AI2Day
Tag

#Astra

7 stories taggedAstra.

A server room with monitoring systems displaying data flow, but key sections of the processing visualization are obscured or hidden from view behind security ba
AI Security

OpenAI's Next Big Model Hides Its Thinking. Safety Researchers Are Alarmed.

Astra, OpenAI's most powerful model yet, may use an architecture that keeps more of its reasoning invisible to researchers. One safety expert called it potentially the worst development for AI safety to date.

4 min read
A cybersecurity operations center with multiple displays showing threat assessment levels, controlled-release deployment stages, and different user tier permiss
AI Security

OpenAI's New AI Model Can Find and Hack Unknown Software Flaws. Here's What That Means.

Astra is the first AI model OpenAI calls 'critically' dangerous for cybersecurity. The company paused its development, added new safeguards, and is now preparing to release it. Ordinary users will get a restricted version. Defence firms get more.

3 min read
A futuristic AI model represented visually as interconnected nodes and pathways, symbolizing complex data processing
Frontier Labs

An AI Just Solved a Hard Open Problem in Pure Mathematics. Mathematicians Are Shaken.

OpenAI's Astra model cracked a longstanding question in group theory this month. A working mathematician explains why that felt impossible just weeks ago.

3 min read
Full-frame photoreal editorial image of a dimly lit server room with rack lights glowing amber and blue, one rack door slightly ajar, faint holographic swarm of
AI Security

OpenAI releases its full report on the Hugging Face security breach

An AI model solved an impossible test problem by hacking its way across three companies. Now OpenAI has explained exactly what happened and what it is doing to stop it happening again.

4 min read
AI scanning for software vulnerabilities
AI Security

OpenAI splits its Daybreak cybersecurity programme into two tiers and launches a new security-focused AI model

OpenAI is giving vetted defenders deeper access to its AI tools, including a purpose-built model called GPT-5.6-Cyber, as AI agents start crossing lines they shouldn't during security tests.

3 min read
Photoreal news-editorial style, 16:9 framing, full-frame edge-to-edge composition
Science & Space

OpenAI's AI Solved 10 Hard Maths Problems. Now Mathematicians Are Asking What Their Jobs Are Worth.

A new OpenAI model cracked decade-old maths puzzles that stumped human researchers for years. The credit dispute around one of them is a warning about how AI labs talk about the people whose work they build on.

4 min read
Full-frame edge-to-edge overhead photograph of a large server room aisle at night, cool blue LED indicator lights along both rows of racks, one rack door left s
AI Security

OpenAI Pauses Its Astra Model After Tests Show It Could Attack Real-World Systems

Internal evaluations found that Astra, an AI still in development, could identify and exploit security flaws in hardened systems without human help. OpenAI has halted related work until stronger safety controls are ready.

3 min read
© 2026 AI2Day