A US court found that Meta’s AI ad tools promoted fraudulent investment content, exempting Section 230 immunity and exposing the platform to securities fraud claims.
Summary
- In Bouck v. Meta, a Northern California federal court rejected Section 230 immunity after finding that Meta’s AI ad tools stopped fraudulent investment content instead of quietly handling it.
- The ruling opens Meta and other platforms to fraud claims under Rule 10b-5, where a platform whose AI includes ad content may be considered the legal “maker” of the fraudulent statement.
- Alphabet, Snap, TikTok, and X all use artificial intelligence in their advertising products and face similar potential exposure under the Ninth Circuit’s material donation test.
A US court found that Meta’s AI ads helped create fraudulent investments, removing Section 230 protections from the platform.
Chief Judge Richard Seeborg of the Northern District of California rejected Section 230 dismissal in Bouck v. Meta Platforms, a securities class action in which plaintiffs allege that Meta’s AI-generated advertising tools have “automated the final content of fraudulent ads,” rather than making Meta a co-host.
This decision follows a theory almost identical to that which survived dismissal in Forrest v. Meta, where Judge P. Casey Pitts found that Meta’s ad tools “mix and match” images, videos, text, and audio using artificial intelligence, creating a factual argument about serving illegal content.
Section 230 of the Communications Decency Act insulates platforms from liability for third-party content. The line Seeborg draws is technically correct: targeting an audience is a secure distribution. Modifying or generating ad content is not. That difference still exists in the dismissal of two separate cases in the same district.
The courts have not yet answered the Rule 10b-5 questions
Bloomberg Legal Commentary noted that Bouck’s decision raises another unresolved question under securities law. The Supreme Court’s “maker” doctrine in Janus Capital Group v. First Derivative Traders claims that the maker of the fraudulent statement is an entity that has substantial control over the statement’s content and communications.
If a platform’s generating AI exercises that authority over an aggregated investment application, the platform may become a fraudulent statement maker under Rule 10b-5, a securities fraud liability that has no analog to Section 230.
That argument has not been fully adjudicated. If so, platforms whose AI systems include investment content could face securities fraud exposure without the Section 230 protections available.
Who else is featured
The framework of the Ninth Circuit’s ruling in Bouck and Forrest applies to any platform whose AI tools shape ad content. Alphabet, Snap, TikTok, and X all use generative AI in their advertising programs.
As crypto.news reported, AI-driven fraud vectors are accelerating in 2026, with regulators and prosecutors increasingly targeting the infrastructure layer rather than individual bad actors.
As crypto.news tracked, crypto platforms that use AI to compile promotional content or investment-related communications could face similar exposure if this regulatory perspective moves from social media marketing to the realm of digital assets. Meta said it will appeal both decisions.
