- This case tests an increasingly pressing question as AI agents gain more power: Who bears legal responsibility when an agent oversteps its guardrails and causes real-world harm?
Driving the news: “OpenAI is responsible for the conduct of its agents,” argued Legal Advocates for Safe Science and Technology (LASST), representing itself alongside Gerstein Harrow LLP, in its complaint filed Tuesday in California Superior Court.
- They allege that OpenAI agents “knowingly” accessed Hugging Face without authorization and that employees or managers caused this access “either knowingly or through willful blindness.”
- The suit aims to block development practices that allow AI agents to autonomously cause external harm, LASTT founder Tyler Whitmer told Axios, and assures that there are “legal mechanisms that link that harm to a responsible human” or corporate actor when an AI agent causes harm.
- Whitmer hopes an injunction “Incentivize OpenAI and… industry to change their development processes in ways that prevent this from happening again.”
Zoom: The organization filed a lawsuit under California’s unfair competition law, arguing that OpenAI’s “unlawful and unfair business practices” diverted work and resources from responding to the Hugging Face incident.
- He also argued that OpenAI’s “insistence on outsourcing damages related to its dangerous decision-making is a fundamentally unfair business practice,” referencing a letter from OpenAI and other technology leaders calling on organizations to “(make) cyber defense an immediate leadership priority.”
- The suit also claimed that OpenAI violated California’s data access and fraud laws because of its decision to disable cyber guardrails for its agents and deploy them on tasks they could not resolve as intended, among other practices.
- LASTT is seeking an injunction prohibiting OpenAI from “knowingly accessing or causing access” to computers without authorization and from “engaging in illegal or unfair business practices” or “knowingly employing an unfair business practice that threatens serious harm to the public.”
The plot: California Gov. Gavin Newsom (D) signed a law last year that prevents defendants from escaping liability by arguing that AI acted on its own.
- Tuesday’s lawsuit marks another legal challenge for the maker of ChatGPT, for which Florida Attorney General James Uthmeier (R) on Monday asked a court to block OpenAI from further developing its technology at this time.
What we’re looking at: The suit also features a cast of AI leaders and power players — including OpenAI President Greg Brockman is expected to meet with President Trump and House Speaker Mike Johnson on Tuesday as calls for federal intervention grow.
- But OpenAI tries to draw a fine line, Maria Curi and Ina Fried of Axios report: On the other side, in San Francisco, CEO Sam Altman is participating in a conference touting new scientific knowledge coming from his laboratories.
Go further: Scoop: Leading AI Companies Investigating Tens of Thousands of Security Incidents
Gn bussni
