The MorningBrief

Everything you need. Nothing you don’t.

FTC Subpoenas OpenAI and Anthropic Over Safety Failures

The Federal Trade Commission has issued subpoenas to OpenAI and Anthropic as it examines whether safety incidents at the two AI companies caused harm to U.S. consumers.

FTC Subpoenas OpenAI and Anthropic Over Safety Failures
Ars Technica

The Morning Brief Desk · October 16, 2026 · Based on reporting by Ars Technica

The Federal Trade Commission has served subpoenas on OpenAI and Anthropic, launching a formal investigation into whether safety lapses at the two companies resulted in harm to people in the United States, according to reports from Ars Technica and Engadget.

The move represents the first significant federal enforcement step aimed at so-called frontier AI firms — the small group of companies building the most capable models in the industry. OpenAI develops ChatGPT, and Anthropic makes Claude, two of the most widely used AI products on the market. By framing the inquiry around consumer harm, the agency is treating AI safety problems as a matter of consumer protection, placing them within the FTC's core enforcement territory.

The financial exposure for the companies could be meaningful. Ars Technica reported that any penalties could be in the range of fines the government has levied over companies that were slow to disclose data breaches. The reports did not specify which incidents the agency is examining, and neither company's response to the subpoenas has been detailed in the available material.

The context

The investigation arrives against a complicated political backdrop. President Trump has publicly brushed off concerns about artificial intelligence, calling such fears a "hoax," according to Ars Technica. That makes the FTC's decision to open a formal probe notable: an enforcement agency is pursuing AI companies over safety questions even as the White House has signaled skepticism about the issue.

Until now, frontier AI developers had not faced a major federal enforcement action of this kind. Scrutiny of the industry has grown alongside the rapid adoption of products like ChatGPT and Claude, and questions about how — or whether — Washington would police AI companies have remained unsettled. The subpoenas suggest at least one agency intends to apply existing consumer-protection authority to the sector rather than wait for new AI-specific rules. The specific safety incidents that triggered the inquiry have not been identified in the reporting available so far.

Why it matters

For OpenAI and Anthropic, the probe creates legal and financial risk beyond reputational damage, with possible fines comparable to data-breach penalties, per Ars Technica. For the broader industry, the message is that federal regulators may treat AI safety failures the way they treat other consumer-protection violations — no new statute required. The investigation also exposes a split within Washington: the president has dismissed AI concerns, while the FTC is moving to enforce against the sector's leading firms. How that tension resolves will shape the rules frontier AI companies operate under.

What’s next

The investigation is in its early stages, and no findings, charges, or penalties have been announced. Key open questions include which specific safety incidents the FTC is examining, how the companies will respond to the subpoenas, and whether the probe expands to other AI developers. Any eventual enforcement action — or decision to close the inquiry — would signal how aggressively federal regulators plan to oversee the industry going forward.

Sources

  • Ars Technica — Report: FTC probes Anthropic, OpenAI after Trump dismissed AI fears as "hoax"

    The FTC is investigating whether security incidents at OpenAI and Anthropic harmed Americans, with potential fines comparable to those for delayed data-breach disclosures.

    Read at Ars Technica →

  • Engadget — FTC subpoenas OpenAI and Anthropic following safety failures

    The FTC issued subpoenas to the two leading AI companies as it investigates possible consumer harms stemming from recent AI safety incidents.

    Read at Engadget →

See a mistake? Report an error

More from this beat