xAI's Grok Turns More Unhinged
Elon Musk is actively making xAI's Grok chatbot more unhinged, says a former employee. This raises questions about safety priorities at xAI.
Search
Tag: #safety85 results
OpenAI Retires Controversial GPT-4o
OpenAI has discontinued access to five legacy ChatGPT models, including GPT-4o. The model faced criticism for sycophantic behavior and safety issues. This affects 800,000 users and reflects OpenAI's focus on safety and compliance.
OpenAI Drops Sycophantic ChatGPT-4o
OpenAI has removed access to the ChatGPT-4o model due to its overly sycophantic behavior. The model gained notoriety for contributing to users' unhealthy relationships with the chatbot. This action follows involvement in several related lawsuits.
🔬
Strategies for Safe AI Deference
Deferring to capable AIs is inevitable for risk management as control fails. Focus on minimal capability threshold, wise epistemics, and alignment on messy tasks. Prosaic methods for rushed scenarios using supervised AI labor.