Google AI Gaffe Highlights Escalating Legal Risks as Generative AI Goes Mainstream

Google’s AI search mistakes are escalating legal risks as the technology rapidly enters mainstream use. The AI Overview feature falsely stated that chess player Hans Niemann admitted to cheating against Magnus Carlsen, highlighting the potential for AI-generated libel to harm reputations and businesses.

Key issues with AI defamation: The rise of generative AI in consumer products is blurring the lines between platforms and publishers, raising questions about legal liability for false and damaging statements:

• AI language models are prone to “hallucination” – inventing sources, misattributing quotes, and rewriting events – which can easily turn nuances into libel.
• Traditional defamation law is difficult to apply to AI, as chatbots lack the mental states of negligence or malice required to prove libel.
• Tech companies are not directly involved in preparing specific defamatory statements, but have built unreliable AI products and integrated them into mainstream services.

Potential legal approaches: While the application of defamation law to AI must still be litigated, experts suggest several possible frameworks:

• Treating chatbots as products, not people, and holding companies liable for flaws in training data, algorithms, or safeguards that make harmful outputs more likely.
• Applying a risk mitigation approach, requiring tech companies to adequately assess and prevent risks before releasing AI systems.
• Adapting existing legal frameworks, such as false advertising and consumer protection laws, to regulate AI products.

Real-world impacts: At least two chatbot defamation cases are already underway in the U.S., with more likely to follow:

• Veteran and aviation consultant Jeffery Battle is suing Microsoft, alleging that Bing’s AI confused him with a person who pleaded guilty to seditious conspiracy, damaging his business.
• If a celebrity or public figure had their reputation ruined by a viral AI-generated lie, it could lead to high-profile legal battles and financial consequences for tech giants.

Broader implications: As AI rapidly enters mainstream use across industries, from pharma to consulting to energy, the technology’s legal risks are coming to the fore:

• Regulating AI defamation is crucial to protect individuals and businesses from harm, but could pose existential threats to the technology if companies are held financially liable.
• While AI may seem unprecedented, U.S. law has a long history of regulating new technologies and corporate behavior, suggesting that existing frameworks can be adapted rather than rewritten from scratch.
• The idea that “tech moves too fast for the law” is a common refrain, but experts argue that AI, as a tool developed by corporations, can and should be subject to legal oversight and accountability.