OpenAI's 'hallucination' problem hit with another privacy complaint...
OpenAI is facing another privacy complaint in the European Union. This complaint, filed by privacy rights nonprofit noyb on behalf of an individual complainant, focuses on the inability of its AI chatbot ChatGPT to correct misinformation it generates about individuals.
The issue of AI tools generating incorrect information has been well-documented, putting them at odds with the EU's General Data Protection Regulation (GDPR) that governs the processing of personal data of regional users. Penalties for GDPR compliance failures can be severe, with fines of up to 4% of global annual turnover.
GDPR and OpenAI
OpenAI had to make adjustments following an intervention by Italy's data protection authority, which led to a temporary shutdown of ChatGPT back in 2023. The latest GDPR complaint, filed by noyb with the Austrian data protection authority, addresses an incident where the AI chatbot provided an incorrect birth date for a "public figure."
Under the GDPR, individuals in the EU have rights concerning their information, including the right to have incorrect data corrected. noyb argues that OpenAI is not fulfilling this obligation regarding ChatGPT's output. Despite requests for correction, OpenAI claimed it was technically impossible to rectify certain inaccuracies.
OpenAI's Response
OpenAI's privacy policy allows users to request corrections for factually inaccurate information generated by the AI chatbot. However, due to the technical complexity of the models, corrections may not always be feasible. In such cases, users are advised to request the removal of their personal data from ChatGPT's output.
GDRP rights are not selective, and individuals have the right to request data rectification or deletion. OpenAI must adhere to these rights, as pointed out by noyb in the latest complaint.
Legal Challenges
Legal challenges regarding data privacy and AI technologies continue to arise, with companies like OpenAI navigating the complex landscape of regulatory compliance and user rights. The outcome of this latest privacy complaint could shape how generative AI tools operate within the EU in the future.
