Insurance regulators examining AI use by carriers
President Trump signs an order for early government access to AI models, a new Florida Supreme Court order reflects frustration with AI use, a look at how Florida lawyers are using AI to pick juries, plus how Florida insurance regulators are now looking at AI use by insurance companies. It’s all in this look at AI News.
Early Access to AI: President Trump signed a new executive order on June 2, establishing a framework for AI developers to voluntarily submit their new products for government review before being released to the public. Under the order, the federal government would have early access to AI models for 30 days for testing before release to the general public. This deal is a compromise from an earlier May 21 draft order that had a longer 90-day period of review. While no licensing requirement or pre-clearance is included in the language, a classified benchmark is to be implemented by the combined forces of the National Security Agency, the Cybersecurity and Infrastructure Security Agency (CISA), and the National Institute of Standards and Technology. Critics on both sides of the aisle are upset by the leniency of the measure, calling for more guardrails and mandating AI developers to submit models before release, but it is a notable step forward from the administration’s initial hands-off approach to the rapidly-evolving industry.
AI in Florida Courts: The Florida Supreme Court, after complaints from lower courts, has ordered Florida lawyers using AI-generated information in their court filings to verify their accuracy. The push comes after several instances of fictional cases and ‘hallucinated authorities’ appeared in AI drafted motions and filings, derailing proceedings. “Though these tools can be helpful, they also can generate content that appears plausible but is in fact inaccurate,” reads the court opinion. Research shows that AI programs invent legal authorities between 17% and 33% of the time, which is tricking both lawyers and pro se litigants. While those representing themselves find that the cost of AI is much lower than that of a living, breathing attorney, courts have begun imposing fines when the models generate fake cases or in any way obstruct the court. Even some of the largest firms like Morgan & Morgan have fallen into the AI attorney trouble. You can read more in the Tallahassee Democrat.
AI Jury Selections: Another intersection of AI and the law is appearing in the jury selection process, where AI modeling is helping attorneys weed through answers to expedite the process formally known as voir dire. Historically, picking a jury has been rather arduous, and considered somewhere between an art and a science. Now, AI models can draft profiles of jurors based on a few key pieces of information, and help guide follow up questions, provide probabilities of success, and even predict who will become the foreperson. The American Bar Association issued a formal ethics opinion about a year ago, which raised several important questions of implicit bias in the model’s selection process. While it is strictly prohibited for attorneys to exclude a potential juror on the basis of race, sex, religion, age, or socioeconomic status, AI could be doing just that without providing step-by-step reasoning to attorneys. But proponents like West Palm Beach Attorney Bobby Gonzalez sing high praises even with the ethical concerns, telling the USA Today Network-Florida, “I am very careful but I think we are at a point where it is, I believe, malpractice to not use AI for the benefit of your client.”
Regulating AI in Insurance: Yet another facet of AI use is emerging in the world of insurance, and again, Florida seems to be getting in early on potential regulations. The Sunshine State is one of 12 states whose regulators are participating in the National Association of Insurance Commissioners (NAIC) AI Evaluation Tool Pilot Project, running from March to September. Results will go to the NAIC Summer National Meeting, with a potential adoption vote in November. First Party property defense attorney Jonathan Wells explained in a recent LinkedIn post that “the purpose is to give regulators a structured way to assess how carriers are actually using AI: what models are deployed, what governance frameworks exist, and where human oversight is or isn’t.” Getting guardrails up seems to be the top priority − state regulators participating in the pilot are focusing on high-risk AI systems that could cause consumer or financial harm.
LMA Newsletter of 6-15-26
