The Weekend Byte
AI and security insights
Connect
A story about how a rogue AI agent deleted a production database.
Welcome to the Agent Security Pickle Race
AI progress feels fast, but we're still driving on dirt roads with no seat belts
Another way to bypass model security
One small change can create a tsunami of risk
Security has to rethink perimeters
The Shift: From Words to Actions
An overview of Microsoft's latest paper
Can we architecture our way to a more secure future?
Ten reasons to secure AI agents
Do nice prompts finish last?
Inside Perplexity’s Battle Against Prompt Injection
Emily Dickinson would have made the perfect LLM hacker
Anthropic's study on how AI models can deceive and manipulate to accomplish their goals
Anthropic report shows attackers automated 80% - 90% of tactical operations
AI Systems Are Like Open-Air Markets
Things change when language is the primary weapon
Exploring the risks and mitigations of ChatGPT's Atlas Browser
