AI writing faces restrictions under new Wikipedia rules

Wikipedia has published a new policy prohibiting editors from using artificial intelligence (AI) to generate or rewrite article content.

Wikipedia—the popular free online encyclopedia—is restricting the use of AI on its platform, following its recent policy banning large language models (LLMs) from writing or rewriting articles.

“Text generated by large language models (LLMs) such as ChatGPT, Gemini, DeepSeek, etc. often violates several of Wikipedia’s core content policies,” Wikipedia’s policy reads.

Despite the new rules, the digital encyclopedia cited two exemptions for copywriting and translation. The policy still allows the use of AI tools, with caution, to suggest basic copy edits to the editors’ own writing, provided the LLM adds new information. In addition, the use of LLMs to translate articles is still allowed, but editors must first follow their guidelines.

“Editors are permitted to use LLMs to suggest basic copyedits to their own writing, and to incorporate some of them after human review, provided the LLM does not introduce content of its own,” it said. “Editors are permitted to use LLMs to translate articles from another language’s Wikipedia into the English Wikipedia, but must follow the guidance laid out at Wikipedia:LLM-assisted translation.”

Although Wikipedia didn’t mention any penalties for the use of AI-generated content, their LLM guidelines stated that feedback that LLM or AI obviously makes “may be struck or collapsed. Repeating such misuse forms a pattern of disruptive editing, and may lead to a block or ban.”

Following the new policy announcement, a spokesperson for the Wikimedia Foundation—the non-profit organization behind Wikipedia—told one news outlet that it does not determine the platform’s editorial policies and guidelines: “Wikipedia’s strength has been and always will be its human-centered, volunteer-driven model.”

In October, the Wikimedia Foundation revealed that the number of humans visiting its site fell by 8% year-over-year as search engines and AI chatbots continue to provide direct answers on their platforms.

“Bots and crawlers have continued to have a significant impact on traffic data to the Wikimedia projects,” Wikimedia said on a blog post. “We are seeing declines in human pageviews on Wikipedia over the past few months, amounting to a decrease of roughly 8% as compared to the same months in 2024. We believe that these declines reflect the impact of generative AI and social media on how people seek information.”

However, just this January, the Wikimedia Foundation announced partnerships with companies exploring AI, including Microsoft (NASDAQ: MSFT), Google (NASDAQ: GOOGL), Amazon (NASDAQ: AMZN), and Meta (NASDAQ: META). These agreements allow these companies to use Wikipedia content through its Enterprise product, a commercial service that facilitates the extensive reuse of the platform’s materials.

How can we make AI and its creators accountable?

With the rising use of generative AI comes its darker side: deepfake or hyper-realistic videos or audio created using AI that could generate fake news.

Across the world, different jurisdictions have been seeking solutions to bring much-needed accountability to creators and users of AI applications. While some regions, such as the European Union, have moved swiftly to introduce regulations, such as the AI Act, some countries have yet to outline a clear strategy. More concerning is that it remains unclear whether regulations can be effectively enforced.

There might be a silver lining, as blockchain can strengthen AI accountability by creating a tamper-proof, timestamped record of how AI is built, trained, and used. By logging key aspects such as data inputs, model changes, and deployment decisions into an immutable public ledger, we can trace the harm caused by deepfakes or faulty LLMs.

“Blockchain can play an incredibly important role by just journaling the truth and putting up a journal that’s unalterable in the world that really lays out what happened,” explained Sebastian Thrun, Founder of Udacity, at the London Blockchain Conference 2025.

Blockchain technology essentially “journals the truth” by having a system that immutably logs records. Meaning, with blockchain, every single AI data source would have a transparent history for users, regulators, and other stakeholders.

In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.

Watch: AI is a double-edged sword