Elon Musk just admitted that Grok was trained using distillation from other AI models - meaning xAI used outputs from competitors like GPT and Claude to teach its own model. This is one of the most contested practices in AI right now. OpenAI has accused others of it. Anthropic has flagged it. The AI Security Institute just found GPT-5.5 matching Claude Mythos in cyberattack capabilities - and distillation is likely part of why capabilities are converging so fast. For crypto, this matters because AI x blockchain is a live narrative. Projects building on-chain AI agents, decentralized inference, and tokenized compute are betting that AI model development will eventually need censorship-resistant infrastructure. If the biggest AI labs are secretly training on each other's outputs, the race is tighter and dirtier than anyone admitted. That makes the case for open, verifiable AI training - exactly what crypto rails enable - a lot stronger. If Musk admits it, how many others are doing it without admitting it? And does this make you more bullish on decentralized AI infrastructure as the only way to have verifiable model provenance? #MuskAdmitsDistillation