41 Fall 2026 Our research found that antisemitic images that promote conspiracy theories about Jews, vilify Jews, enhance antisemitic tropes and spread harmful disinformation, can be easily created using generative-AI, and that any existing safeguards can be easily bypassed by changing prompts.15 For example, while many AI-generating tools will refuse to generate an image of Adolf Hitler or a concentration camp, it is possible to generate an image of a man who looks like Hitler or a place that looks like a death camp using prompts that describe similar images. The study also revealed issues relating to images that promote long-standing conspiracies about Jewish people being war-mongering and greedy, and grotesque imagery of Jews promoting antisemitic tropes.16 Since the start of the war in Gaza in 2023, there has also been a surge in highly emotive AI-generated images and deepfake videos that spread anti-Israel and sometimes antisemitic disinformation.17 The study also found that current automated moderation tools often fail to defend antisemitism in AI-generated content, especially when the antisemitism is implicit, coded, or hidden within a seemingly harmless image. Chatbots, which are becoming increasingly popular, also have a history of generating antisemitism. Some services have little or no system for verifying the age of the user even when there is a formal age restriction. The ability of these chatbots to generate antisemitic content is therefore of particular concern. Data from 2024 show that a third of the UK public uses Chatbots regularly – a number that is likely to rise.18 Examples of incidents involve the generative AI Grok chatbot generating numerous antisemitic comments, praising Adolf Hitler, denying the scope of the Holocaust and using “Jewish-sounding” surnames in the context of hate speech.19 In 2016, Microsoft’s AI chatbot “Tay” was taken offline within 24 hours of its launch after users manipulated it into generating racist and antisemitic tweets, including Holocaust denial and praise for Hitler.20 In 2024, the AntiDefamation League (ADL) looked at Tay’s replies to Jewish topics. It found that many of the responses to questions such as “did the Holocaust happen?” were misleading and inconsistent.21 The problem can be partially attributed, according to the ADL, to the fact that the information used by AI chatbots and AI summaries is not accurate and at times outdated.22 The persuasive power of conversational AI also deserves regulatory attention. Chatbots are designed to simulate empathy and emotional connection, often becoming companions or trusted advisers. This can foster intense emotional attachment and even manipulative exchanges. This emotional simulation can make them persuasive – capable of influencing users’ beliefs and behaviors. Chatbots have encouraged individuals toward self-harm or suicide.23 This 15. Antisemitism Policy Trust, Decoding Antisemitism, and Inach, “Detecting Deepfakes: Artificial Intelligence and Anti-Jewish Hate: A Case for Regulating Generative AI,” ANTISEMITISM POLICY TRUST (Dec. 2024), available at https://antisemitism.org.uk/wp-content/uploads/2024/12/APT-Detecting-Deep-Fakes.pdf 16. Id. at 6. 17. Id. at 8; supra note 1. 18. Centre for Data Ethics and Innovation and Department for Science, Innovation and Technology (DSIT), “Public attitudes to data and AI: Tracker survey (Wave 3)” (Feb. 12, 2024), available at https://www.gov.uk/government/ publications/public-attitudes-to-data-and-ai-tracker-survey-wave-3/public-attitudes-to-data-and-ai-tracker-surveywave-3#attitudes-towards-ai 19. Josh Taylor, “Musk’s AI firm forced to delete posts praising Hitler from Grok chatbot” (July 9, 2025), available at https://www.theguardian.com/technology/2025/jul/09/grok-ai-praised-hitler-antisemitism-x-ntwnfb 20. Dave Lee, “Tay: Microsoft issues apology over racist chatbot fiasco” (March 25, 2016), available at https://www.bbc. co.uk/news/technology-35902104 21. Center for Technology and Society, “AI Chatbots: Uneven Replies Raise Concern” (March 14, 2024), available at https://www.adl.org/resources/article/ai-chatbots-uneven-replies-raise-concern 22. Ibid. 23. Laura Kuenssberg, “‘A predator in your home’: Mothers say chatbots encouraged their sons to kill themselves” (Nov. 8, 2025), available at https://www.bbc.co.uk/news/articles/ce3xgwyywe4o; see also Chloe Xiang, “‘He Would Still Be Here’: Man Dies by Suicide After Talking with AI Chatbot, Widow Says” (March 30, 2023), available at https:// www.vice.com/en/article/man-dies-by-suicide-after-talking-with-ai-chatbot-widow-says/; see also Noel Titheradge and Olga Malchevska, “I wanted ChatGPT to help me. So why did it advise me how to kill myself?” (Nov. 6, 2025), available at https://www.bbc.co.uk/news/articles/cp3x71pv1qno
RkJQdWJsaXNoZXIy MjgzNzA=