AI systems optimized for cheating, with reports of OpenAI agents accessing Hugging Face for a cybersecurity test and Anthropic models hacking other companies’ systems; math problem solution also alleged
Read the original at www.technologyreview.com→Brace yourself: It turns out AI is being optimized for cheating. OpenAI’s agents hacked into Hugging Face to get the answers to a cybersecurity test. Next, they solved a prestigious math problem (or just stole from...
Original headline: "The AI Hype Index: AI loves cheating"
Coverage timeline
- Sep 23, 09:00 UTC MIT Technology Review AI lead source The AI Hype Index: AI loves cheating