The Download: AI “coworkers” and stratospheric internet
infonewsLLM-Specific
safetyindustry
Source: MIT Technology ReviewJune 30, 2026
Summary
A Boston University study found that managers caught 18% fewer errors when reviewing work attributed to an AI agent (described as a digital "coworker") compared to the same work labeled as coming from a chatbot, suggesting that humanizing AI tools may reduce human oversight and vigilance. Major tech companies like Microsoft, OpenAI, Anthropic, and Google are releasing tools to manage teams of AI agents marketed as digital colleagues, which researchers warn could be problematic for workers and workplace quality control.
Classification
Attack SophisticationModerate
Impact (CIA+S)
safety
AI Component TargetedAgent
Affected Vendors
MicrosoftOpenAIAnthropicGoogle
Related Issues
Monthly digest — independent AI security research
Original source: https://www.technologyreview.com/2026/06/30/1139954/the-download-ai-agents-coworkers-solar-powered-internet/
First tracked: June 30, 2026 at 02:00 PM
Classified by LLM (prompt v3) · confidence: 82%