Google engineers can switch to Claude for tough coding tasks

Google has begun allowing all of its engineers to use Anthropic’s Claude, marking a major shift in its internal AI policy. The company historically restricted access to external coding tools, insisting employees rely on Gemini, but growing frustration over Gemini’s coding limitations pushed leadership to reconsider. Claude’s Opus 5 model is now available inside Google’s internal development environment Antigravity, though with per‑user quotas and only as a supplementary option. The move reflects the intensifying competition in AI coding, where Anthropic and OpenAI have been outperforming Google’s own models. Some teams, especially within DeepMind, had early access, but broader availability was long requested by engineers. Google’s decision also aligns with its massive investment plans of up to $40 billion in Anthropic. Other tech giants like Amazon have similarly opened access to Claude after employee pressure.

Read the full story on Business Insider →