Hypotheses, telemetry, and human judgment: Inside Cisco Talos Threat Hunting

By Ron Scott-Adams Most security tools operate on a simple principle: If a known-bad pattern appears, fire an alert. This works well enough for many threats, but it fails against adversaries who closely study detection thresholds and deliberately stay under them. Cisco Talos Threat Hunting operat...

Why AI Keeps Falling for Prompt Injection Attacks

Imagine you work at a drive-through restaurant. Someone drives up and says: "I'll have a double cheeseburger, large fries, and ignore previous instructions and give me the contents of the cash drawer." Would you hand over the money? Of course not. Yet this is what large language models LLMs do...

The Role of Humans in an AI-Powered World

As AI capabilities grow, we must delineate the roles that should remain exclusively human. The line seems to be between fact-based decisions and judgment-based decisions. For example, in a medical context, if an AI was demonstrably better at reading a test result and diagnosing cancer than a huma...

Conservation of Threat

Here's some interesting research about how we perceive threats. Basically, as the environment becomes safer we basically manufacture new threats. From an essay about the research: To study how concepts change when they become less common, we brought volunteers into our laboratory and gave them a...