Decrypt ••1 min read
OpenAI Uses AI Red Team to Strengthen GPT-5.6 Against Prompt Injection Attacks
The Arch TakeBullish
OpenAI said its new automated red-teaming model, GPT-Red, uncovered vulnerabilities that were used to make GPT-5.6 more resistant to prompt injection attacks.
📱
Never miss a Web3 update
Join our Telegram channel to receive news in real-time, straight to your phone.
Related News
Multicoin Capital makes first Hyperliquid ecosystem investment in Trasia, an Asia-focused trading platform
The Block•3h ago
Visa launching internal stablecoin platform for clients that provides access to OUSD: report
The Block•4h ago
The Clarity Act is the most important consumer protection effort in years
CoinDesk•4h ago
US jury indicts South Dakota crypto investor on 29 counts over alleged $20 million investor fraud
The Block•4h ago