant 和 Adept AI。

Body

他指出大厂采用这种模式是为了规避反垄断审查。

Original · Gergely Orosz

Zero surprise: Google basically acquihired the founding team, leaving a zombie company behind.

There have been countless other examples of Big Tech doing the same: Microsoft with Inflection AI, Amazon with Covariant and Adept AI

All so that Big Tech doesn't trigger antitrust

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Industry" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI regulation, compliance frameworks, and sovereign compute policy. Key entities and technical topics under continuous tracking include Google, Inflection AI, Covariant, Adept AI.

Source attribution: curated from Gergely Orosz (@GergelyOrosz).

#Google#Inflection AI#Covariant#Adept AI#antitrust

Keep reading

Recent stories with the same tag or section
ModelsUS vs ChinaMulti-source84°00:58
Chinese contentGoogle 发布 Gemini 4 Hydrogen，6 周内第三款模型

Google 半个月左右就发一款 Gemini 4 新模型，这回的 Hydrogen 主打 agentic 工作流和日常使用，速度是卖点。从 Argon 到 Carbon 再到 Hydrogen，6 周三连发的节奏在 OpenAI 和 Anthropic 面前摆明了要拼迭代速度。做智能体应用的开发者可以拿它和 Carbon 对比一下延迟和价格。

#Gemini 4#Hydrogen#Google#agents
Event
Thang Luong@lmthangSource
PapersBuildersSelected00:34
Google paper introduces SynthCorp benchmarks for data agents

The key shift is moving evaluation from fixed, easily saturated datasets to generated environments with ground truth derived from simulation logs, which makes skill-targeted benchmarking reproducible at scale. Teams building enterprise data agents or text-to-SQL systems can use the seven estates and 256 tasks to identify exactly where multi-event chaining breaks their pipelines. The finding that state-of-the-art agents still fail on chained tasks gives labs a concrete target for the next round of agentic improvements.

#SynthCorp#Google#benchmarks#agents
DAIR.AI@dair_aiSource
ModelsUS vs ChinaMulti-source76°00:02
Chinese contentGoogle 发布 Gemini 4 Hydrogen：对标 Claude Sonnet 5.5 与 GPT-6.1 Terra

Google 这次把推理、速度和 1M 上下文塞进一个模型，比 3.8 Flash 快 2.5 倍，还用 Carbon 编排器调度多模型。如果你在 Claude、GPT 和 Gemini 之间挑主力模型，这条值得对比着看，做长视频或大文件分析的开发者尤其该关注。

#Gemini 4 Hydrogen#Google#Claude Sonnet 5.5#GPT-6.1 Terra
Event
Shubham Saboo@Saboo_Shubham_Source
ModelsUS vs China23:49
Chinese content群聊讨论称 Gemini 3.5 flash-lite 翻译性价比高，Gemini 4 Carbon 中文写作强

有人在小群里说 Gemini 3.5 flash-lite 翻译最省钱，Gemini 4 Carbon 写中文最像人，就是慢。不过这些都是口口相传的结论，没有具体跑分，博主自己也说要再验证。做翻译或中文写作选型的朋友可以先围观，别急着换模型。

#Gemini#Google#translation models#model selection
Event
向阳乔木@vista8Source