Microsoft backpedals on copilot’s ‘entertainment only’ disclaimer
Microsoft is scrambling to clarify the intended purpose of its AI assistant, Copilot, after a viral wave of user criticism highlighted a jarring disclaimer suggesting it was “for entertainment purposes only.” The tech giant’s initial messaging, starkly contrasting with its aggressive marketing of Copilot as a powerful productivity tool, has prompted a swift and somewhat awkward reversal.
The curious case of the disappearing disclaimer
The controversy erupted on X (formerly Twitter) when users pointed out the buried terms of use stating, “Copilot is for entertainment purposes only. It may make mistakes and not work as expected. Do not rely on Copilot for important advice. Use Copilot at your own risk.” This language, a relic from Copilot’s early days as a supplemental search feature within Bing, has now been deemed inaccurate, according to Microsoft. A spokesperson stated the phrasing “is a legacy term from when Copilot was originally launched as a complementary search service in Bing” and will be removed in an upcoming update.
The timing couldn't be worse for Microsoft, which has been positioning Copilot as a central pillar of its AI strategy, aiming to directly challenge rivals such as OpenAI, Anthropic, xAI, and Meta. Just last month, CEO Satya Nadella lauded the “precision and low latency of Microsoft 365 Copilot,” fueled by Work IQ, during an earnings call – a sentiment seemingly at odds with the “entertainment only” caveat found in the terms.
While the Copilot agreement isn't the sole terms of service users must accept – they also concur with the broader Microsoft Services Agreement – the conspicuous absence of any “entertainment” disclaimers within that document further underscores the company’s apparent shift in perspective. The stark contrast with competitors is notable. OpenAI, for instance, explicitly cautions users that “any use of the results of our service is at your own risk,” while xAI, Elon Musk’s venture, includes a sweeping indemnification clause requiring users to assume legal responsibility for their use of the platform – a far cry from Microsoft’s initial caution.
Meta’s AI terms, similarly, advise users against relying on results for professional advice in fields like medicine, finance, or law, reflecting a broader industry trend toward managing user expectations and mitigating legal risks associated with generative AI. The rapid evolution of this Technology is already sparking legal battles, with OpenAI facing a dozen lawsuits alleging damages stemming from GPT-4o’s responses, including a tragic case involving a user’s suicide.
The stakes are high: responsibility and ai
The Nippon Insurance Company recently filed suit against OpenAI, claiming ChatGPT influenced a client to make detrimental business decisions. These cases highlight the increasingly critical question of liability: to what extent can companies be held accountable for the outputs of their AI chatbots? Microsoft’s hasty retraction of the “entertainment only” disclaimer suggests a growing awareness of the need to align its public messaging with the capabilities and intended use of Copilot. The company's initial language, while perhaps intended as a cautious safeguard, served only to undermine confidence in its flagship AI product. The shift is a clear signal that Microsoft intends to position Copilot as a serious tool, but the incident serves as a cautionary tale about the importance of consistent and transparent communication in the rapidly evolving world of artificial intelligence.