Ten million users in a few months is not a footnote. It is a signal. Meta's AI assistant, Muse, has crossed that threshold largely on the strength of two use cases: shopping and personal productivity. That tells us something important about where AI adoption is actually happening, not in futuristic labs, but in the everyday tasks of choosing what to buy and getting work done. We think that reality is more transformative than any headline-grabbing demo, and it carries real consequences for anyone building or buying productivity tools today.
Consider what Meta's own numbers imply. The company is simultaneously pursuing a massive Meta explores new stock class to fund its expansive AI investment to bankroll its AI ambitions. That kind of spending makes sense only if the user base converts into long-term engagement. Muse's 10 million users suggest the bet is working, but the nature of that engagement matters. Shopping and productivity are not speculative use cases. They are the core workflows of modern knowledge work and consumer behavior. When an AI assistant becomes the default tool for comparing products or organizing a to-do list, it stops being a novelty and starts being infrastructure. That shift is what we should be watching.
The practical takeaway for our readers is direct: if you manage data or build tools for people who do, the era of the passive spreadsheet is ending. AI assistants like Muse are not just answering questions, they are making decisions about what to recommend and how to prioritize. That creates a new set of responsibilities. As we recently saw with An AI-Discovered Vulnerability in PHP Shows the Risk of Progress, the same technology that accelerates progress can also introduce blind spots. A shopping assistant that optimizes for speed might steer users toward lower-quality products. A productivity assistant that prioritizes urgency might bury important long-term tasks. The tool is powerful, but the user still needs to understand its logic.
This is where the contrast with legacy tools becomes sharp. Traditional spreadsheets force you to build the logic yourself. AI-native tools like Muse embed that logic in the interface, which makes them faster but also more opaque. The question is not whether these tools will replace spreadsheets, they already are for millions of people. The question is whether users will demand transparency alongside speed. Blackstone's AI head recently argued that Blackstone on AI buildout debt: Why the worry is overblown market jitters about AI infrastructure debt are overblown, because the underlying demand is real. We agree on the demand side. But demand for convenience does not automatically translate into demand for rigor. The metric to watch over the next year is not user count. It is how many of those 10 million users can articulate why Muse gave them the answer it did.