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.
Related Insights
The primary obstacle for tools like OpenAI's Atlas isn't technical capability but the user's workload. The time, effort, and security risk required to verify an AI agent's autonomous actions often exceed the time it would take for a human to perform the task themselves, limiting practical use cases.
Unlike human attackers, AI can ingest a company's entire API surface to find and exploit combinations of access patterns that individual, siloed development teams would never notice. This makes it a powerful tool for discovering hidden security holes that arise from a lack of cross-team coordination.
For AI agents, the key vulnerability parallel to LLM hallucinations is impersonation. Malicious agents could pose as legitimate entities to take unauthorized actions, like infiltrating banking systems. This represents a critical, emerging security vector that security teams must anticipate.
The ease of finding AI "undressing" apps (85 sites found in an hour) reveals a critical vulnerability. Because open-source models can be trained for this purpose, technical filters from major labs like OpenAI are insufficient. The core issue is uncontrolled distribution, making it a societal awareness challenge.
Managing human identities is already complex, but the rise of AI agents communicating with systems will multiply this challenge exponentially. Organizations must prepare for managing thousands of "machine identities" with granular permissions, making robust identity management a critical prerequisite for the AI era.
Recent security breaches (e.g., Gainsight/Drift on Salesforce) signal a shift. As AI agents access more data, incumbents can leverage security concerns to block third-party apps and promote their own integrated solutions, effectively using security as a competitive weapon.
Vercel is building infrastructure based on a threat model where developers cannot be trusted to handle security correctly. By extracting critical functions like authentication and data access from the application code, the platform can enforce security regardless of the quality or origin (human or AI) of the app's code.
Organizations must urgently develop policies for AI agents, which take action on a user's behalf. This is not a future problem. Agents are already being integrated into common business tools like ChatGPT, Microsoft Copilot, and Salesforce, creating new risks that existing generative AI policies do not cover.
An AI agent capable of operating across all SaaS platforms holds the keys to the entire company's data. If this "super agent" is hacked, every piece of data could be leaked. The solution is to merge the agent's permissions with the human user's permissions, creating a limited and secure operational scope.
While AI models excel at gathering and synthesizing information ('knowing'), they are not yet reliable at executing actions in the real world ('doing'). True agentic systems require bridging this gap by adding crucial layers of validation and human intervention to ensure tasks are performed correctly and safely.
