AI safety tests found to rely on 'obvious' trigger words; with easy rephrasing, models labeled 'reasonably safe' suddenly fail, with attacks succeeding up to 98% of the time. New corporate research ...

The Department of Bioengineering at the Indian Institute of Science has invited organisations to host its MTech ...

Sasha Stiles turned GPT-2 experiments into a self-writing poem at a Museum of Modern Art installation—and a new way to think about text-generating AI optimization ...

Art made with AI is selling for over $1 million and being embraced by some of the world's most prestigious museums, but ...