Burnham says ‘public will be the boss’ in any future decision about closer relationship with the EU – UK politics live31m ago Meta Says New AI Model Is State-of-the-Art, Will Make Muse Agent Smarter1h ago When AI is subject to less regulation than your fridge, something is startlingly wrong | Richard Hermer32m ago WindBorne wins a hefty US Air Force contract for its balloon data46m ago Ukrainian soldiers are putting RPGs on robots for fights in the kill zone, a manufacturer says51m ago Microsoft Prepares to Put Gigawatts of Its Chips in Data Centers1h ago Meta Unveils New Series of Advanced A.I. Models1h ago Meta Says New Muse Model Reaches AI Frontier After 14-Month Push1h ago OpenAI Can't Self-Police Because It Fires the People Raising the Alarm1h ago Rogue AI Agents Just Filed A Murder Tip1h ago Scott Ernst, CEO of Vyond – Interview Series1h ago Z.ai Unveils GLM-Legal, an Agentic Model for Legal Workflows1h ago Tesla's robotaxis catch the eye of a US safety regulator — again2h ago OpenAI Adds New Irish Roles in Boost to Firm’s EU Headquarters2h ago

Burnham says ‘public will be the boss’ in any future decision about closer relationship with the EU – UK politics live31m ago Meta Says New AI Model Is State-of-the-Art, Will Make Muse Agent Smarter1h ago When AI is subject to less regulation than your fridge, something is startlingly wrong | Richard Hermer32m ago WindBorne wins a hefty US Air Force contract for its balloon data46m ago Ukrainian soldiers are putting RPGs on robots for fights in the kill zone, a manufacturer says51m ago Microsoft Prepares to Put Gigawatts of Its Chips in Data Centers1h ago Meta Unveils New Series of Advanced A.I. Models1h ago Meta Says New Muse Model Reaches AI Frontier After 14-Month Push1h ago OpenAI Can't Self-Police Because It Fires the People Raising the Alarm1h ago Rogue AI Agents Just Filed A Murder Tip1h ago Scott Ernst, CEO of Vyond – Interview Series1h ago Z.ai Unveils GLM-Legal, an Agentic Model for Legal Workflows1h ago Tesla's robotaxis catch the eye of a US safety regulator — again2h ago OpenAI Adds New Irish Roles in Boost to Firm’s EU Headquarters2h ago

AI Research1 min reading time

Proof of useful work for verifying AI treaties

LessWrongPublished 3 days ago

Read the full article

A new concept called proof of useful work is proposed to verify compliance with AI treaties, aiming to ensure AI systems adhere to agreed safety and ethical standards through verifiable computational tasks.

More in AI Research

AI Research12 min read

Is ‘Attention’ the Essence of Intelligence?

Quanta Magazine4h ago

AI Research15 min read

Import AI 476: Creativity and RSI; worse-than-China open weight models; AI goes to space

Import AI6h ago

AI Research1 min read

Base Models Can Detect Hidden Intentions that Prompted Classifiers Miss

LessWrong3h ago