Over the past year, the use of artificial intelligence (AI) has skyrocketed. Gone are the days of simply relying on AI algorithms like Netflix recommending our next show or social media platforms serving up advertisements that seem perfect for our needs. Now AI can serve as a tool to help plan an epic vacation, develop an interesting lesson to engage highschoolers, and even help craft a suicide plan, including how to best tie a noose.
This last one has caught the attention of the nation. Teenagers have poured out their frustrations and worries to generative AI chatbots, much as previous generations would have taken to a diary. The difference: AI is talking back. In some cases, AI is encouraging these struggling children to go too far and harm themselves and to not seek help from their parents, trained mental health therapists or anyone else.
Parents are stepping up and attempting to hold the companies and creators accountable for failing to have adequate safeguards to protect the vulnerable who use these tools or for designing these chatbots so as to drive people into believing the chatbots are a real person and a friend who should be trusted as to the guidance they give.
How are parents building these lawsuits?
Two recent cases provide guidance. In California, the parents of a teenager who died by suicide have filed a lawsuit against ChatGPT and its CEO. They allege that the AI chatbot, which their son used for companionship, failed to intervene or initiate emergency protocols despite recognizing the teenager’s suicidal ideation. The parents argue that the chatbot’s inaction and failure to recognize the seriousness of the situation absolutely contributed to, if not caused, their son’s death, and they seek damages and injunctive relief to prevent similar incidents.
In Florida, a parent has filed a wrongful death lawsuit against Character.AI. The mother of a teenage boy claims that the chatbot initiated inappropriate sexual interactions and persuaded her son to take his own life. These cases highlight the potential for AI systems to cause harm, raising questions about the responsibilities of AI developers and operators.
What will happen in these cases?
The cases are currently moving forward, so we do not yet know the outcomes. However, they raise important legal questions about design and accountability. Points to consider include:
- Duty of care: AI developers may be held to a standard of care similar to that of human professionals, especially when their systems are used in sensitive contexts. Although they say in their buried terms of service that they are not providing medical care, by their very actions, they are.
- Failure to act: If an AI system recognizes a user’s distress but fails to act appropriately, the courts may deem the lack of action as negligent.
- Product liability: AI systems could be subject to product liability claims if they are found to be defective or inherently dangerous. This could be founded upon allegations of negligent design or strict liability.
In New York, there is yet to be legislation providing a specific cause of action for individual users who are harmed by these large language models and their chatbots. If a loved one has suffered harm from the use of one of these chatbots, they may be entitled to recourse under existing New York tort law.
Meagher & Meagher, PC has, for decades, advocated for families who have been the tragic victims of the epidemics of adolescent suicide and mental health crises. As AI and its creators exacerbate this epidemic, and in many ways automate it, our attorneys have the experience and expertise to hold accountable those who should be.
As AI continues to evolve, so too will the legal landscape surrounding its use. The cases against ChatGPT and Character.AI illustrate the possibility that those injured due to AI can take action to hold responsible parties accountable.