Are you building a team of AI passengers coasting toward irrelevance?
with Greg Shove, Section
Are you building a team of AI passengers coasting toward irrelevance?
Show Notes
Greg Shove is a 7-time founder with roughly $250M in exits and leadership roles at Apple, AOL, and Sun Microsystems. Today he runs Section , an enterprise AI adoption platform that combines software and services to help organizations move from passive AI awareness to active, high-leverage use. In this conversation, he makes a sharp distinction between companies that have AI tools and companies that have AI cultures - and explains why most enterprises are building teams of passengers when they desperately need drivers.
Greg's vantage point is rare: he has seen multiple technology transitions from the inside of major institutions and from the founding seat of his own companies. That experience shapes a perspective that is both urgent and patient - urgent about the stakes of falling behind on AI, patient about the human change management required to actually move the needle.
Drivers vs. Passengers: The Only Distinction That Matters
The central frame of Greg's leadership philosophy at Section is simple: every person on your team is either an AI Driver or an AI Passenger. Drivers own the task, own the decision, and own the conviction behind their output. When they use AI, they interrogate it, push back against it, and take responsibility for the result. Passengers let AI do the thinking and present the output as if it is their own conclusion - "this is what AI thinks" rather than "this is what I think, informed by AI."
Greg's policy on passengers is a one-strike rule. Not because he is harsh, but because a passenger in a knowledge work environment is a liability disguised as a contributor. The organizational cost of normalizing deferred thinking - of building a culture where AI output substitutes for human judgment - compounds invisibly until it collapses visibly.
The 10% Adoption Ceiling Every Enterprise Hits
One of the most operationally useful insights in this conversation: every organization can reach approximately 10% AI adoption through self-motivated, intrinsically driven employees. These are the people who would have figured out Excel pivot tables on their own in 1997. They do not need permission or training - they seek it. The problem is that 10% is also the ceiling without active intervention.
Getting from 10% to 30–50% requires deliberate change management - leadership modeling behavior, structured learning, accountability mechanisms, and a cultural frame that makes AI fluency an expectation rather than a differentiator. This is the gap Section is built to close. Most enterprise AI "success stories" are actually 10% stories told as if they represent the whole organization.
Capability Boundaries: Punching Above Your Weight Class
AI does not just make work faster - it shifts capability boundaries. A team of 32 at Section does work that previously required 45. That is not a headcount story; it is a quality-of-output story. The same 32 people are producing better strategy, better content, better customer interactions - not because they are working harder, but because AI has extended what each person can reach.
For founders, this reframes the build vs. hire decision. Before adding a function, the question is now: what is the AI-extended capability of the people we already have? In many cases, the answer is: more than you assumed. The 45-person equivalent is not a curiosity - it is a structural advantage that compounds as adoption deepens.
AI as Thought Partner: The Underutilized Use Case
Greg identifies AI as a thought partner - specifically for confirming decisions and surfacing blind spots - as one of the most undervalued applications in business today. Not AI as writer, not AI as researcher, but AI as the entity that asks you the uncomfortable question before you make the call you are already leaning toward.
This use case requires a particular relationship with the tool: one of genuine dialogue rather than prompted extraction. It also requires the cognitive security to sit with an AI's counterargument without dismissing it - to let it slow you down when slowing down is warranted. Greg practices this deliberately, and considers it one of the highest-leverage applications available to any founder or executive.
Pivot vs. Take the Loss: What Experienced Founders Get Wrong
Greg's most counterintuitive observation has nothing to do with AI: experienced founders, he argues, are often more failure-averse than first-time founders - and that manifests as an over-willingness to pivot. When a company is not working, the instinct of a seasoned operator is to restructure, reposition, find an adjacent angle. Anything but accept the loss.
Sometimes the right move is to take the loss, rest, and start clean. Pivoting into a marginally better version of a broken thesis is not resilience - it is avoidance. The scar tissue of seven exits gives Greg the credibility to say this without it sounding like defeatism. It is one of the more honest things any experienced founder has said on this show.
Events as an Underutilized Enterprise GTM Channel
Section generated 100,000 virtual event attendees last year. Their approach: do not pitch. Deliver genuine, actionable value at every touchpoint. In an era where enterprise buyers are increasingly skeptical of webinars-as-sales-calls, Section's events work because the content earns attention rather than demanding it.
For founders building in the enterprise space, the lesson is not to copy Section's scale - it is to copy their orientation. Every event, every touchpoint, should make the attendee's job easier before asking for anything in return. The pipeline follows. Greg's newsletter at 130K subscribers operates on the same principle.
Cognitive Dependency and the AI Sobriety Question
Greg is unusually candid about the seductive quality of AI tools - and the risk that fluency shades into dependency. His frame: subscription is addiction. Not in a sensationalist way, but in the behavioral-economics sense. AI lowers the friction of thinking to such a degree that not using it starts to feel cognitively uncomfortable.
His recommendation: build AI-free time into your routine. Not as a rejection of the technology, but as a practice of maintaining the cognitive muscle that makes you valuable with or without it. The middle path is not abstinence - it is conscious use, which requires periodic contrast to calibrate.
