Microsoft has heavily invested in integrating Copilot across its product suite, positioning it as a crucial AI coworker. However, its Terms of Use contradict this, stating that Copilot is “for entertainment purposes only” and advising against relying on it for important advice. This discrepancy has gained attention as less than one in 30 potential users actually pay for Copilot.
Within Microsoft’s Terms of Use for Copilot, a section updated in October 2025 and noticed in April 2026 declares: “Copilot is for entertainment purposes only. It can make mistakes, and it may not work as intended. Don’t rely on Copilot for important advice. Use Copilot at your own risk.” Additional terms indicate no warranty is made about Copilot, users are responsible for its outputs, and the enterprise Microsoft 365 Copilot is excluded from these terms.
While publicly promoting Copilot as a tool enhancing productivity, and spending $80 billion on AI in fiscal 2025 with a significant investment in OpenAI, Microsoft’s pricing doesn’t align with labeling the tool as entertainment-only. Legal experts interpret the disclaimer as a strategy to limit liability, a contrast from marketing that suggests reliability.
Released adoption data reveals Copilot’s limited traction—only 3.3% of eligible Microsoft 365 users are paid subscribers, amounting to 15 million out of 450 million. Accuracy concerns persist, with a declining Net Promoter Score and significant user distrust. Copilot’s erroneous outputs, like false criminal allegations and misinformation, exacerbate reliability issues.
Microsoft CEO Satya Nadella is prioritizing AI development, shifting responsibilities to directly manage this area. The company’s recent introduction of proprietary AI models signals a move to reduce reliance on external models. Internally, Copilot’s limitations are acknowledged, aligning with the Terms’ cautious language amidst rising expectations for AI to be both dependable and accountable. Despite these challenges, Copilot is not alone in facing scrutiny over AI inaccuracies, as similar issues plague competitors like ChatGPT. Microsoft’s significant investment and marketing assert AI’s readiness as a workplace necessity, yet its Terms suggest a more cautious approach. As competitors vie for AI reliability, bridging the gap between disclaimers and true reliability becomes an industry focal point.