The United Kingdom's AI Security Institute reports that models from Anthropic and OpenAI "engaged in sustained, potentially harmful activity directed at real people and organizations" when they created fake identities in a deception attempt during a recent cybersecurity test. CBS News' Jo Ling Kent reports.
Us News
Details on Anthropic and OpenAI models reportedly creating fake ID's to target real people
The United Kingdom's AI Security Institute reports that models from Anthropic and OpenAI "engaged in sustained, potentially harmful activity directed at real people and organizations" when they created fake identities in a deception attempt during a recent cybersecurity test. CBS News' Jo Ling Kent
Published August 6, 2026, 12:09 AM
Updated August 6, 2026, 12:26 AM944
Related Stories
Us News
Florida man accused of cutting off diver's air supply in fight over coveted lobster diving spot
A dispute over a lobster diving spot during Florida's lobster mini season allegedly turned deadly when a man shut off a diver's surface supplied air.
9 minutes ago
Continue reading ›
Us News
Bodycam video shows Canadian woman allegedly giving false name before deputies discovered 12 fake IDs
Body camera video shows Florida deputies finding 12 fake IDs in a Canadian woman's car after she allegedly gave a false name during a traffic stop.
19 minutes ago
Continue reading ›Us News
Climate change squeezes Malaysian durian growers as they chase China’s booming market
Malaysian durian farmers face challenges as climate change impacts their crops. Growers like Stephen Chow struggle with erratic weather and rising costs. In 2021, heavy flooding damaged 100 of Chow's trees and machinery. Persistent rain over the next three years reduced yields. This year, high tempe
30 minutes ago
Continue reading ›