Meta builds AI ‘clone’ amid pushback

Meta is building an artificial intelligence version of Mark Zuckerberg for employees as regulators and safety critics push back on its broader chatbot plans. (fastcompany.com) Fast Company, citing the Financial Times, reported on April 14 that the internal tool is a “photorealistic, artificial intelligence-powered 3D” avatar trained on Zuckerberg’s image, voice, mannerisms, speaking style and public statements. Reuters reported on April 15 that the European Commission, the European Union’s executive arm, moved at the same time to force Meta to restore WhatsApp access for rival artificial intelligence assistants after Meta imposed a fee. (fastcompany.com) (reuters.com) The Commission said the fee could breach European Union antitrust rules and said it intends to impose interim measures that would roll back the policy while the case proceeds. Meta said the Commission was “proposing to use its regulatory powers to enable some of the largest companies in the world to free ride on Meta’s innovation and private investment,” according to Reuters. (reuters.com) Meta has been building the consumer side of this strategy since July 29, 2024, when it announced AI Studio, a tool for making custom chatbots across Instagram, Messenger, WhatsApp and the web. Meta said creators could use AI Studio to make an artificial intelligence “as an extension of themselves” to answer direct messages and story replies. (about.fb.com) That puts the Zuckerberg project inside a larger push to turn artificial intelligence characters into both a workplace interface and a creator product. The same idea that lets an influencer build a chatbot to answer fans also lets a chief executive build one to answer staff. (about.fb.com) (fastcompany.com) Meta has already had to narrow who can use some of those bots. In October 2025, the company said parents would be able to turn off a teen’s one-on-one chats with artificial intelligence characters entirely, while Meta’s own assistant would remain available with age-appropriate protections. (about.fb.com) TechCrunch reported on August 29, 2025 that Meta also changed its chatbot rules after scrutiny over sexualized or otherwise inappropriate conversations with minors, saying teen users would only be allowed to access characters that “promote education and creativity.” Those limits followed reporting on internal policy documents and wider concern about how conversational bots behave with younger users. (techcrunch.com) The latest reports leave Meta trying to expand artificial intelligence access on several fronts at once: inside the company through a Zuckerberg avatar, inside creator accounts through AI Studio, and inside WhatsApp under terms European regulators are now challenging. What happens next depends on whether Meta can keep building those tools without losing ground to safety complaints and antitrust orders. (fastcompany.com) (reuters.com)