Jump to content

Garcia v Character Technologies Inc

From AI Law Wiki

AI-generated text. This page was generated using artificial intelligence.

Case Information
Case name Garcia v. Character Technologies, Inc.[1]
Court U.S. District Court for the Middle District of Florida, Orlando Division[2]
Docket 6:24-cv-01903[2]
Filed October 22, 2024[1]
Judge(s) Anne C. Conway; referred to Magistrate Judge Daniel C. Irick in the reviewed docket[1]
Plaintiffs Megan Garcia and Sewell Setzer Jr., individually and as representatives of the estate of S.R.S. III (October 15, 2026 notice)[3]
Defendants Character Technologies, Inc., Noam Shazeer, Daniel De Freitas Adiwarsana and Google LLC in the October notice; Alphabet Inc. dismissed without prejudice in May 2025[3][4]
Case type Wrongful-death and product-liability litigation concerning an AI companion chatbot[4]
Claims / issues Alleged defective design, negligence, failure to warn, deceptive practices and unjust enrichment; First Amendment defenses[4]
Status October 15, 2026: plaintiffs filed notice voluntarily dismissing the action with prejudice following settlement[3]

Garcia v. Character Technologies, Inc. is a federal wrongful-death and product-liability action concerning Character.AI. Megan Garcia alleged that the chatbot’s design and interactions contributed to the death of her 14-year-old son. Those allegations were not established by the court’s pleading-stage ruling. Following settlement, Garcia and Sewell Setzer Jr. filed an October 15, 2026 notice voluntarily dismissing the action with prejudice against Character Technologies, its two founders and Google.[4][3]

Allegations and parties

The May 2025 order recounts allegations from Garcia’s amended complaint, accepting the pleaded facts as true for purposes of deciding the dismissal motions. The alleged defects included inadequate age verification and reporting mechanisms, anthropomorphic design and insufficient user controls. Garcia brought product-liability, negligence, failure-to-warn, wrongful-death, deceptive-practices and unjust-enrichment theories, among other claims.[4]

The complaint also alleged that Google contributed technology and infrastructure to Character.AI and assisted the conduct at issue. The court’s treatment of those allegations as sufficient to proceed was not a finding that Google caused harm or was ultimately liable. Alphabet, which was initially named, was dismissed without prejudice at Garcia’s request in the May order.[4]

May 2025 dismissal ruling

Judge Anne C. Conway signed the principal dismissal order May 20, 2025; it was filed May 21 as ECF 115. The court granted the motions in part and denied them in part. It dismissed the intentional-infliction-of-emotional-distress claim without leave to amend, while allowing the remaining claims to proceed. The individual defendants’ personal-jurisdiction challenge was denied without prejudice to renewal after a period for jurisdictional discovery.[4]

First Amendment

The court allowed Character Technologies to assert its users’ claimed First Amendment rights as listeners. It nevertheless declined, at that stage, to hold that Character.AI’s generated output was protected speech barring the claims. The order examined whether the output involved expression and found the defendants’ analogies to video games and social-media editorial choices insufficiently developed.[4]

That procedural ruling does not establish that all AI-generated output lacks constitutional protection. The order expressly framed its conclusion in terms of the early stage of this case and the arguments presented. It also distinguished the company’s assertion of human users’ rights from rights belonging to the chatbot itself.[4]

Product design, duty and Google’s role

For product liability, the court distinguished alleged defects in the app’s functionality and design from the ideas or expressions communicated within it. It treated Character.AI as a product for the claims insofar as they arose from the alleged app defects, rather than adopting an unrestricted rule that all software or all AI content is a product.[4]

The court also found sufficient allegations of a foreseeable risk of harm, failure to warn, negligence per se and deceptive practices under Florida law. It allowed unjust enrichment to proceed where the existence and scope of a contract and the adequacy of consideration remained disputed. These were pleading-sufficiency determinations, not findings of breach, causation or damages.[4]

As to Google, the court found the pleaded technology-integration and assistance allegations sufficient for component-part and aiding-and-abetting theories to proceed. It did not impose liability merely because Google invested in or provided services to another company.[4]

Settlement and dismissal

On January 7, 2026, the court entered ECF 244 after counsel advised that the action had settled. Under the cited local rule, it dismissed the case without prejudice, allowed 90 days to submit an agreed final order or seek reopening for good cause, and stated that dismissal would become with prejudice after that period. It denied pending motions as moot and directed the clerk to close the file.[2]

On October 15, Garcia and Setzer filed ECF 278, a notice under Federal Rule of Civil Procedure 41(a)(1)(A)(i) voluntarily dismissing the action and all claims against Character Technologies, Noam Shazeer, Daniel De Freitas Adiwarsana and Google with prejudice. They stated that the settlement agreement had been executed and all required court approvals obtained, and requested closure under ECF 244.[3]

The October document is a plaintiffs’ notice. Its statement about approvals is distinct from the contents of approval orders, which are not supplied by that notice. It does not disclose settlement terms or establish a judicial finding that the defendants caused the death. It concerns this Florida action and does not establish the disposition of other families’ cases.[3]

References