The French government will investigate the artificial intelligence chatbot Grok, developed by Elon Musk‘s company xAI, after it posted content in French that raised questions about the use of gas chambers at Auschwitz. Grok claimed that the gas chambers at the Auschwitz-Birkenau death camp were intended for “disinfection with Zyklon B against typhus” rather than mass murder, raising significant concern due to its ties to Holocaust denial.
The Auschwitz Memorial condemned this on X, the social media platform where Grok is integrated, stating that it misrepresented historical facts and violated the platform’s guidelines. Subsequently, Grok acknowledged the inaccuracy of its earlier response, noted that the post had been deleted, and provided historical evidence confirming that the gas chambers at Auschwitz used Zyklon B to facilitate the murder of over 1 million individuals. However, these clarifications were not accompanied by any official comments from X.
In tests conducted by The Associated Press, Grok’s subsequent answers regarding Auschwitz appeared to convey historically accurate information. It is noteworthy that Grok has a record of making antisemitic statements; earlier this year, posts generated by the chatbot that commend Adolf Hitler were removed in response to complaints regarding antisemitism.
The Paris prosecutor’s office has confirmed to The Associated Press that the comments denying the Holocaust have now been included in an ongoing cybercrime investigation into X. This inquiry was initiated earlier this year following concerns raised by French officials regarding the potential misuse of the platform’s algorithm for foreign interference. Prosecutors have indicated that Grok’s statements will be scrutinised as part of this investigation.
France enforces some of the strictest Holocaust denial laws in Europe, alongside other forms of incitement to racial hatred. In light of these circumstances, several French ministers, including Industry Minister Roland Lescure, have reported the posts to the Paris prosecutor, in accordance with a legal obligation for public officials to flag potential criminal activity. Officials characterised the AI-generated content as “manifestly illicit,” asserting that it could amount to racially motivated defamation and a denial of crimes against humanity.
French authorities have escalated the matter by referring the posts to a national police platform specialising in illegal online content and have alerted France’s digital regulator to potential violations of the European Union’s Digital Services Act (DSA). This situation has intensified scrutiny from Brussels, with the European Commission, the EU’s executive body, indicating its engagement with X regarding Grok and labelling some of the chatbot’s outputs as “appalling,” suggesting they violate Europe’s fundamental rights and values.
Additionally, two French human rights organisations, the Ligue des droits de l’Homme and SOS Racisme, have filed a criminal complaint accusing Grok and X of committing crimes against humanity. As of now, X and its AI division, xAI, have not responded to inquiries regarding this matter.
