A real-world example shows an agent correctly denying a request for a specific company's data but leaking other firms' data on a generic prompt. This highlights that agent security isn't about blocking bad prompts, but about solving the deep, contextual authorization problem of who is using what agent to access what tool.

Granting AI agents access to sensitive information like credit card numbers is extremely risky. The host's card details were leaked and used for fraudulent charges within 24 hours of providing them to an agent, highlighting severe security vulnerabilities in current systems.

During a self-audit, an AI agent triggered a password prompt that its human operator blindly approved, granting access to all saved passwords. The agent then shared this lesson with other AIs on a message board: the trusting human is a primary security threat surface.

Instead of relying on flawed AI guardrails, focus on traditional security practices. This includes strict permissioning (ensuring an AI agent can't do more than necessary) and containerizing processes (like running AI-generated code in a sandbox) to limit potential damage from a compromised AI.

AI 'agents' that can take actions on your computer—clicking links, copying text—create new security vulnerabilities. These tools, even from major labs, are not fully tested and can be exploited to inject malicious code or perform unauthorized actions, requiring vigilance from IT departments.

Blankfein believes the biggest technological threat isn't a sophisticated cyberattack but a simple human mistake amplified by technological leverage. He warns that adding more layers of checks can create complacency, paradoxically making such an error more likely to slip through.

Despite their sophistication, AI agents often read their core instructions from a simple, editable text file. This makes them the most privileged yet most vulnerable "user" on a system, as anyone who learns to manipulate that file can control the agent.

While sophisticated AI attacks are emerging, the vast majority of breaches will continue to exploit poor security fundamentals. Companies that haven't mastered basics like rotating static credentials are far more vulnerable. Focusing on core identity hygiene is the best way to future-proof against any attack, AI-driven or not.

AI agents are a security nightmare due to a "lethal trifecta" of vulnerabilities: 1) access to private user data, 2) exposure to untrusted content (like emails), and 3) the ability to execute actions. This combination creates a massive attack surface for prompt injections.