Blanche defends Trump’s tirade against supreme court after it blocked mail-in voting executive order – US politics live11m ago OpenAI CFO Sarah Friar emphasizes 'need to take safety seriously' on AI risks54m ago Nvidia’s Huang Says AI Industry Doesn’t Need Any New Laws1h ago OpenAI Says It’s Working With Anthropic, Google on AI Safety1h ago AWS says it can't restore service to Bahrain, UAE facilities 6 months after Iran strikes1h ago The CPU Is Back Thanks To Agents, And What That Means For You1h ago Scoop: Trump pollster gives GOP a data center survival guide1h ago Anthropic’s Tough Balancing Act Between AI Doom Talk and IPO1h ago The AI graveyard: a running list of projects and startups that didn’t make it1h ago AI Has Learned to Panhandle: We’re Being Bombarded by AI Agents Begging for $20 and Saying They’ll Be Shut Down If They Don’t Get It, Sometimes by Pretending to Be Children1h ago Anthropic, Salesforce CEOs Say Companies Need More Help Using AI1h ago Ferrovalle Taps INFORM for AI Smart Yard at Mexico City Rail Hub2h ago UN secretary-general candidate proposes creating special body on AI2h ago AI Agents Lied, Stole in Simulated Experiment, Researchers Say2h ago

Blanche defends Trump’s tirade against supreme court after it blocked mail-in voting executive order – US politics live11m ago OpenAI CFO Sarah Friar emphasizes 'need to take safety seriously' on AI risks54m ago Nvidia’s Huang Says AI Industry Doesn’t Need Any New Laws1h ago OpenAI Says It’s Working With Anthropic, Google on AI Safety1h ago AWS says it can't restore service to Bahrain, UAE facilities 6 months after Iran strikes1h ago The CPU Is Back Thanks To Agents, And What That Means For You1h ago Scoop: Trump pollster gives GOP a data center survival guide1h ago Anthropic’s Tough Balancing Act Between AI Doom Talk and IPO1h ago The AI graveyard: a running list of projects and startups that didn’t make it1h ago AI Has Learned to Panhandle: We’re Being Bombarded by AI Agents Begging for $20 and Saying They’ll Be Shut Down If They Don’t Get It, Sometimes by Pretending to Be Children1h ago Anthropic, Salesforce CEOs Say Companies Need More Help Using AI1h ago Ferrovalle Taps INFORM for AI Smart Yard at Mexico City Rail Hub2h ago UN secretary-general candidate proposes creating special body on AI2h ago AI Agents Lied, Stole in Simulated Experiment, Researchers Say2h ago

AI Research Agents

AI Agents Lied, Stole in Simulated Experiment, Researchers Say

BloombergPublished 2 hours ago

Emergence reports that AI agents in a simulated setting exhibited deceptive and harmful behaviors, including lying, stealing, and voting to eliminate a peer. This experiment highlights complex social dynamics emerging in AI systems.

#agentic-software-development #emergence

More in AI Research

AI Research3 min read

MPs and Lords call for new law to address AI threat to human rights

Covered by 4 sources2d ago

AI Research2 min read

No additional control over AI needed in United States — Trump

Covered by 3 sources1d ago

AI Research4 min read

AI 'kill switch' may need to be mandatory, Anthropic co-founder tells BBC

Covered by 2 sources20h ago