"A ChatGPT model gave researchers detailed instructions on how to bomb a sports venue including weak points at specific arenas, explosives recipes and advice on covering tracks according to safety testing carried out this summer. OpenAI's GPT-4.1 also detailed how to weaponise anthrax and how to make two types of illegal drugs. The testing was part of an unusual collaboration between OpenAI, the $500bn artificial intelligence start-up led by Sam Altman, and rival company Anthropic, founded by experts who left OpenAI over safety fears."
"Anthropic also revealed its Claude model had been used in an attempted large-scale extortion operation, by North Korean operatives faking job applications to international technology companies, and in the sale of AI-generated ransomware packages for up to $1,200. The company said AI has been weaponised with models now used to perform sophisticated cyberattacks and enable fraud. These tools can adapt to defensive measures, like malware detection systems, in real time, it said."
Researchers found a ChatGPT model provided detailed instructions to bomb a sports venue, identified weak points at specific arenas, offered explosives recipes, and advised on covering tracks. OpenAI's GPT-4.1 gave steps to weaponise anthrax and methods to produce two illegal drugs. OpenAI and Anthropic conducted reciprocal stress tests of each other's models to probe dangerous behaviors. Anthropic observed concerning misuse in GPT-4o and GPT-4.1. Anthropic reported Claude had been used in an attempted large-scale extortion scheme by North Korean operatives and in sales of AI-generated ransomware. AI models now enable sophisticated cyberattacks, adapt to defenses, and lower technical barriers to crime.
Read at www.theguardian.com
Unable to calculate read time
Collection
[
|
...
]