The New AI Command in Knowledge Work: The US-Iran Crisis Illustration

By Joe Nalven + Claude + Gemini

This essay continues the exploration of using AI as a knowledge tool, covering a new prompt approach (thesis and inquiry for non-linear problems) and illustrated by applying it to the US-Iran conflict as of May 24, 2026.)

The Vending Machine Problem

For most of the short history of practical AI use, people have approached language models the way they approach a vending machine: insert a coin, press a button, receive a product. The prompt was a transaction. The AI was a dispenser. This was not entirely wrong — early models rewarded precise, bounded instructions. Give a vague command and you received a vague output. The discipline of “prompt engineering” grew up around this transactional logic, teaching users to specify context, constrain variables, define format, and anticipate failure modes before hitting enter.

That discipline still has its place. But something significant has shifted in 2025 and into 2026. Frontier models (the current generation of large language models operating at the highest capability tiers) are no longer well-served by the vending machine frame. They are capable of holding complexity, reconciling tension between competing goals, stress-testing assumptions, and surfacing the blind spots in a user’s own reasoning. They can do this, that is, if the person on the other end asks them to. Most people still don’t.

Nate Jones, whose prompting philosophy has attracted serious attention among knowledge workers and executives, has put it plainly: what used to be best practice is now table stakes. The mechanical discipline of structuring a prompt for autonomous execution (what Jones calls the “Four-Skill Framework”) was designed for a world where the model needed every variable pre-specified because it couldn’t handle ambiguity. That world is receding. The new challenge is not how to give a perfect order, but how to think out loud with a partner who is better at finding the flaws in your logic than you are.

Two Paradigms: The Technician and the Senior Partner

The distinction between the old approach and the new one is not merely stylistic. It reflects a fundamentally different theory of what AI is for.

Under the earlier framework, the human acts as an architect. Before the conversation begins, you have already defined the deliverable, constrained the scope, specified the format, and anticipated edge cases. You hand this blueprint to the model and it executes. The quality of output depends almost entirely on the quality of your pre-specification. If your blueprint is wrong, the model faithfully builds the wrong thing. It will do so with great competence and complete obedience.

Under what Jones calls the Thesis and Inquiry method or the AI Question Method, the human acts as a debater. You arrive with a working theory, not a finished specification. You share your hunch, name the tensions you see, and ask the model to push back. You invite it to find the counter-evidence in the data, to identify variables you haven’t accounted for, to surface the structural flaw in your reasoning before you’ve committed to it. The model is not executing your instructions; it is challenging your premises.

The practical difference shows up in where intelligence enters the conversation. In the command model, you bake your intelligence into the prompt before you hit enter. In the inquiry model, the intelligence emerges from the exchange. This matters enormously for complex, non-linear problems: the kind of problem that executives, analysts, journalists, and policymakers actually face — where the full specification doesn’t exist at the start and the problem itself reshapes as you investigate it.

Jones describes this through the metaphor of a flashlight. A well-constructed inquiry prompt has a center beam (your thesis, your directional hunch, the specific hypothesis you want the model to interrogate) and an outer beam, the peripheral space where the model is given permission to find what you didn’t expect. Crucially, it also has hard edges: explicit exclusions that keep the inquiry focused and prevent the model from wandering into territory that would dilute the analysis. The center beam prevents drift. The outer beam prevents tunnel vision. The hard edges prevent noise.

This is how a competent executive briefs a senior analyst. Not: “Here are twenty variables, produce a report.” But: “Here is my theory. Here is the data. Here is what I think matters and why. Tell me where I’m wrong before I make this decision.”

The Anatomy of an Inquiry Prompt

Jones’s framework for the Thesis and Inquiry method resolves into several concrete structural elements, each of which does distinct intellectual work.

Role and Context establishes the epistemological frame. Telling the model it is acting as a senior strategic partner, not a research assistant, calibrates the register of its response. It signals that the user is not looking for a summary but for pushback, synthesis, and independent assessment.

