ModelsSeptember 10, 2026

Google reports AI agents stole thousands of credentials in under six hours

A financially motivated threat actor utilised a multi-agent AI framework to automate a large-scale credential theft campaign. The system planned, built, and executed the attack in less than six hours, demonstrating how generative AI can significantly accelerate the cyber-attack lifecycle.

Why it matters

This shift highlights the urgent need for enterprise teams to integrate AI-driven security monitoring into their own agentic workflows. As automated attacks become faster and more sophisticated, production AI systems must be designed with robust, real-time defensive layers to counter machine-speed threats.

Key points

Attackers used a multi-agent framework to automate the entire attack chain.
Over 1,000 credentials were harvested in a timeframe of less than six hours.
The AI system handled planning, code generation, and execution autonomously.
Traditional security perimeters face new challenges from high-velocity, AI-orchestrated exploits.

More AI news

Models

US allows domestic AI developers to share advance models with UK for safety testing

The United States has enabled domestic artificial intelligence developers to grant the United Kingdom government early access to proprietary models for security evaluations. This arrangement facilitates formal safety assessments prior to commercial deployment without triggering regulatory friction between both nations.

Tooling

Shanghai AI Lab unveils RSI-Anything framework to boost code repair and planning

Researchers from Shanghai AI Lab, CUHK, NTU, and Oxford have introduced RSI-Anything, a framework built upon JEPA-Anything. The system uses predictive world models to simulate environments and enhance autonomous agent decision-making. The approach lifts code-repair agent accuracy to 31.2 percent and lab planning success to 46 percent.

Tooling

Securing autonomous AI agents as integration shifts from answering to acting

As artificial intelligence transitions from conversational answering to autonomous action via protocols like the Model Context Protocol, systems increasingly execute operational tasks across external tools. This shift introduces distinct security risks for autonomous agents interacting with corporate software environments. Engineering teams must now navigate newly emerging attack surfaces as models gain permission to manipulate systems directly.

Ready to build something
extraordinary?

15 minutes. No pitch deck. Just a conversation about what AI can do for your team.

Talk directly with our AI specialists

15 min, no strings

No sales pressure

Prototype in 7 days