- AI Drivers vs. AI Passengers - Every team member is either an AI Driver (owns the task, the decision, the conviction) or an AI Passenger (defers to AI output without taking responsibility for it). Drivers interrogate AI; passengers present its output as their own thinking. See Frameworks.
- The 10% Adoption Ceiling - Self-motivated employees will independently reach ~10% AI adoption in any org. Getting to 30–50% requires deliberate change management: leadership modeling, structured learning, and cultural expectation-setting. See Frameworks.
- Sometimes Just Take the Loss - Experienced founders often pivot when they should accept the loss, rest, and start clean. Pivoting into a marginally better version of a broken thesis is avoidance, not resilience. See Frameworks.
- Section - Enterprise AI adoption platform (software + services) that helps organizations move from 10% to 30–50% AI adoption through structured change management, upskilling programs, events, and a 130K-subscriber newsletter. Greg Shove's company.
- AI Driver - A team member who uses AI as a tool while retaining full ownership of the task, decision, and conviction behind the output. Interrogates AI rather than accepting its output uncritically. See Glossary.
- AI Passenger - A team member who defers to AI output and presents it as their own thinking without exercising independent judgment. A liability in knowledge work environments. See Glossary.
- 10% Adoption Ceiling - The natural plateau of AI adoption (~10% of staff) reachable through self-motivated employees alone. Breaking through to 30–50% requires deliberate change management. See Glossary.
- Capability Boundary - The outer limit of what a person or team can produce. AI extends this boundary, enabling smaller teams to do the work previously requiring larger ones. See Glossary.
- Cognitive Dependency - The behavioral state in which reliance on AI tools becomes habitual to the point that working without them creates discomfort. See Glossary.
- AI Thought Partner - Using AI not for output generation but for decision confirmation and blind-spot surfacing - a dialogue-based relationship where AI is asked to challenge, not just assist. See Glossary.
- Software + Services Model - A business model combining software platforms with human-led services delivery. Ensures adoption actually happens rather than just licensing. See Glossary.
What is the difference between an AI Driver and an AI Passenger?
A Driver owns the task, the decision, and the conviction. When they use AI they interrogate it, push back against it, and sign their name to the result. A Passenger lets AI do the thinking and presents the output as their own conclusion - 'this is what AI thinks' rather than 'this is what I think, informed by AI.' One is an amplifier. The other is a liability. Greg's policy is one strike.
Why is 10% the natural ceiling for AI adoption, and how do you get past it?
Every organization can reach about 10% adoption through self-motivated employees who would have figured out the technology regardless. These people do not need permission - they seek it. But 10% is also the ceiling without active intervention. Getting to 30–50% requires leadership modeling the behavior, structured learning programs, accountability, and a cultural frame where AI fluency is an expectation rather than a differentiator. Section's entire business model is built around closing that gap.
What does it mean to punch above your weight class with AI?
It means AI extends your capability boundary - the outer limit of what you and your team can produce. Section's 32 people do the work that previously required 45. That is not a headcount story. It is a quality-of-output story. The same people are producing better strategy, better content, better customer interactions - not because they are working harder, but because AI has extended what each person can reach. For founders, this reframes the build vs. hire decision: before adding a function, ask what the AI-extended capability of your existing team actually is.
What is the AI as Thought Partner use case, and why is it undervalued?
Using AI as a thought partner means using it not to generate output but to confirm decisions and surface blind spots before you make them. It is the AI that asks you the uncomfortable question before you commit to the call you are already leaning toward. This requires genuine dialogue rather than prompted extraction, and the cognitive security to sit with a counterargument without dismissing it. Greg considers it one of the highest-leverage applications available to any founder or executive - and almost no one is doing it deliberately.
Why do experienced founders pivot when they should take the loss?
Experienced founders are often more failure-averse than first-time founders, and that manifests as an over-willingness to pivot. When a company is not working, the instinct of a seasoned operator is to restructure, reposition, find an adjacent angle - anything but accept the loss. But sometimes the right move is to take the loss, rest, and start clean. Pivoting into a marginally better version of a broken thesis is not resilience. It is avoidance. The scar tissue of seven exits is what gives Greg the standing to say this.
How does Section use events as a GTM channel?
Section generated 100,000 virtual event attendees last year. The approach is simple: do not pitch, deliver genuine value at every touchpoint. In an era where enterprise buyers are skeptical of webinars-as-sales-calls, Section's events work because the content earns attention rather than demanding it. The pipeline follows naturally. Greg's newsletter at 130K subscribers operates on the same principle - make the reader's job easier before asking for anything. The lesson for enterprise founders is not to copy the scale; it is to copy the orientation.
What is cognitive dependency, and how do you avoid it?
Cognitive dependency is the behavioral state where reliance on AI becomes habitual to the point that working without it creates discomfort. Greg's frame: subscription is addiction - not in a sensationalist way, but in the behavioral-economics sense. AI lowers the friction of thinking so dramatically that not using it starts to feel cognitively wrong. His recommendation: build AI-free time into your routine. Not as rejection of the technology, but as a practice of maintaining the cognitive muscle that makes you valuable with or without it. The middle path is not abstinence - it is conscious use.