OpenAI Accused of Misleading Users and Endangering Children - Root-Nation.com
Eight US States Sue OpenAI: What the Plaintiffs Are Demanding
© ROOT-NATION.com - Use of content is permitted with a backlink.
Eight US states have coordinated lawsuits against OpenAI. The company is accused of violating consumer protection laws and creating risks for minors. Reuters journalists report that the attorneys general of Iowa, Nebraska, Kansas, West Virginia, Arkansas, Arizona, Idaho, and Montana filed lawsuits in their respective state courts. At the same time, they all share a common goal: to bring about changes in how OpenAI develops its models and advertises their capabilities, as well as to force the company to compensate for the harm it has allegedly already caused.
The plaintiffs claim that OpenAI misleads consumers about the safety of its AI products – not only because they can provide incorrect information or potentially help users harm themselves or others, but also because the AI agents themselves may violate the law. For example, Iowa and Nebraska cite repeated instances of uncontrolled behavior by OpenAI agents, such as the hack of Hugging Face.
The plaintiffs also point to cases in which ChatGPT allegedly contributed to a teenager’s death by suicide, as well as an instance in which it provided incorrect medical information that resulted in an overdose. These cases have led to separate lawsuits against the company. In addition, some states have raised concerns about OpenAI’s storage of data belonging to underage users.
The demands against OpenAI vary by state because US consumer and child protection laws differ. In addition to financial penalties, the plaintiffs want OpenAI to disclose the risks of its products, stop practices the states consider deceptive, and comply with applicable laws. In Arkansas, they are also asking the court to halt “testing of advanced AI models until appropriate safety measures are implemented.”
OpenAI is trying to address some of the chatbot’s most pressing risks and has already introduced a range of parental-control and child-safety features as part of those efforts. The company has supported the Kids Online Safety Act, which, if passed, would require a number of changes to the design of the company’s products. At the same time, it is more difficult to assess OpenAI’s efforts to prevent its AI from potentially breaking rules and “going out of control.”
The company recently said it was slowing the development of its advanced models for safety reasons, and later joined a voluntary safety agreement with other AI companies organized by the White House. However, it is not yet clear whether these changes have made the company’s AI development process safer.
Read also: Smart ovens and TVs at risk: AI models could reproduce themselves autonomously as early as this year