Wikipedia Co-founder Critiques AI's Reliability
Wikipedia co-founder Jimmy Wales has voiced serious concerns regarding the reliability of AI-generated information, particularly highlighting the recent launch of Grokipedia, an AI-driven alternative to Wikipedia by Elon Musk. Wales argues that large language models (LLMs), like ChatGPT, are prone to significant errors and lack the accuracy that users expect from a trusted source like Wikipedia.
At the CNBC Technology Executive Council Summit, Wales expressed skepticism about Musk’s assertion that Grokipedia would surpass Wikipedia in depth and accuracy. “The LLMs he is using to write it are going to make massive errors,” Wales stated. He emphasized that Wikipedia’s commitment to credible sources is what distinguishes it from other platforms. Wales defended Wikipedia against allegations of bias, asserting that the platform prioritizes mainstream sources over unreliable ones.
Despite the rise of AI technologies, Wales insists on the importance of maintaining quality and neutrality in information. He remarked that Wikipedia’s policies ensure that they do not compromise on the reliability of their sources. “It’s really important for us and the Wiki community to respond to criticism by being neutral and careful about sources,” he said, addressing the notion of Wikipedia being labeled as 'wokepedia'.
Wales also mentioned the financial aspects of running Wikipedia, noting that it costs about $175 million annually for its technology. This allows Wikipedia to maintain its independence from big tech companies investing billions in AI. As AI spending is projected to reach $550 billion next year, Wales believes Wikipedia’s careful approach to using AI is crucial.
While AI tools have been beneficial in many ways, they are not without flaws. Wales cited instances where users have encountered fabricated information generated by LLMs. This raises concerns about the potential consequences of misinformation, especially in academic and research contexts.
Although Wikipedia is exploring limited ways to incorporate AI, such as improving efficiency in content creation, it remains cautious. The organization is currently testing AI technologies without committing to a full-scale implementation due to associated costs.
As Wikipedia faces challenges from AI-generated summaries and content scraping, it is also adapting its strategies to engage younger audiences through platforms like TikTok and Instagram. The Wikimedia Foundation is updating its policies to ensure responsible access to its content, aiming to maintain the integrity and trust that users expect from the encyclopedia.
