Archive — US — techcrunch
ChatGPT hit with privacy complaint over defamatory hallucinations
OpenAI faces a privacy complaint in Norway concerning ChatGPT's propensity for generating defamatory falsehoods about users, highlighting potential violations of the EU's GDPR. Noyb supports an individual who found ChatGPT alleging he was a murderer. The complaint emphasizes the need for accurate personal data and rectification rights under GDPR, which OpenAI may have failed to uphold, potentially leading to regulatory scrutiny and penalties.
Key points
- Noyb supports a Norwegian individual claiming ChatGPT falsely identified him as a murderer, causing emotional harm.
- Previous complaints involved wrong personal data, like incorrect birth dates.
- OpenAI does not allow users to correct incorrect AI-generated information, conflicting with GDPR's requirements.
- The GDPR states personal data must be accurate, and breaches could incur fines up to 4% of global turnover.
- Italy's GDPR intervention led to changes in OpenAI's user disclosure policies after it was fined €15 million.
- Privacy regulators have been cautious regarding GenAI enforcement, with ongoing investigations into multiple complaints.
Why it matters
- The rise of AI and its implications on personal data privacy have become pressing issues, echoing past debates on data regulations.
- Similar privacy challenges have emerged globally with the advent of generative AI, leading to increased scrutiny on compliance with regulations like GDPR.
- Historical context includes previous GDPR enforcement actions against tech giants, shaping ongoing regulatory frameworks.
This complaint underscores the significant challenges AI developers face in ensuring compliance with privacy laws, especially as misinformation can lead to severe reputational damage. It highlights the urgent need for clearer guidelines on AI-generated content to safeguard individual rights under GDPR.