Stop asking ChatGPT to tell you a joke because it’s not playing anymore. OpenAI’s research points to a 'capability overhang,' noting that power users engage with roughly seven times more advanced AI features than typical users. Moreover, some countries show 3x higher per-person adoption of sophisticated tools. On the surface, it suggests a gap between what AI can do and how people use it, but it’s unclear whether it’s true untapped potential or simply that users haven’t found compelling applications for the tech yet. Across countries, leading adopters are in both high-income and emerging markets, proving that adoption is driven by factors beyond infrastructure or wealth, like workflow fit or organizational priorities. One question feels unanswered: if AI capabilities continue to grow faster than practical adoption, could it translate into meaningful productivity gains, or will users just keep searching for what actually works for them? The gap between power users and typical users is real enough to rethink how AI might scale across industries and economies. TLDR; AI’s got the brains, but it’s still waiting for bodies.