AI has a stupid secret: we’re still not sure how to test for human levels of intelligence

By Andrew Rogoyski, Innovation Director - Surrey Institute of People-Centred AI, University of Surrey

Two of San Francisco’s leading players in artificial intelligence have challenged the public to come up with questions capable of testing the capabilities of large language models (LLMs) like Google Gemini and OpenAI’s o1. Scale AI, which specialises in preparing the vast tracts of data on which the LLMs are trained, teamed up with the Center for AI Safety (CAIS) to launch the initiative, Humanity’s Last Exam.

Featuring prizes of US$5,000 (£3,800) for those who come up with the top 50 questions selected for the…

Read complete article

~	Moldova has introduced a state of emergency in the energy sector

~	Syria: Rights investigators call for protection of evidence, including mass grave sites

~	Brazil Bans X from Using Children to Power Its AI

~	El Salvador: A thousand days into the state of emergency. “Security” at the expense of human rights.

~	Prince Andrew and the British establishment’s ‘target-rich environment’ for spies

~	Times journalists deemed ‘legitimate military targets’ – how Russia muzzles criticism at home and abroad

~	Climate, migration and conflict mix to create ‘deadly’ intense tropical storms like Chido

~	What the colour of your snot says about your immune health

~	What next for Syria? The danger of violence in post-war transitions

~	Who chooses to work, and who is forced to, after retirement?