The new benchmark, called Elephant, makes it easier to spot when AI models are being overly sycophantic—but there’s no ...

Separating AI reality from hyped-up fiction isn’t always easy. That’s why we’ve created the AI Hype Index—a simple, ...

A pro-AI Reddit forum is cracking down on users exhibiting chatbot-induced delusions, banning over 100 users in a few weeks.

ChatGPT is revolutionizing the way we work, create, and use the internet, but it also has creeping effects on how we interact ...

As AI chatbots grow into large-scale businesses, companies may use engagement optimization techniques even at the expense of ...

One of the godfathers of AI is creating a new AI safety company called LawZero to make sure that other AI models don't go ...

ver dinner in Brooklyn a few weeks ago a writer friend told me he had asked ChatGPT to critique his manuscript. Had I tried it? No, I said, I’m not interested in what it would say about my work, or ...

New research finds that AI chatbots like ChatGPT and therapy-specific bots can pose risks when relied upon as AI therapists.

We're only beginning to understand the effects of spending more and more time talking to AI chatbots on a daily basis. As the ...

ChatGPT is more than just a chatbot—it ... Recently, OpenAI had to roll back an update to GPT-4o that made the model too “sycophantic,” emphasizing the point that these models are constantly ...

A new study from consultancy Accenture reveals some insights into how consumers perceive and use AI, and it's seen as a lot ...