FAA tees up $875M AI tool to help manage air traffic congestion19m ago Quoting Thariq Shihipar30m ago MariaDB 13 Expands Oracle Compatibility and Improves Developer Experience and Observability39m ago This tiny cybersecurity startup managed to hack OpenAI using Claude, and won a $6,500 bounty42m ago Spanberger Signs Data Center Accountability Order, Creates AI Task Force48m ago A new kind of AI model from a ChatGPT inventor is thrilling developers50m ago Why Europe has been absent from the great AI safety debate1h ago Disney Beefs Up Tech Ambitions With Newly Created CTO Role1h ago Newsom Wants AI ‘Kill Switch’ to Rein in Tech Industry1h ago 5 Defining Traits Every Self-Driving Network Needs1h ago The Week’s 10 Biggest Funding Rounds: Large Rounds For AI Infrastructure, Space Tech And Investment Management Lead1h ago Virginia governor creates an AI task force and moves to restrain data centers1h ago PrismML hopes its tiny LLM will change how we all use AI2h ago Meta’s AI Agent Has a Trust Problem2h ago

FAA tees up $875M AI tool to help manage air traffic congestion19m ago Quoting Thariq Shihipar30m ago MariaDB 13 Expands Oracle Compatibility and Improves Developer Experience and Observability39m ago This tiny cybersecurity startup managed to hack OpenAI using Claude, and won a $6,500 bounty42m ago Spanberger Signs Data Center Accountability Order, Creates AI Task Force48m ago A new kind of AI model from a ChatGPT inventor is thrilling developers50m ago Why Europe has been absent from the great AI safety debate1h ago Disney Beefs Up Tech Ambitions With Newly Created CTO Role1h ago Newsom Wants AI ‘Kill Switch’ to Rein in Tech Industry1h ago 5 Defining Traits Every Self-Driving Network Needs1h ago The Week’s 10 Biggest Funding Rounds: Large Rounds For AI Infrastructure, Space Tech And Investment Management Lead1h ago Virginia governor creates an AI task force and moves to restrain data centers1h ago PrismML hopes its tiny LLM will change how we all use AI2h ago Meta’s AI Agent Has a Trust Problem2h ago

AI Research Society & Culture1 min reading time

Persuasion Undermining Control: Can AI Talk its Way Out of Human Control?

LessWrongPublished 3 hours ago

A recent analysis explores whether AI systems could use persuasive language to override human control, raising concerns about AI autonomy and the challenges of maintaining oversight.

More in AI Research

AI Research3 min read

OpenAI reports 6 new instances of 'concerning model behavior' since March

Covered by 12 sources2d ago

Our framework for reporting model misalignment

Covered by 8 sources2d ago

AI Research1 min read

[Paper] Stringological sequence prediction III

Covered by 2 sources3h ago