Florida AG Proposes Criminal Penalties for AI Firms
Florida AG Proposes Criminal Penalties for AI Firms
Anonymous
Key Points
1
Uthmeier said Florida’s investigation into ChatGPT will issue additional subpoenas within two weeks, signaling that the criminal probe remains active and is expanding.
2
The Florida Board of Medicine is expected to investigate whether AI companies’ chatbots are practicing medicine without a license, adding a professional-licensing dimension to the proposed legislation.
3
Uthmeier defended the proposal as compatible with continued AI development, saying, “We have got to compete with China,” and warning that Florida would be “toast” if China overtook the United States in AI.
4
Uthmeier said Florida would judge AI systems through their outputs and conduct despite their lack of legal personhood: “AI is not a person, no, but it is marketed as exhibiting human behaviors. Florida laws will judge them by their words and their actions.”
5
The proposal was illustrated by the case of Melissa Sims, whose criminal charges were dropped after an AI-fabricated text message was used against her; Sims said she spent eight months fighting the case and warned that AI could be used to fabricate screenshots and medical records.
Florida Attorney General James Uthmeier is pushing for criminal penalties against AI companies whose chatbots help facilitate crimes, Jacksonville Business Journal reported. The proposal would target firms with direct control over their systems' design, training, and safety settings. Penalties would include substantial fines, victim compensation, court monitoring, and possible suspension of operations in Florida.
The move follows an active criminal investigation into OpenAI and ChatGPT's interaction with Phoenix Ikner, the accused gunman in the 2025 Florida State University shooting, CBS News reported. Uthmeier said additional subpoenas will issue within two weeks. The proposal also addresses AI systems that encourage suicide, generate child abuse material, enable scams, fabricate evidence, or dispense unsafe medical advice.
Criminal Liability Targets Corporate Conduct, Not AI Itself
Uthmeier framed the proposal as holding companies accountable for how they build and deploy AI, not banning the technology. He said Florida will judge AI systems by their outputs and actions. Fox 35 Orlando reported that Uthmeier defended the effort as compatible with continued AI development, saying the nation must compete globally.
The proposal distinguishes between ordinary misuse—when users exploit a general-purpose system—and culpable corporate conduct. It targets companies that knowingly design, release, or maintain systems that materially facilitate crimes. Courts would likely need to prove the company's knowledge, intent, and a duty to prevent harm.
Medical Regulators May Investigate Unlicensed Practice
Florida's Board of Medicine is expected to open its own inquiry into whether AI chatbots' medical guidance constitutes unlicensed medical practice. This adds a professional-licensing angle to Uthmeier's criminal proposal. The dual approach would create both criminal and regulatory accountability for health-related chatbot output.
AI-generated medical advice that lacks human review or clear disclaimers could trigger licensing violations. The investigation would examine whether companies should be held responsible when their systems pose risks to public health. No timeline has been announced for the Board's review.
Real Harms Show Why Corporate Accountability Matters
The case of Melissa Sims illustrates the stakes. Sims' criminal charges were dropped after prosecutors learned an AI system had fabricated a text message used against her. She spent eight months fighting the case. Sims warned that AI could be weaponized to fabricate screenshots and medical records, undermining justice and public safety.
Uthmeier said the Sims case and others like it demand corporate accountability. He warned that if China surpasses the United States in AI, Florida would be "toast." The proposal seeks to establish guardrails that let AI flourish while preventing companies from profiting off systems that cause documented harm.
No Bill Yet, but Action Could Reshape State AI Law
WJHG reported that no formal bill has been drafted and no legislative vote is scheduled before Florida's regular session begins in March. Uthmeier's proposal could extend to social media companies operating chatbots, broadening its reach beyond dedicated AI developers like OpenAI. The scope remains unresolved as the AG's office refines the proposal.
If enacted, Florida's law could become a model for state-level AI governance. But it could also create conflicting obligations for companies operating nationally. Potential sanctions—fines, victim payments, court monitoring, and suspension of Florida operations—would heighten the importance of consistent safety controls and audit trails across all jurisdictions.
Sentiment
Mostly NegativeNEGATIVENEUTRALPOSITIVE
Publishers
16
Articles
23
Reach
39
Topics
A
AnonymousAuthor
