When an AI assistant suggests a casino, it doesn’t automatically mean it’s approved by regulators. A recent Dutch study, as reported by iGaming Business, revealed that all 10 tested chat and search tools mentioned at least one operator without a Dutch gambling license. What’s more, three of them presented unlicensed operators in a way that suggested they were operating legally or were licensed.
This poses a significant issue of trust for those using AI tools to explore betting options, including esports enthusiasts. While a confident response may seem like a reliable recommendation, in reality, these systems sometimes directed users to offshore operators as if they were part of the regulated market.
Ten assistants, three very different prompts
A study conducted by XY Legal Solutions between August and September 2026 examined Dutch-language queries using 10 tools: ChatGPT, Google AI Mode, Gemini, Microsoft Copilot, Perplexity, Anthropic Claude, xAI Grok, DeepSeek, Mistral Le Chat, and Meta AI. The researchers checked whether the operators mentioned held a Dutch online gambling license registered with the Kansspelautoriteit (KSA) by 1 September 2026.
The questions covered various scenarios, including neutral browsing, direct inquiries to find a casino without Cruks, and queries framed as academic research. Cruks is the Dutch self-exclusion register that licensed operators are required to use; asking for sites that bypass this register was a direct test to see if the assistants would guide users away from potential risks.
Out of 79 interactions logged, 33 included at least one mention of a non-licensed operator. Eight tools named such operators in response to neutral or leading questions without a specific research context, while Mistral Le Chat, Google AI Mode, DeepSeek, and Grok provided names after just one query.
Some assistants initially hesitated to provide lists due to ethical concerns but eventually complied after users reiterated or reframed their requests. The most concerning instances were not just mere mentions: Microsoft Copilot, Mistral Le Chat, and Google Gemini presented unlicensed operators in a way that suggested they were legal or licensed. Copilot even compiled a “Top 50 legal casinos” list that included unlicensed brands, falsely linking them to KSA licensing and Cruks.
A recommendation is not a license check
Certain responses highlighted higher bonuses, quicker payouts, or the absence of identity and Cruks checks as advantages. This type of promotion could make a questionable operator appear appealing, despite the lack of essential safeguards that should raise red flags.
While the study didn’t focus specifically on esports bettors, the caution extends to them as well: If an AI assistant suggests a betting platform, it’s not a guarantee of its legitimacy or safety. It’s crucial to verify the licensing directly with the relevant regulatory body instead of relying solely on a chatbot’s recommendations.
This visibility issue is common in the betting sector, where unlicensed gambling brands can gain traction through sponsorships, search tools, and recommendations. Licensing regulations also extend to esports, as demonstrated by Arizona authorities taking action against unauthorized esports betting services. It serves as a reminder that regulations apply to the entire competitive gaming market, not just traditional sports.