OpenAI faces political and legal heat
OpenAI's public standing looks fragile: critics say the company's momentum feels off, and Florida's attorney-general has opened an investigation into ChatGPT over safety and national-security concerns. That rising scrutiny makes traceability, reviewer calibration and auditable human-feedback workflows more valuable to labs and their vendors. (theverge.com) (theverge.com)
Florida’s attorney general said on April 9 that subpoenas are coming for OpenAI, and he tied the probe to ChatGPT’s alleged use in the April 2025 Florida State University shooting, plus concerns about minors, privacy, and national security. (politico.com) James Uthmeier, Florida’s attorney general, said his office is investigating whether OpenAI’s tools can be linked to child exploitation material, self-harm, and foreign access to sensitive data, and local Florida outlets reported the state plans to issue subpoenas as part of the inquiry. (wusf.org) (wptv.com) OpenAI said after the Florida State University shooting that it identified a ChatGPT account believed to be associated with the suspect and proactively shared information with law enforcement. (wptv.com) That would be a serious political problem for any artificial intelligence company, but it lands at a moment when OpenAI is already looking less like an untouchable winner and more like a company juggling too many fronts at once. The Verge reported this week that executive reshuffles, canceled efforts, and strategy changes have raised questions about how steady the company really is. (theverge.com) The contrast is sharp because OpenAI closed a $122 billion funding round on March 31 at an $852 billion post-money valuation, one of the biggest private capital raises ever, and then walked straight into a week of headlines about probes, departures, and instability. (openai.com) (cnbc.com) Bloomberg reported on April 3 that chief operating officer Brad Lightcap moved into a special-projects role and two other senior executives went on leave because of health issues, which added to the sense that the company’s leadership bench is shifting during a possible public-market run-up. (bloomberg.com) At the same time, OpenAI has been buying more of the plumbing around safety and testing. On March 9, it said it would acquire Promptfoo, a startup whose software helps companies find vulnerabilities in artificial intelligence systems before those systems go live. (openai.com) (cnbc.com) That kind of tooling matters because a modern chatbot is not just one model answering one question. Big customers now want logs, reviewer checks, repeatable evaluations, and a paper trail that shows why a system answered the way it did and who signed off on the safeguards. (openai.com) (promptfoo.dev) So the Florida probe is not just a state-level political story. It is also a signal that the next fight around artificial intelligence may be less about who has the flashiest model and more about who can prove, in courtrooms and procurement reviews, that their systems can be audited after something goes wrong. (politico.com) (openai.com)
