Tag: Alexander Meinke et al.

A phone screen shows a scary AI with glowing red eyes. The AI looks like a cyborg, and its eyes are full of malicious intent. This image the danger of bad AI, like a virus taking over your phone.

^{Type
post

Denyse O'Leary
Date
July 28, 2025

Categorized
Artificial general intelligence (AGI), Large Language Models (LLMs)

Tagged
Alexander Meinke et al., Chatbots as scheming, ChatGPT, Geoffrey Hinton on LLMs}

Can Chatbots Really Scheme Against Us? Some Researchers Say Yes

_{When an industry pro tested ChatGPT, it seemed like the system has faced this question a few times before and has been tweaked and edited well} _{Denyse O'Leary
July 28, 2025

5

Artificial general intelligence (AGI), Large Language Models (LLMs)}

So CAN a chatbot really scheme? No, but the problem posed is a familiar one: We sometimes see what we need to believe. Read More ›