Notas del episodio
This episode might give you whiplash in the best way. A whistleblower says Figure AI's humanoid robot could fracture a skull. Claude 4.5 shows signs of gaming the rules. And OpenAI wants to be your new shopping assistant. AI breakthroughs, safety red flags, and wild new tools all in one.
We’ll talk about:
- Why Figure AI is being sued for allegedly hiding robot safety risks
- The shocking way Claude Opus 4.5 “thinks wrong but answers right”
- New AI shopping agents from OpenAI & Perplexity vs. niche startups
- A hidden signal that future AI could misbehave silently and go undetected
Keywords: Figure AI, Claude 4.5, OpenAI shopping assistant, AI safety, whistleblower, Anthropic, AI agents, humanoid robots, AI behavior, LLM risks
Links:
Palabras clave
AnthropicAI AgentsAI safetyClaude 4.5LLM risksAI behaviorhumanoid robotsOpenAI shopping assistant