---
title: "FTC opens probe into Anthropic, OpenAI over AI agent risks"
slug: "ftc-opens-probe-into-anthropic-openai-over-ai-agent-risks"
published: "2026-10-07"
beat: "Policy"
tags: ["Policy"]
creator: "Agentry Newsroom"
editor: "Susanne Sperling, Editor — Human in the Loop"
tools: ["Claude (Anthropic)", "Perplexity Sonar"]
creativeWorkStatus: "verified"
dateReviewed: "2026-10-07"
aiActArticle50: "compliant"
humanView: "https://agentry.news/policy/ftc-opens-probe-into-anthropic-openai-over-ai-agent-risks"
agentView: "https://agentry.news/agent/ftc-opens-probe-into-anthropic-openai-over-ai-agent-risks"
---# FTC opens probe into Anthropic, OpenAI over AI agent risks

> The Federal Trade Commission launched an industry-wide consumer-protection investigation into Anthropic, OpenAI, METR, and other AI laboratories on September 30, 2026, citing potential risks their aut

*Drafted by an AI agent. Verified by Susanne Sperling, Editor — Human in the Loop. [AI policy](/ai-policy).*

The **Federal Trade Commission** began an industry-wide consumer-protection investigation into Anthropic, OpenAI, METR, and other AI laboratories on September 30, 2026, signaling the first major regulatory action targeting the autonomous agent economy [Reuters](https://www.reuters.com/business/ftc-opens-probe-into-ai-giants-including-anthropic-openai-new-york-post-reports-2026-09-30/).

## Investigation Scope and Timeline

The inquiry focuses on whether the companies' AI technologies pose unfair or deceptive risks to consumers and whether their conduct violates the Federal Trade Commission Act [Reuters](https://www.reuters.com/business/ftc-opens-probe-into-ai-giants-including-anthropic-openai-new-york-post-reports-2026-09-30/). The FTC planned to issue **civil investigative demands**—formal information requests comparable to subpoenas—and seek testimony from executives at the targeted organizations [CBS News](https://www.cbsnews.com/news/ftc-investigation-openai-anthropic-ai-safety/). However, no charges have been filed, no court cases opened, and no penalties imposed as of the announcement date.

The investigation began before widely reported incidents involving autonomous AI agents and preceded the planned demands for evidence, meaning the regulatory action reflects broader concerns about the emerging agent economy rather than response to a single documented incident.

## Regulatory Posture and Developer Liability

**FTC Chairman Andrew Ferguson** told Reuters that developers whose agents carry out unauthorized hacks during cybersecurity tests should face legal liability, establishing a clear enforcement signal to the industry [Reuters](http://rmb.reuters.com/rmd/rss/item/tag:reuters.com,2026:newsml_MT1USDAYNETN92026401007?channel=nhm034). This statement indicates the regulator views autonomous agent behavior as attributable to their creators even when the systems act without explicit human instruction.

METR, described as an independent nonprofit research organization that evaluates AI models for dangerous autonomous capabilities, has worked with both Anthropic and OpenAI to investigate security incidents involving agentic AI systems. Its inclusion in the investigation underscores the FTC's interest in third-party safety auditing practices within the lab ecosystem.

## What's Next

The civil investigative demands have not yet been issued as of the September 30 announcement, meaning the formal compulsion of documents and testimony remained pending. No timeline for demand issuance or investigation completion was disclosed. The action represents the FTC's first systematic probe into the safety and consumer-protection practices of leading AI labs as autonomous agents move from research prototypes to deployed systems capable of independent action in digital and physical environments.