IntelThe next AI conversation needs a better question.
Commercial momentum, an unexpected food-safety use case and a way to put leadership curiosity to work.
Good morning. This week’s Palantir filing makes an AI conversation feel distinctly like a business conversation: commercial revenue grew faster than government revenue.
Put that beside an unexpected food-safety application and a useful question emerges. What important problem has your company left out of its AI discussion? This edit follows the possibilities, then brings the experiment back to a mid-market scale.
The commercial story is getting harder to ignore.
This week: November 3 filing · September reading
Palantir reported global commercial revenue growth of 113% year over year in the third quarter, versus 81% for government. The November 3 filing measures revenue, not deployment counts—but it is a striking comparison for a company so closely associated with government work.
WIRED’s September story about Chipotle working with Palantir on food-safety risks is the companion read. Quality, risk and the information behind an operating decision deserve a place beside the writing assistant in an owner’s imagination.
My read: the enterprise example should expand the question, even when its implementation would be the wrong size for your business. The work starts with finding a consequential problem your own team can investigate.
Borrow the ambition. Design your own experiment.
Selected perspective · illustrative application
Save the World Economic Forum’s case for a mid-market AI moment, then bring it down to a task somebody owns.
An illustrative finance question: could a team use a small, approved set of records to identify why invoices keep getting held up? Compare its findings with the existing review. Have the practitioner explain what it missed and what would actually save time.
That is the translation I care about. Borrow an enterprise example’s ambition, then give your people a question they can answer. A better Tuesday is a more useful starting point than an abstract transformation roadmap.
Two bookmarks. Two different jobs.
Use-case library + learning resource
Two places to explore, with different jobs. Start with the one that answers your team’s next question.
For the leadership agenda: Deloitte.
Browse the AI use-case library outside your own function. A quality example might change a service discussion; an operations idea might expose a question about risk. Bring back the business problem that caught your attention, rather than a list of technologies to buy. Bring a practitioner into the conversation before treating an interesting example as an implementation plan.
For getting more familiar: OpenAI Academy.
Explore the first-party learning resources before the next team conversation. Familiarity gives people something specific to ask about. It also helps separate what a tool can do from what the business should entrust to it. Choose a resource that helps your team ask a better next question.
Make curiosity somebody’s responsibility.
Tim’s proposed leadership practice
Ask each functional leader to bring one application and the problem behind it. Choose one for a small tiger team: the outcome owner, a practitioner and the technical or data support the question requires.
Agree on authorized information, a baseline and a date for a continue/change/stop decision. Count accuracy and review effort alongside speed. That gives curiosity a place to go—and leadership something better to assess than enthusiasm.
Sources & further reading
Palantir / SEC EDGAR · 2026-11-03
WIRED · 2026-09-16
World Economic Forum · 2026-01-19
Deloitte · Undated source
OpenAI · Undated source