Anthropic anticipates AI-powered virtual employees to begin navigating corporate networks in the next year (2026). Virtual employees would take that automation a step beyond from agents. These AI identities would possess their own “memories,” their own roles in the organization and even their own corporate accounts and passwords. AI workers would have a level of independence that far surpasses what agents have today. Virtual employees could be the next AI innovation hotspot, Jason Clinton, Anthropic’s chief information security officer. “In that realm, there are so many problems that we haven’t resolved yet from a security perspective that we need to solve,” Clinton said. For example, AI employees could go rogue and infiltrate the company’s continuous integration system — where new code is merged and tested before it’s deployed. Clinton: “In an old world, that’s a punishable offense,” he said. “But in this new world, who’s accountable for an agent that was executing for a couple of weeks and got to that point?”
Problems to tackle for AI workers:
- how to protect the AI employee’s user accounts,
- what network admission it should be given and
- who is responsible for controlling its actions
Virtual employee security is one of the biggest security domains where AI companies could be making investments in the next few years. There’s a demand for solutions that provide visibility into what an AI employee account is performing on a system and also on tools that create a new account classification framework that better accounts for virtual employees. Okta launched a unified control platform in February to better safeguard non-human identities and constantly monitor what systems each company account has access to and monitors for suspicious activity Integrating AI into the workplace is already causing complications, and figuring out how to administer virtual employees won’t be easy.
References
https://www.axios.com/2025/04/22/ai-anthropic-virtual-employees-security
https://www.shrm.org/topics-tools/news/technology/lattice-scraps-plans-to-treat-ai-bots-as-employees-after-backlash