security Source: BleepingComputer

📅 28 October 2026 at 04:30 AM UTC

OpenAI's GPT-6.1 Terra is safer overall, but shows more destructive-action flags in tests

Verified Primary Feed Ingestion

↗ Open Source Publication

⚡ Executive Takeaways & Intelligence Summary

OpenAI confirmed GPT-6.1 Terra is safer overall than some of its older models, but its safety testing also found other problems. [...]

Publication Reference: https://www.bleepingcomputer.com/news/artificial-intelligence/openais-gpt-61-terra-is-safer-overall-but-shows-more-destructive-action-flags-in-tests/

← Back to News Wire

📰 Related Surveillance Disclosures

BleepingComputer

Critical Linux kernel flaw allows remote root attacks on MPTCP servers

The Hacker News

Chinese Hackers Use AI-Powered ARTEX to Target Over 200 Organizations Worldwide

BleepingComputer

Trezor warns of suspected crypto-theft implant in hardware wallet