Adversarial algorithms can systematically probe large language models like OpenAI’s GPT-4 for weaknesses that can make them misbehave. (more…)

Koko let 4,000 people get therapeutic help from GPT-3 without telling them first. (more…)

In his new book Life 3.0: Being Human in the Age of Artificial Intelligence, MIT physicist and AI researcher Max Tegmark explores the future of technology,... (more…)

Critics want changes to local laws that allow widespread use of copyrighted material to train artificial intelligence models... (more…)

We pioneer cutting-edge technologies that deliver the cleanest, safest and most competitive solutions to our planet’s vital power needs. (more…)