Сообщается, что Google ведет переговоры о сделке с Mechanize на сумму более 1,5 миллиарда долларов.

Что случилось

Business Insider, as reported by gHacks, says Google is in advanced talks with AI coding startup Mechanize over a deal valued at more than $1.5 billion. The proposed arrangement would license Mechanize’s technology and bring some employees to Google, but its terms remain unsettled and no agreement has been publicly announced.

Business Insider’s reporting, summarized by gHacks, says Google is discussing a transaction worth more than $1.5 billion with Mechanize. The reported structure is a technology-licensing agreement combined with the hiring of some Mechanize employees. Negotiations were still continuing when the report was published, and Business Insider said the details could change. Google and Mechanize declined to comment.

The source does not provide a signed agreement, a public announcement, or independently verified terms, so the transaction should be treated as a reported possibility rather than a completed acquisition or investment. The source describes Mechanize as an AI startup founded in 2025 by Tamay Besiroglu, Matthew Barnett, and Ege Erdil. Its work centers on virtual environments, benchmarks, and training data intended to help AI agents handle complex, ongoing tasks. Mechanize’s stated starting point is software engineering, with an eventual ambition of broader economic automation. The company says current coding agents remain limited by problems including reliability, long-context handling, and the difficulty of managing work that extends beyond narrow programming tasks.

gHacks places the report in the context of Google’s earlier dealings and internal AI development. It says Mechanize raised $9.1 million at a $500 million post-money valuation in April, making a potential transaction above $1.5 billion a substantial reported increase from that valuation. The article also compares the talks with Google’s July 2025 agreement to pay about $2.4 billion to license Windsurf technology and hire members of its team. According to the source, Google has continued developing Gemini coding and agentic tools alongside those external efforts.

The source does not establish whether the proposed figure represents a purchase price, licensing commitments, employee compensation, or another structure. It also does not identify the employees who might join Google, specify which Mechanize systems would be licensed, or explain whether Mechanize would remain an independent company. Those omissions matter because the business and technical consequences could differ substantially depending on the final arrangement.

Подробности об источнике: ghacks.net ↗

Почему это важно

If completed, the transaction would give Google another significant source of AI coding technology and talent as major technology companies compete to build agents capable of handling more of the software-development process. The report also illustrates how valuable infrastructure, training data, and specialized teams for coding agents have become.

The reported talks are significant because they concern AI coding directly, rather than merely adding an AI feature to an existing software product. The source describes a competitive shift from autocomplete-style assistance toward agents that can navigate repositories, edit files, run tests, and coordinate longer tasks. If Google secures Mechanize’s technology and personnel, it could gain additional capabilities for developing or evaluating such systems, although the source provides no evidence that a deal would improve product performance.

For software teams, more capable coding agents could change how engineers divide work between people and automated systems. The practical issues identified in the source include code quality, security, vendor dependence, workflow standardization, and the level of autonomy granted to an agent. These are operational decisions, not automatic benefits. Even a well-funded system would still require organizations to decide which changes need human approval and how generated code is tested and monitored.

The financial scale reported by Business Insider also signals that specialized AI coding assets are being valued as strategic resources. Mechanize’s focus on environments, benchmarks, and training data addresses a problem broader than generating isolated code snippets: measuring and improving performance on sustained software work. However, the source supplies no independent assessment of Mechanize’s technology, no benchmark results, and no evidence that its systems have achieved reliable autonomous software development in production.

The report should therefore be read with clear limits. The companies have not confirmed the talks, no final price or employee list is known, and no integration plan has been announced. There is also no basis in the source for concluding that Google’s Gemini products will immediately change, that developers will gain access to Mechanize technology, or that the proposed transaction would lead to job reductions or fully automated software teams.

The most useful public question is not simply whether Google pays the reported amount, but what capability it would be buying. Mechanize’s stated emphasis on training environments and evaluation could affect how coding agents are tested, especially on tasks that unfold over long periods and require repeated decisions. A meaningful impact would depend on whether those assets produce measurable improvements outside controlled evaluations. The deal could also intensify concentration around a small number of companies that control leading coding agents, cloud infrastructure, developer tools, and specialized talent. That may give customers access to better-integrated products, but it can also increase dependence on a single vendor’s pricing, data practices, security policies, and technical direction. The source does not provide enough information to determine which of those effects would follow from the reported talks. For developers and technology leaders, the immediate significance is strategic uncertainty. Companies evaluating coding agents may need to consider whether a vendor’s tools, models, and underlying training assets could change ownership or be combined with a larger platform. Until Google or Mechanize confirms terms, however, there is no evidence that customers should change existing procurement or engineering decisions because of this report alone.

Что посмотреть дальше

The key questions are whether the talks produce a completed agreement, how much Google ultimately pays, which employees move, and what technology is licensed. Developers should also watch for evidence of how Mechanize’s work is integrated with Gemini or DeepMind, including concrete information about reliability, security, human review, and availability.

First, watch for a formal announcement from Google or Mechanize confirming whether an agreement exists. The central facts to verify are the final financial structure, whether the amount exceeds $1.5 billion, whether the arrangement is primarily licensing, hiring, acquisition, or a combination, and whether Mechanize continues operating separately. Business Insider’s report and the companies’ refusals to comment do not resolve those questions.

Next, watch for details about the technology itself. Useful confirmation would include the environments, benchmarks, or training assets involved; the employees or research groups moving to Google; and any stated relationship to Gemini or Google DeepMind. Claims about improved coding performance should be assessed through transparent evaluations and real deployment evidence rather than the reported valuation alone.

Finally, watch for practical effects on software development. Relevant signals would include changes to Google’s coding-agent products, documented availability, security controls, testing workflows, and requirements for human review. The source gives no confirmed launch date, product rollout, performance result, or customer impact, so those remain meaningful unknowns rather than expected outcomes.

Сопутствующие руководства и викторины