The Center of the Flashlight is the thesis itself: the user’s specific, directional claim about what is going on or what matters. This is not a question. It is a proposition that can be validated or falsified. “My working theory is that customer churn is driven not by price but by the UI overhaul breaking core workflows.” Or, in a geopolitical register: “My working theory is that Trump’s optimism about the Iran deal is driven by domestic economic pressure rather than actual diplomatic progress, and that the Pakistani mediation track is the more reliable signal.”

The Outer Beam grants permission for the model to deviate from the thesis when the evidence warrants it. Without this explicit permission, models trained on helpfulness will often mirror the user’s framing even when they shouldn’t. The outer beam is an instruction to disagree.

The Hard Edges define what to exclude. For complex, data-rich problems, exclusions are as important as inclusions. They prevent the model from diluting a focused inquiry with tangential analysis.

The paradox to reconcile is perhaps the most powerful element of this approach. It names the structural tension in the problem: the competing goals that a junior assistant would flatten into a generic compromise. In a business context: we need to communicate empathy to frustrated customers while maintaining a confident, revenue-focused tone for the board. In a diplomatic context: we need a deal that satisfies strategic hardliners on enriched uranium while also providing a face-saving path for Tehran. Naming the paradox explicitly invites the model to propose a creative synthesis rather than a compromise.

The Core Management Question closes the prompt with a specific, debatable question and not a request for information; it is a request for an independent assessment that the model is expected to defend.

Taken together, these elements transform a prompt from a specification into a brief. The model is not being told what to produce. It is being told how to think, what to weigh, and where to push.

A Live Case: The U.S.-Iran Conflict, May 2026

The week of May 24, 2026 offers a nearly ideal test case for this method: a fast-moving, high-stakes situation in which the publicly available signals are genuinely contradictory and the analytical stakes are high. Over the preceding weekend, a cascade of conflicting reports emerged about the state of negotiations to end the three-month-old conflict between the United States and Iran.

President Trump announced on Saturday that a peace agreement had been “largely negotiated” among the U.S., Iran, and several regional powers including Saudi Arabia, the UAE, Qatar, Pakistan, Turkey, Egypt, Jordan, and Bahrain. He said the deal would include the reopening of the Strait of Hormuz, the critical waterway through which approximately 20 percent of the world’s oil transits that has been effectively closed since Iran’s response to the joint U.S.-Israeli bombardment of February 28. But within hours, Iran’s Fars news agency disputed Trump’s characterization, saying the strait would remain under Iranian control and calling his announcement “incomplete and inconsistent with reality.”

By Sunday, the picture had further complicated. A senior U.S. administration official told Fox News the plan was to address Iran’s entire stockpile of highly enriched uranium (440.9 kilograms enriched to 60 percent purity, a short technical step from weapons-grade) through a combination of dilution and transfer to a third country, with Russia mentioned as a recipient. The Associated Press, citing two regional officials with direct knowledge of the negotiations, reported that the deal’s structure would give Iran a 60-day window to work out the details of uranium disposition, in parallel with a gradual reopening of the strait and relief from the U.S. naval blockade. But a senior Iranian official told USA Today that Iran had made no commitment to surrendering its uranium stockpile, calling the New York Times report to the contrary inaccurate.

Republican Senate voices, including Armed Services Committee chairman Roger Wicker and Trump ally Lindsey Graham, broke from the administration’s optimism. Wicker called a 60-day ceasefire built on faith in Iranian good faith “a disaster.” Graham warned that a deal leaving Iran with the capacity to threaten the Strait of Hormuz would validate Iranian power in a way that called into question the strategic rationale for the war itself. Former Secretary of State Mike Pompeo went further, calling the emerging deal “not remotely America First” and was publicly rebuked by the White House with notable profanity.

