To manage security risks, treat AI agents like new employees. Provide them with their own isolated environment—separate accounts, scoped API keys, and dedicated hardware. This prevents accidental or malicious access to your personal or sensitive company data.

Each AI agent acting on a user's behalf creates a new "non-human identity" with its own keys and API access. This proliferation of autonomous agents dramatically increases the number of potential exploit points, a problem traditional security models weren't designed to handle.

The future of work involves potentially millions of AI agents operating within a company. This requires a new governance layer, including agent inventories, inspectable reasoning traces, identity management, and sandboxed execution environments to maintain security and control.

A practical security model for AI agents suggests they should only have access to a combination of two of the following three capabilities: local files, internet access, and code execution. Granting all three at once creates significant, hard-to-manage vulnerabilities.

To control costs, security, and governance, enterprises are moving from interactive, ad-hoc agent use to a 'software factory' model. This approach systematizes the entire work lifecycle, automating processes and minimizing the risks associated with inconsistent human operation of powerful AI tools.

To address security concerns, powerful AI agents should be provisioned like new human employees. This means running them in a sandboxed environment on a separate machine, with their own dedicated accounts, API keys, and access tokens, rather than on a personal computer.

Securing AI agents requires extending the concept of 'least privilege' (access to data) to 'least agency' (scope of autonomous actions). This OWSAP-coined term means an agent should only be granted the minimum capability to perform its function, constraining its potential 'blast radius' if compromised.

For maximum security, run different AI agents on separate physical machines (like Mac Minis). This creates a hard barrier, preventing an agent with access to sensitive data (e.g., finances) from interacting with an agent that has external communication channels (e.g., scheduling via iMessage), minimizing the risk of accidental data leaks.

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.