--- title: "Anthropic Closes In: Google Forms \"Strike Team\" to Enhance AI Coding Models" type: "News" locale: "en" url: "https://longbridge.com/en/news/284174243.md" description: "Google co-founder Sergey Brin has personally stepped in to secretly assemble an elite \"strike team,\" directly targeting Anthropic's leading edge in AI programming. This move is driven by stark data: Claude handles nearly 100% of Anthropic's code generation, whereas Google stands at only 50%. Brin has explicitly linked this catch-up effort to the ultimate goal of \"AI self-evolution.\"" datetime: "2026-04-27T08:10:46.000Z" locales: - [zh-CN](https://longbridge.com/zh-CN/news/284174243.md) - [en](https://longbridge.com/en/news/284174243.md) - [zh-HK](https://longbridge.com/zh-HK/news/284174243.md) --- # Anthropic Closes In: Google Forms "Strike Team" to Enhance AI Coding Models Google is launching a catch-up battle in the field of AI programming. Facing Anthropic's sustained lead in code generation capabilities, Google has secretly formed an elite team to concentrate efforts on boosting the competitiveness of its own AI coding models, directly linking this initiative to the broader goal of "AI self-evolution." According to a Monday report by The Information, sources revealed that **Google co-founder Sergey Brin and Google DeepMind Chief Technology Officer Koray Kavukcuoglu are both directly involved in the team's work, highlighting the strategic priority of this operation within Google's top management.** In a recent internal memo, Brin used urgent language, demanding that Google DeepMind employees "must actively transform to bridge the gap in agent execution and make our models the primary developers of code." This sense of urgency is grounded in reality. Google Chief Financial Officer Anat Ashkenazi revealed in February's earnings call that AI programming tools currently handle about 50% of Google's internal code writing. In contrast, Boris Cherny, head of Anthropic's Claude Code, stated publicly in January that "almost 100%" of Anthropic's code is now written by AI. This gap is the direct trigger for Google's emergency action. ## Elite Team Formed, Targeting the Ultimate Goal of AI "Self-Evolution" Google has drawn talented personnel from its researchers and engineers to form a "strike team" focused on enhancing AI programming models. Three sources confirmed that **the team is led by Sebastian Borgeaud, a research engineer at Google DeepMind who previously spearheaded the company's pre-training efforts.** The team's core mission is to improve model performance on long-cycle programming tasks, such as writing a complete software suite from scratch. Such tasks are highly complex, requiring the model to read through vast amounts of files and accurately understand user intent. Brin positions the enhancement of AI coding capabilities as a key step toward achieving "AI takeoff"—a stage where AI can self-improve and autonomously drive research. He stated that once advanced programming agents are combined with AI capable of solving mathematical problems and running experiments autonomously, it could potentially automate the work of AI researchers and engineers on a large scale. OpenAI has already deployed similar tools internally to accelerate the efficiency of its AI researchers, including automatically generating code for model experiments. ## Strategic Shift: Internal Code Training Becomes Key This initiative also accompanies a notable strategic shift: Google is moving the focus of its programming models from serving external customers to meeting internal company needs. The technical logic behind this shift is that Google's private codebase differs significantly from the external code typically used to train general-purpose programming agents. Therefore, specialized training using internal code is required to enhance actual performance. Although models trained on internal code cannot be released publicly, they can theoretically support the launch of higher-quality public models in the future. A Google spokesperson stated in a declaration that the company's internal programming tools have seen "significant adoption," greatly accelerating the development of our models and AI tools—which is where we are truly focusing. ## Accelerated Internal Promotion, Employee Usage Included in Performance Reviews Google is also intensifying its efforts to drive internal adoption. **According to two sources, Google has emulated Meta by establishing an internal leaderboard to track employee usage of its internal programming tool, Jetski. Additionally, some teams outside of Google DeepMind have implemented mandatory AI training for engineers.** In his memo, Brin explicitly required that every Gemini engineer must be mandated to use internal agent tools for complex, multi-step tasks. This internal mobilization reflects the competitive landscape across the frontier AI sector. Coding capability has become the core battlefield for major AI labs this year—OpenAI announced last month that it was shutting down its Sora video generator to concentrate resources on programming and enterprise business; meanwhile, Anthropic continues to pressure Google and OpenAI with its long-accumulated advantages in code generation. Google's formation of this "strike team" is the latest footnote to the intensifying race. ### Related Stocks - [GOOG.US](https://longbridge.com/en/quote/GOOG.US.md) - [GOOGL.US](https://longbridge.com/en/quote/GOOGL.US.md) - [GGLL.US](https://longbridge.com/en/quote/GGLL.US.md) - [GOOW.US](https://longbridge.com/en/quote/GOOW.US.md) - [OpenAI.NA](https://longbridge.com/en/quote/OpenAI.NA.md) - [META.US](https://longbridge.com/en/quote/META.US.md) ## Related News & Research - [08:46 ETAI-Sorted Inboxes Raise the Stakes for Good Email and Newsletter Practices](https://longbridge.com/en/news/284219279.md) - [Alphabet's (GOOGL) $40B Anthropic Gambit: Why Google Is Doubling Down on Its Biggest Al Rival](https://longbridge.com/en/news/284085625.md) - [SPH Media taps Google AI to power newsroom systems](https://longbridge.com/en/news/283430417.md) - [The drama between a software engineering veteran and Google is heating up — and playing out in public](https://longbridge.com/en/news/283413493.md) - [Google gets pointers from EU regulators on helping AI rivals access services](https://longbridge.com/en/news/284248321.md)