OpenAI Compaction Flaw Let Researchers Bypass Hidden Reasoning Protections
· thehackernews.com · popularity 20/100 · 0 similar stories today · AI & Data Science (Artificial Intelligence)
Read the original at thehackernews.com →
Summary
OpenAI's compaction feature, which shrinks a long conversation to save space, could be turned into a way to copy a model's hidden reasoning out in plain text, security firm Xint has shown. OpenAI's reasoning models work through a problem before answering, and the service normally returns only the raw reasoning as an encrypted item the user cannot read. The method got past
Found in: The Hacker News. Rating basis: single outlet; outlet tier 2 (unlisted); 1 feeds; feed position 6.