OpenAI faces a new privacy complaint in the European Union over ChatGPT’s hallucination issue.

Non-profit organization Noyb is filing a complaint against OpenAI because of ChatGPT’s inability to correct false information, the organization wrote in a press release. The so-called hallucination problem mainly concerns false personal information. Furthermore, the company cannot say where it obtains this data from. OpenAI openly acknowledges this problem, but has not yet taken any action to combat it, according to the organization. Noyb then filed a complaint with the Austrian data protection authority.

Hallucinations

It is already known that generative AI tools dare to hallucinate and therefore find answers. ChatGPT has also been struggling with this for some time. According to Noyb, OpenAI does not acknowledge this problem and hides the fact that “factual accuracy in large language models continues to be an area of ​​active research.”

Users use ChatGPT for various purposes. Although the company has extensive training data, there is currently no guarantee that ChatGPT will display accurate information to users. This is relatively acceptable for homework, but unacceptable when personal data is involved. Such hallucinations are not in line with the General Data Protection Regulation (GDPR), which states that information about people must be accurate.

Filed a complaint

Noyb has asked the Austrian Data Protection Authority (DSB) to investigate OpenAI’s data processing and the measures the company has taken to ensure the accuracy of the personal data processed. Additionally, Noyb is calling for OpenAI to be fined to ensure future compliance. Noyb suspects that this problem will be solved through EU cooperation.