At the All-In Summit 2026, Nvidia CEO Jensen Huang dismissed "AI doom" predictions as unscientific, while acknowledging that safety discussions in Anthropic CEO Dario Amodei's post deserve attention. He explained that techniques like LoRA make recursive self-improvement controllable and said the term "RSI" is being weaponized. He also revealed Nvidia has frontier open-source models in five verticals including autonomous driving and biopharma, and partners with Blackstone and Goldman Sachs on the Cloverleaf project to finance AI infrastructure.