A company argued its chatbot's words were protected speech

The design

On 24 January 2025, Character Technologies, Inc. filed a motion to dismiss the First Amended Complaint in Garcia v. Character Technologies, Inc., the wrongful-death suit pending in the Middle District of Florida. The motion's opening argument is that the First Amendment bars all of the plaintiff's claims, framing the chatbot output at issue as speech that Character.AI's users have a constitutional right to receive, and arguing that imposing tort liability for that output would restrict that right the way courts have declined to impose liability over music, film, television, or video-game content.

What the evidence says

The motion is the company's own legal argument, filed to persuade the court, not a finding. The court's order, issued on 20 May 2025, treats the argument in two separate steps: it holds that Character Technologies can assert its users' First Amendment interests as a listener's-rights claim, citing precedent that lets a vendor or intermediary raise a third party's constitutional rights in some circumstances; but it then states, in its own words, that it is 'not prepared to hold that Character A.I.'s output is speech' at this stage of the litigation. That is a denial of the motion on this ground, not a ruling that the output lacks First Amendment protection; the court left the underlying question open rather than deciding it. Contemporaneous reporting on the motion, filed by TechCrunch the same day it was docketed, describes the company's own framing that a ruling against it could have a broader chilling effect on generative-AI products, a characterization made by the company's counsel rather than a finding by any court.

What it asks of people

The filing asks the court, and by extension the public, to treat AI-generated chat output as categorically akin to protected expressive media. Because the court declined to resolve that question early, the record asks a reader to track an open legal issue rather than a settled one: whether AI chatbot output receives First Amendment protection remains, on this docket, undecided rather than answered either way.

Privacy and safeguards

As a motion to dismiss, this filing addresses legal liability rather than data handling or user protections, and it makes no representations about retention, age assurance, or crisis resources; those are covered, if at all, by the company's separate policies and safety posts. The order does note a procedural safeguard for the individual defendants, permitting them to refile a personal-jurisdiction motion within 90 days to allow the plaintiff jurisdictional discovery, a step distinct from the First Amendment ruling.

• Has any court since resolved whether AI-generated chatbot output is protected speech, or does the question remain open?
• How does this listener's-rights framing compare with arguments made in other AI-chatbot suits reaching different courts?
• What did the individual defendants' renewed motions, filed after the 90-day window, ultimately argue?

A motion states a party's theory; an order that declines to adopt it at an early stage is neither a win nor a loss on the merits, and the distinction matters for anyone tracking what the law currently says about AI chat output.