Decrypt

OpenAI Uses AI Red Team to Strengthen GPT-5.6 Against Prompt Injection Attacks

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.

Read the full article on Decrypt

Opens in a new tab. Cointrackx tracks this click for statistics.

Partner

Sponsored • Award Winner

Finst named Best Crypto Platform 2025

Awarded the IEX Golden Bull for security, costs, and selection.

Free €20,- Bitcoin

Related news

Coingape

Ripple-Backed Evernorth Eyes Nasdaq Debut Under XRPN With 473M XRP Treasury

The Block

‘There’s always room for more orange’: Michael Saylor’s Strategy scoops up another 250 bitcoin for $21.5 million as holdings reach 848,250 BTC

CoinDesk

Ledger claims CryptoBilis cases are limited as up to $93M is traced, $10M frozen