

Google’s AI chatbot may face fresh copyright issues in Europe over how it uses news reports to answer queries. Reportedly, Gemini AI uses press stories to answer queries from users. On November 3, 2026, an adviser to the European Union’s top court said that chatbots could break copyright rules if they collect news articles from websites and reuse their content in answers.
This may happen when a chatbot gets around paywalls or repeats parts of an article. The opinion came from a case involving Hungarian website operators Like Company and Google. Like Company claimed that Google used its articles to train Gemini and that the chatbot later shared parts of those articles in its answers.
However, Advocate General Maciej Szpunar said there was no proof that the website’s articles were used to train the AI model. According to his statement, “When an LLM-driven chatbot retrieves a publication from a press publisher's website, reuses that publication to generate an answer, and provides that answer to the user, the press publisher's right may be infringed in certain circumstances.”
AI models learn from large amounts of text from websites, books, articles and other sources. Chatbots also use a method called Retrieval-Augmented Generation, which allows them to search websites in real time, collect fresh details and use them to answer questions.
The EU opinion makes a clear difference between these two activities. AI training helps a model learn from large amounts of text. RAG allows a chatbot to find and reuse existing online content. According to Szpunar, AI training is generally covered by EU rules that allow certain types of text and data use. However, copying protected writing into chatbot answers or getting around paywalls could break copyright rules.
Also Read: Google Expands Scam Protection; Google Messages Gets New AI Shield Gemini-Powered Checks
The case could have a wider effect on AI products. Companies may need to find new ways to give users quick answers without copying protected content. News websites could also get more control over how AI tools use their reports.
The EU court's final ruling is still awaited. However, the opinion has raised an important question for the AI industry: how can chatbots share useful information without taking away the rights of the people who create it?