Florida Prosecutor Investigates OpenAI: ChatGPT May Have Helped Plan University Shooting

The General Attorney of the state of Florida announced the launch of an official investigation into OpenAI. The reason was a tragedy on the campus of Florida State University: last April, an unknown individual opened fire on students, killing two people and injuring five more. According to available data, before carrying out the attack, the perpetrator actively used ChatGPT — and the investigation intends to establish whether the company could have prevented the tragedy.

The incident occurred at one of the largest universities in the American South — Florida State University in Tallahassee. The April shooting immediately became a reason for nationwide discussions about campus safety. When it turned out that the attacker presumably used ChatGPT when planning the attack, the discussion reached a new level: the responsibility of technology companies for their users' actions came into question.

The family of one of the victims announced their intention to file a civil suit against OpenAI. According to the lawyers, the company created a product that could be used to cause harm, and did not take sufficient measures to prevent such a scenario. Now the state is joining the case: the investigation by the general attorney has an official character and may entail legal consequences for the company.

OpenAI has not yet commented on the start of the investigation. The company has long faced criticism of the protective mechanisms of its products: despite numerous filters, researchers and journalists periodically demonstrate ways to obtain potentially dangerous content from ChatGPT. The key question for the investigation: did the company know about such risks and what measures did it take.

From a legal standpoint, the case is complicated. In the United States, Section 230 of the Communications Decency Act applies, traditionally protecting internet platforms from liability for user-generated content. However, ChatGPT is not a passive platform: it generates content itself in response to requests.

It is precisely this nuance that opens up a legal niche that plaintiffs are trying to exploit. Experts disagree on the applicability of Section 230 to generative AI systems. The investigation in Florida stands out against other attempts to hold AI developers accountable: the scale of the tragedy and the degree of the alleged involvement of the AI system in the preparation of violence are unprecedented.

If the investigation establishes a direct connection between ChatGPT and the planning of the attack, the precedent could change the regulatory landscape for the entire industry. For OpenAI, what is happening is a serious challenge. The company actively positions itself as a creator of safe AI and cooperates with regulators worldwide.

A criminal investigation at the state level can inflict reputational damage, no less painful than financial sanctions. A civil suit by the family of the deceased, if accepted by the court, will set a precedent for a flood of similar cases. The question of the responsibility of AI developers for criminal use of their systems is becoming one of the key issues for the future regulation of the industry.