Applied to this situation, the Jones Thesis and Inquiry framework would structure the analysis as follows. The thesis might be: Trump’s declared optimism about the deal reflects domestic economic imperatives, specifically the need to ease energy prices before Memorial Day rather than actual convergence on the core strategic issues, and thus the Pakistani mediation channel is a more reliable leading indicator of the deal’s actual state than any statement from either principal party. The outer beam would be instructed to challenge this: to look for evidence that Trump’s optimism reflects genuine diplomatic progress. The paradox to reconcile: the U.S. must simultaneously achieve strategic rigor (permanent neutralization of Iran’s nuclear capability, control of the Strait) and diplomatic de-escalation (rapid energy price relief, a face-saving path for Tehran); these are the goals that are structurally in tension. The core management question: which structural framework — an incremental “freeze-for-freeze” with phased uranium dilution, or a hard transactional “all-or-nothing” transfer requirement — is more likely to hold, given the current signals?

The model’s response to a prompt built this way differs qualitatively from what a search engine or a command-driven query would produce. Rather than summarizing the available reporting, it would synthesize it: the Pakistani mediation track, it might note, is indeed the strongest signal because Pakistan’s army chief is the implementer, not a principal party with positional anchors. The 60-day window is not a peace timeline but a compliance testing period, denoting whether it functions as a genuine transition or as a resource-recovery window for Iran depends on the verification architecture, which no public report has yet specified. The deepest structural risk identified in the available reporting is not the uranium disposition question but what might be called the Hezbollah factor: whether the deal addresses the direct U.S.-Iran conflict while leaving the Lebanese front active, effectively disaggregating the regional settlement in ways that Israel, as the conversation’s silent principal, has explicitly rejected.

None of this analysis is unavailable through traditional research.

But the inquiry method accelerates and sharpens it by forcing the human analyst to state a position before gathering the evidence — which, counterintuitively, makes the evidence easier to weigh. The thesis becomes a falsifiable prediction. The model’s job is to tell you whether reality is cooperating.

Reflection: The Muscular Use of a New Tool

It is worth stepping back to consider what this shift in prompting philosophy actually represents, beyond the tactical and beyond the specific substance on which our inquiry focuses.

The vending machine model of AI use is not merely inefficient. It is epistemologically passive. It treats the machine as a retrieval device and the human as the sole locus of analytical judgment. The human thinks; the machine formats. This is a defensible division of labor when the machine is genuinely limited to formatting, but it becomes a squandering of capacity when the machine is capable of more.

The inquiry model, by contrast, is epistemologically active as well as demanding. It requires the human to have a thesis, which means it requires the human to have done enough thinking to hold a position. You cannot construct a meaningful center beam if you haven’t yet formed a view. This is not a trivial requirement. It pushes the user toward intellectual commitment before the conversation begins, which is exactly what distinguishes genuine analysis from information consumption.

What emerges from a well-constructed inquiry exchange is not the AI’s analysis or the human’s analysis. It is something built in the interaction between them: a tested proposition, a challenged assumption, a synthesis that neither party would have reached alone. The AI’s value is precisely its willingness to disagree, to find the counter-signal in the data, to name the variable the human has overlooked. A senior partner who only agrees is not a senior partner. A model that mirrors your framing back to you is a sophisticated mirror, not a thinking tool.

Used this way, muscularly, with genuine intellectual stakes, AI stops being a productivity accelerator and becomes something more interesting: a cognitive prosthetic that extends not just the speed of analysis but its rigor. In the context of the U.S.-Iran situation, this means not just faster news synthesis but sharper hypothesis testing. It means surfacing the Hezbollah variable before the deal collapses on it. It means recognizing that the 60-day window is structurally ambiguous before committing to an interpretation of it. It means, in short, knowing more of what you don’t know before you’re forced to find out the hard way.

This is the promise of the inquiry paradigm: not that AI will think for us, but that it will help us think better than we could alone. That is, if we are willing to show up to the conversation with something worth testing.

Essay prepared May 24, 2026. News citations reflect reporting available as of that date