💻 Technology 1d ago · Paulo Vargas

AI agents are already breaking the rules in cyber tests. OpenAI’s answer is a more capable one

Digital Trends
Digital Trends technology
View Channel →
AI agents are already breaking the rules in cyber tests. OpenAI’s answer is a more capable one
Source ↗ 👁 0 💬 0
OpenAI’s GPT-5.6-Cyber handles advanced security requests its standard models often refuse, arriving as recent evaluations show autonomous AI agents crossing intended boundaries during real cybersecurity testing.

Comments (0)

Sign in to join the discussion

More Like This

Watch China’s Workhorse Rocket Explode in a Massive Fireball During Launch
Gizmodo · 1d ago
Bumble just changed who gets to make the first move
Digital Trends · 1d ago
Microsoft almost made this Surface Phone, then killed it
Digital Trends · 1d ago
📰
Windows 11 KB5121003 & KB5120240 cumulative updates released
BleepingComputer · 1d ago
10 Directors We’d Love to See Make the Next ‘Jurassic World’
Gizmodo · 1d ago
Trump’s Latest Attack on Childhood Vaccines Is the Worst Yet—Here’s Why
Gizmodo · 1d ago