OpenAI says prompt injections will always be a risk for AI browsers with agentic capabilities, like Atlas. But the firm is beefing up its cybersecurity with an “LLM-based automated attacker.”
OpenAI says AI browsers may always be vulnerable to prompt injection attacks
Published on Dec 22, 2025
•
1 min read
Related Articles
cybersecurity
France’s postal and banking services disrupted by suspected DDoS attack
France’s postal service, La Poste, said it was hit by a disruptive cyberattack that knocked…
AI
ChatGPT launches a year-end review like Spotify Wrapped
The experience includes awards, poems, and pictures referencing your year in chat.
AI
John Carreyrou and other authors bring new lawsuit against six major AI companies
These authors rejected Anthropic’s class action settlement, arguing that “LLM companies should not be able…
cybersecurity
Inside Uzbekistan’s nationwide license plate surveillance system
The Uzbek government’s national license plate scanning system was discovered exposed to the internet for…