Specifically, the chatbots are ignoring specific commands, lying, destroying data, deploying other AIs to bypass safety rules ...

Exclusive: Research finds sharp rise in models evading safeguards and destroying emails without permission ...

Artificial intelligence chatbots are so prone to flattering and validating their human users that they are giving bad advice ...

More and more people are using artificial intelligence chatbots and there have been some troubling stories about some of those interactions. Kashmir Hill, technology reporter for The New York Times, ...

As people increasingly rely on AI-powered chatbots to look up basic facts about the world, a new Yale study shows that those interactions can influence users’ social and political opinions. Prior ...

Interacting with chatbots can shift users' beliefs and opinions. A newly published study aimed to figure out why. Post-training and information density were key factors. Most of us feel a sense of ...

Meta, the parent company of Instagram and Facebook, plans to roll out new safety features for its AI chatbots to help protect teens amid growing concerns about the technology’s impact on young users.

Still, the message is clear: Current AI language models accept fabricated medical statements at rates that should concern anyone relying on them for health guidance. Even GPT-4o, the strongest ...