A researcher developed a method for bypassing GPT-6.2 Astra’s safeguards before its official release. In single-turn cyber tests, the method succeeded in half the queries needed to break through GPT-6 Astra. OpenAI subsequently changed its security safeguards. However, as of their account on November 17, UK testers had not yet independently evaluated the latest changes.
Source: Read the original article
Disclaimer: this content is for information purposes only and is not financial advice. Cryptocurrencies are highly volatile: you may lose all of your capital. Always do your own research. Legal notice


