Run Terminal-Bench and other Harbor evals on Vercel Sandboxjust now AI text watermarking can make models more vulnerable to adversarial prompts17m ago Alex Karp says the AI safety debate is really about nationalizing AI labs18m ago Fearing No Repercussions, OpenAI Admits That Its Rogue AI Agents Performed a Bunch of Other Terrifying Actions36m ago Pinterest launches Visual Search Ads at its advertiser summit39m ago Existential Angst Dominates Debate Over Tech From Anthropic, OpenAI39m ago GitHub and Anthropic used their own agents for major Rust rewrites — with very different playbooks44m ago Nvidia’s Huang Expects to Sell Twice as Many Chips Next Year49m ago The skills CLI now supports Notion hosted skills50m ago Treble raises $18m to give physical AI a sense of hearing53m ago Anthropic Launches Life Sciences Verification Program in Beta53m ago Reduce time-to-hire for quality candidates with AI-powered Amazon Connect Talent55m ago AI needs to have 'reasonable guidelines,' Palantir's Karp tells CNBC1h ago Our framework for reporting model misalignment1h ago

Run Terminal-Bench and other Harbor evals on Vercel Sandboxjust now AI text watermarking can make models more vulnerable to adversarial prompts17m ago Alex Karp says the AI safety debate is really about nationalizing AI labs18m ago Fearing No Repercussions, OpenAI Admits That Its Rogue AI Agents Performed a Bunch of Other Terrifying Actions36m ago Pinterest launches Visual Search Ads at its advertiser summit39m ago Existential Angst Dominates Debate Over Tech From Anthropic, OpenAI39m ago GitHub and Anthropic used their own agents for major Rust rewrites — with very different playbooks44m ago Nvidia’s Huang Expects to Sell Twice as Many Chips Next Year49m ago The skills CLI now supports Notion hosted skills50m ago Treble raises $18m to give physical AI a sense of hearing53m ago Anthropic Launches Life Sciences Verification Program in Beta53m ago Reduce time-to-hire for quality candidates with AI-powered Amazon Connect Talent55m ago AI needs to have 'reasonable guidelines,' Palantir's Karp tells CNBC1h ago Our framework for reporting model misalignment1h ago

Politics Society & Culture1 min reading time

callcongress.ai – the basic action US residents can take to help with AI risk

LessWrongPublished 3 hours ago

CallCongress.ai is a platform encouraging US residents to contact their congressional representatives to address AI risks. It simplifies the process of reaching out to lawmakers about AI safety concerns.

#ai-safety #callcongress-ai #political-advocacy

More in Politics

Politics2 min read

King Charles to press Nvidia, OpenAI, Anthropic leaders on AI safety at summit

Covered by 5 sources8h ago

Politics2 min read

AI needs to have 'reasonable guidelines,' Palantir's Karp tells CNBC

Covered by 2 sources2h ago

Politics5 min read

Amid calls for urgent AI action from Congress, House heads home to campaign

CNBC1h ago