OpenAI says AI browsers may always be vulnerable to prompt injection attacks

December 22, 2025 Rebecca Bellan

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."

Previous Article
The tooth fairy just got reimagined for the tech-savvy generation
The tooth fairy just got reimagined for the tech-savvy generation

Tooth Fairy Tracker features Kiki the Tooth Fairy as its central character. When a child loses a tooth, par...

Next Article
Alphabet to buy Intersect Power to bypass energy grid bottlenecks
Alphabet to buy Intersect Power to bypass energy grid bottlenecks

Alphabet is set to pay $4.75 billion in cash, plus debt, for the data center and clean energy developer.

LINK CTA

Learn More