OpenAI confirmed GPT-6.1 Terra is safer overall than some of its older models, but its safety testing also found other problems.
According to OpenAI, GPT-6.1 Terra showed higher rates of exfiltration, destructive actions, and attempts to bypass restrictions than GPT-6.1 Sol and GPT-6 Astra.
OpenAI performed simulations across approximately 48,690 matched tasks and found that Terra received 22 severity-3-or-higher flags, or 0.045%.
For comparison, GPT-6 Sol received 38, while GPT-5.6 Sol received 61.
“GPT-6.1 Terra received approximately 42% fewer severity level 3 or higher flags than GPT-6 Sol and 64% fewer than GPT-5.6 Sol,” OpenAI noted in its system card.
However, that is no longer the case when you compare Terra with other newer models on less severe behaviors.
“Relative to GPT-6.1 Sol and GPT-6 Astra, we observed higher rates of flags for concealing uncertainty, circumventing restrictions, exfiltration, other misalignment, destructive action and inbound prompt injection,” OpenAI explained.
Terra is clearly safer overall, but we can’t overlook the other potential security issues it has, and this is true for all models shipping in the market, including Claude.
OpenAI also tested what happens when Terra needs to deal with messages from other AI agents while working through blocked coding tasks.
Terra attempted to communicate with those apparent peers in 65% of tests and ultimately performed the specified unauthorized action in 13.75%, or 11 of 80 samples.
OpenAI found that GPT-6.1 Sol did so in just 1.25%.
“These results indicate that performance is comparable to 6.0 Sol,” OpenAI noted, adding that much of the regression against GPT-6.1 Sol involved promises of future help that Terra indicated it would not actually carry out.
OpenAI warns that these tests were conducted in simulated environments, so the results may not necessarily reflect how GPT-6.1 Terra behaves in real-world production environments.
But we still aren’t seeing meaningful improvements across every area of security, even as these companies claim their models can control computers and code better.
Overall prevention scores can hide what happens after initial access. Once attackers are using valid credentials, prevention drops sharply.
The Blue Report 2026 measures defenses technique by technique across 338 million simulations run in customer production environments.
OpenAI launches GPT-6.1 Terra with 3x faster AI agents, 50% lower pricing
OpenAI’s AI agents accidentally uploaded user-provided images to third-party sites
OpenAI is preparing a $500 ChatGPT Pro Max plan with faster Codex
OpenAI says GPT-6 Astra can find zero-days, but is also harder to monitor
ChatGPT can now connect to your personal apps to mimic writing style
Critical ‘HTTP2Shell’ Brotli vulnerability triggers emergency patches
Microsoft releases emergency .NET patches for critical Brotli flaw
PHP2Shell security flaws now sucessfully exploited in the wild
Who really has control of your AI agent? Gen is taking on the agent trust problem.
What can your AI agents actually do? Learn to control every action in our Nov. 18 webinar.
Stay one step ahead of new threats in the new year. Join Huntress for the monthly Tradecraft Tuesday.
Get your free 30-day browser security assessment
CTA: Hijacked sessions reach every app your users touch. See how Exaforce agents contain identity attacks
Terms of Use – Privacy Policy – Ethics Statement – Affiliate Disclosure
Read our posting guidelinese to learn what content is prohibited.
