EU court adviser backs publishers in AI case vs Google

An EU court adviser has expressed support for publishers in a legal dispute with Google over the use of copyrighted articles in the Gemini chatbot's responses. The case highlights the escalating conflict between publishers and major online platforms regarding the utilization of press stories by AI chatbots without appropriate compensation.

This controversy has led to complaints to both EU and national antitrust authorities. Advocate General Maciej Szpunar, advising the Court of Justice of the European Union, is currently mediating in the case, which was initiated by Hungarian website operator Like Company against Google's AI chatbot Gemini. A ruling in favor of the publishers, potentially obliging Google to pay fees for content usage, may provide a significant advantage to online media facing a decline in traffic due to the increasing prevalence of AI-generated news summaries that overshadow traditional search engine results.

Advocate General Szpunar stated that when an LLM-driven chatbot extracts content from a publisher's website, repurposes it to craft an answer, and presents that answer to the user, the publisher's right to reproduction might be violated under certain conditions. Google's Gemini chatbot's responses, as the adviser explained, are considered infringing upon the publishers' right to reproduction when they bypass paywalls, circumvent technical barriers, or replicate the publisher's form of expression.

Copyright laws do not shield the mere facts reported by press publications but only the specific form of expression. Nonetheless, the adviser emphasized that using copyrighted material to train LLMs, AI systems capable of generating human-like responses and summarizing information using extensive text data, is freely permitted under the text and data mining exception.

The court, which adheres to the majority of its advisers' non-binding opinions, is expected to deliver its verdict in the coming months. A Google representative stated that "European law's text and data mining exceptions are crucial for Europe's competitiveness and AI development, and we recognize Advocate General's opinion acknowledging that EU law permits AI training on publicly available information."

They further clarified that "Gemini learns linguistic patterns to create original outputs rather than replicating content, backed by extensive publisher controls that grant rightsholders clear choices and the ability to opt out."