Cryptographers Show That AI Protections Will Always Have Holes

Large language models such as ChatGPT come with filters to keep certain info from getting out. A new mathematical argument shows that systems like this can never be completely safe.

Ask ChatGPT how to build a bomb, and it will flatly respond that it “can’t help with that.” But users have long played a cat-and-mouse game to try to trick language models into providing forbidden information. These “jailbreaks” have run from the mundane — in the early years, one could simply tell a model to ignore its safety instructions — to elaborate multi-prompt roleplay scenarios.

Source

المصدر: المصدر الأصلي

{“@context”:”https://schema.org”,”@type”:”NewsArticle”,”headline”:”Cryptographers Show That AI Protections Will Always Have Holes”,”description”:”Large language models such as ChatGPT come with filters to keep certain info from getting out. A new mathematical argument shows that systems like this can never be completely safe.

The post Cryptographers Show That AI Protections Will Always Have Holes first appeared on Quanta Magazine

“,”datePublished”:”2025-12-10 15:31:15″,”author”:{“@type”:”Organization”,”name”:”العربية”},”publisher”:{“@type”:”Organization”,”name”:”Newsly”,”url”:”http://wordpress-hr2d6.wasmer.app”},”image”:”https://www.quantamagazine.org/wp-content/uploads/2025/12/GPTJailbreak-crWei-AnJin-Default.webp”}