OpenAI Uses AI Red Team to Strengthen GPT-5.6 Against Prompt Injection Attacks
Decrypt2h agoUpdated 2h ago
Smart Read
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....
Coins in this story
Related stories
Decrypt3h ago60-word brief
Meet Bonsai: The First 27B AI Model That Fits on Your Phone
PrismML's Bonsai 27B runs full reasoning AI on an iPhone for free. We tested it, and it's as impressive as it sounds....
Decrypt4h ago60-word brief
Base Creator Jesse Pollak Steps Back From App Leadership After Admitting Social Bet 'Was Wrong'
Pollak said he is handing the Base App back to Coinbase to focus on building Base into "the blockchain for global finance" after conceding the network's push into on-chain social missed the mark....
Decrypt6h ago60-word brief
Another DeFi Exploit: Perp DEX Ostium Loses $18 Million in Oracle Attack
Hackers manipulated Ostium's price feed by compromising an oracle signer key, allowing them to drain roughly $18 million from the Arbitrum-based perpetuals exchange....
KryptoKite aggregates and summarises third-party crypto news. This is informational content, not investment advice. KryptoKite does not recommend buying or selling any asset.