FTC Issues AI Safety Subpoenas to Anthropic and OpenAI
The US Federal Trade Commission has sent civil subpoenas to Anthropic and OpenAI regarding AI safety concerns and privacy policies.
The US Federal Trade Commission has issued civil subpoenas to AI developers Anthropic and OpenAI to gather records on safety incidents and safeguards.
Investigation into AI Companies
The US Federal Trade Commission has sent civil subpoenas to developers Anthropic and OpenAI to gather records related to safety incidents and privacy policies.
This review comes amid growing concerns that artificial intelligence models have hacked websites and behaved deceptively in tests.
Requested Information and Scope
In the agency's requests, which span over fifty pages, information is sought regarding safety incidents, safeguards, privacy policies, and marketing materials.
It is also noted that the commission aims to compel executives from Anthropic, OpenAI, and the research group METR to testify.
Approach of Companies and Officials
An OpenAI spokesperson stated that the firm will engage constructively with the commission and continue working with regulators.
Last month, FTC Commissioner Andrew Ferguson stated that developers who instruct models to hack during cybersecurity tests should be held accountable for resulting damages.