Google is in talks to pour more money into Anthropic, which could push the AI startup's value to $350 billion (www.businessinsider.com)

Google is in early talks to deepen its investment in Anthropic, potentially via fresh funding (convertible note or priced round) or expanded strategic cloud support, according to people familiar with the discussions. The deal is still fluid; one source said it could value Anthropic at more than $350 billion. Google already owns roughly 14% of Anthropic after a previous $3 billion investment, and in October the companies struck a major cloud deal giving Anthropic access to up to one million Google TPUs. Anthropic raised $13 billion in September at a $138 billion valuation; OpenAI recently reached a $500 billion secondary valuation. The talks matter because they sharpen a broader “compute and partnership” split among Big Tech in the multitrillion-dollar race to dominate foundation models. Anthropic is backed by Google and Amazon (which provides Trainium2-based Project Rainer capacity), while OpenAI has deep ties to Microsoft and NVIDIA. A larger Google stake or expanded TPU access would reinforce vendor-specific stacks, preferential compute pipelines, and potential lock‑in for training and inference at extreme scale. Technically, the deal could shift how Claude-family models are trained and deployed—more TPU capacity, pricing and architecture optimization—while intensifying competition over chip resources (TPUs vs Trainium vs NVIDIA GPUs) and strategic influence over model roadmaps.