A survey of US adults: only ~40% report using AI for work and 60% say they use AI to find information at least some of the time, rising to 74% of under-30s
Most U.S. adults say they use artificial intelligence to search for information, but fewer are using it for work, drafting email or shopping. Bluesky: @karlbode.com Bluesky: Karl Bode / @karlbode.com : I've got all sorts of tech bouncing around my house, I'm a consistent early adopter and tinkerer, and the only time I've found AI integration useful is in some vehicle hands-free driving situations, and even then it fucks up a lot — mostly it's clumsy forced integration that provides no value to me [embedded post]
Context & Ripple Effects
Earlier polling showed awareness of major chatbots exceeded actual use, with younger adults leading adoption. This survey indicates that gap has shifted most clearly toward information-seeking, while work use remains a separate, less-penetrated behavior.
The finding also fits earlier evidence that adoption has raced ahead of demonstrated productivity gains: people may incorporate AI first where it can supplement an existing search habit rather than replace a consequential workflow.
First-order effects
- AI providers and search products gain evidence that information lookup is their broadest consumer entry point, especially among adults under 30.
- The gap between information use and work use makes consumer search-oriented experiences a more immediate priority than email, shopping, or generalized workplace integrations.
Second-order effects
- Search incumbents and AI assistants face sharper pressure to make answers reliable and useful enough to retain information-seeking users; distribution inside existing search surfaces becomes consequential.
- Workplace AI vendors will need to show workflow-specific value rather than infer business adoption from broad public familiarity, consistent with separate surveys showing work use is growing but remains uneven.
Third-order effects
- If information seeking remains the dominant use case, AI may reshape the discovery layer before it materially reshapes how most people produce work—concentrating competition around answer quality, trust, and default access.
- A sustained split between consumer experimentation and workplace use would make measurable task-level value, not headline adoption, the key test for enterprise AI deployment.
The trend: AI adoption is moving first through low-friction information discovery, while deeper workflow adoption depends on proving dependable value in specific tasks.