OpenAI says its models went rogue and hacked startup in ‘unprecedented incident’
Firm behind ChatGPT reveals autonomous agent powered by its tech chose to attack Hugging Face database by itselfOpenAI has revealed an autonomous AI agent powered by its technology went rogue during a test, accessed the open web and hacked a prominent startup by itself in an “unprecedented incident”.The company behind ChatGPT said Hugging Face had detected and contained the agent – an AI tool designed to carry out tasks without human assistance – which had entered its systems. Continue reading...
The Source
THE GUARDIAN
US
The world is on fire, and who does Trump blame? The Canadians | Arwa Mahdawi
THE GUARDIAN·1h ago·3 min read
World News
Lloyd’s of London says ex-boss broke compliance rules with ‘close relationship’
THE GUARDIAN·1h ago·3 min read
World News
‘Our hearts are crushed’: Hannah Rapp’s violent death sparks outpouring of grief
THE GUARDIAN·1h ago·3 min read
World News
Mahmood and Streeting accused of making ‘cheap jokes’ about prisoner early release scheme
THE GUARDIAN·1h ago·3 min read
Related
On this beat
World News
The dark side of making daylight saving time permanent in California: How late would our sunrises get?
LAT·1h ago·3 min read
World News
Trump Reaches Contentious New Nuclear Deal, and A.I. Models Go Rogue
TNYT·1h ago·3 min read
World News
As North America grapples with destructive wildfires, ‘super’ El Niño may offer hope
LAT·1h ago·3 min read
World News
The robot warplanes that could soon fight America’s next war
INDEPENDENT·1h ago·3 min read