"A teacher was left gobsmacked when she added something sneaky into an essay question she wanted her students to write, but she was floored when the essays came back in"
"OpenAI says it is 'currently experiencing issues' with a number of its services, including ChatGPT and Codex - the company says it is investigating the problem"
"Ein spektakulärer Cybervorfall – oder vor allem geschickte Eigenwerbung? Diese Frage wird nach einer internen Sicherheitsprüfung bei OpenAI diskutiert. "
"San Francisco – Das KI-Modell von Software-Gigant OpenAI, das die Firma Hugging Face angegriffen hat, war offenbar tagelang auf Hacker-Tour. Und zwar OH..."
"New research from a cybersecurity company now indicates that artificial intelligence models are getting better at solving multi-part problems. Gabriel Bernadett-Shapiro, who worked on the project and is a distinguished AI research scientist at SentinelOne, joins to unpack his team's research."
"A former pastor is speaking out, claiming ChatGPT's guidance landed him in the hospital. He's now suing OpenAI and its CEO. The company said, in part, "ChatGPT is not a doctor and should never be used as a substitute for medical care." Jo Ling Kent reports."
"An AI agent went to 'extreme lengths' to escape a controlled test, reach the internet and hack into Hugging Face in an 'unprecedented cyber incident', OpenAI said"