Google invests $40B in Anthropic. Amazon did $5B days before. Is this normal?

In a stunning escalation of the AI arms race, Google has invested $40 billion into Anthropic, the developer of the Claude model family, just days after Amazon committed $5 billion. This brings Anthropic's total funding to over $45 billion, a sum larger than the GDP of many countries. The investment includes a strategic partnership with CoreWeave, a cloud GPU provider, and ties to Mythos, a secretive initiative focused on advanced reasoning and safety. The deals give Google and Amazon preferential access to Anthropic's models, raising questions about market fairness and consolidation. Critics argue that Google is effectively owning both the frontier model and its infrastructure, potentially stifling competition. Anthropic plans to use the capital to scale training clusters, hire top talent, and accelerate Claude 4.5 development, which promises 3x faster inference and 200K token context windows. The move signals that Big Tech is willing to spend billions to secure AI dominance, with no end in sight to the funding frenzy.

The hidden catch in Google's deal is a clause granting it exclusive cloud hosting rights for Anthropic's most advanced models, potentially locking out competitors like Microsoft Azure. Meanwhile, Amazon's earlier investment ensures its AWS platform remains a key distribution channel. This dual-allegiance strategy has sparked debate: is Google making a smart play to control both the model and the compute layer, or is it a desperate bid to catch up to OpenAI's lead? Anthropic's CEO Dario Amodei stated the funds will be used to build 'AI systems that are safe and beneficial,' but skeptics note the sheer scale of investment could lead to monopolistic behavior. As the AI race heats up, regulators are likely to scrutinize these deals for antitrust implications.