Winning with AI by Becoming Data Inspired | DisrupTV Ep 443
Winning with AI by Becoming Data Inspired | DisrupTV Ep 443
Winning with AI isn't a technology problem. It's a leadership, culture, and strategy problem rooted in how organizations actually use data to drive change.
Key Takeaways
- AI is a leadership test, not a tech rollout. Organizations that hand AI to IT and ask them to “find something to do with it” almost always stall at pilots. AI should serve business strategy, not sit beside it.
- Kill your pilots. Endless departmental pilots create the illusion of progress. Winners pick a small number of high-value, cross-functional initiatives and treat them as enterprise bets.
- Speed is the new moat. Everyone has access to similar cloud infrastructure and foundation models. The real differentiator is how fast your people, processes, and operating model can adapt.
- AI is value reclamation, not just cost cutting. The better question isn’t “how many people can I cut” — it’s “what new capabilities and services become possible” when you 10x your team’s capacity.
- AI fluency has four building blocks. Mindset, skill set, tool set, and decision set. Most organizations provide tools and skip the time and psychological safety needed for real fluency to take hold.
- When everyone owns AI, no one does. Organizations need a clear AI value owner ensuring a coherent roadmap and helping every employee answer where the company is headed and what their role is in getting there.
- The purpose of data is change — not dashboards. If data isn’t changing decisions, behaviors, or how the business runs, the expensive tech stack behind it isn’t earning its keep.
- Three myths undermine data-driven decisions. Data is not objective, it does not speak for itself, and it rarely provides clean, definitive answers. Pretending otherwise sets organizations up for disappointment.
- Culture determines whether data can change minds. Psychological safety, rewarding constructive dissent, and visible leadership humility matter more than any dashboard.
Part 1: AI Is Not a Tech Program. It's a Strategy and Leadership Test.
Charlene Li opens with a blunt assessment: most organizations underperform with AI because they treat it as a technology rollout, not a strategic, leadership-driven transformation.
Many organizations toss AI over to IT and ask them to go find something to do with it, instead of leading with business strategy and rethinking how decisions and workflows will actually change. Her core argument is that AI is a leadership and strategy challenge, not a tooling challenge. Leaders who abdicate AI to IT without owning the vision and value agenda almost always stall at pilots — because the real work is organizational adaptation, not just tech adoption.
One of her favorite examples of a sound approach is a strategy document that listed six traditional strategic objectives, then added a seventh: we will use AI to achieve the first six. That framing is the shift — AI should serve business strategy, not be a separate AI strategy off to the side.
Kill Your Pilots: Why “Pilot Purgatory” Destroys Momentum
Li doesn’t mince words on the way most enterprises do AI: endless pilots across departments, no real scaling, no enterprise-level value, and leaders declaring progress simply because they’re running lots of pilots.
“Kill all your pilots.”
Her prescription is to stop boasting about pilot volume — that’s just messing around — and instead pick a small number of high-value, cross-functional initiatives tied directly to strategic objectives. Treat these as enterprise bets, not departmental experiments.
She names a widespread AI hesitancy gap: on one side, excitement, FOMO, and pressure from boards, customers, and employees; on the other, fear of risk, speed of change, and uncertainty — what she jokingly labels FOGY, fear of getting in. Leaders caught in this gap become spectators rather than participants. The organizations that win are those whose leaders step into the mess, champion AI personally, and move decisively from pilots to production value.
Speed Is the New Moat: Why First Movers Win (Again)
Citing investor and AI thinker Vikram Mahidhar, Li argues that speed is the new moat in AI. Everyone can access similar cloud infrastructure and foundation models, and many enterprises are already rich in internal data. The differentiator is how fast you can adapt the organization — people, processes, governance, and operating model.
Li connects this to her long-standing work on disruption: disruptive growth is never incremental. Any outsized growth path will be messy, non-linear, and uncomfortable. AI amplifies this dynamic — it forces relational transformation, redefining roles, workflows, KPIs, and skills all at once.
Vala Afshar underscores this from his own vantage point inside a large AI-driven company, noting it is going through the messiest period in its 27-year history — not just because of the technology itself, but because of the sheer scale of reskilling, process redesign, and organizational change required.
In this environment, traditional distinctions between strategy and operations collapse. As R "Ray" Wang puts it, they are now effectively the same thing because the feedback loops are so fast.
Beyond Cost Cutting: AI as Value Reclamation and Capability Expansion
Much of the classic AI narrative focuses on efficiency and headcount reduction: do more with less, expand margins, cut jobs. Both Afshar and Li push back hard on that framing.
Li describes how many leaders initially think: if I can 10x my people, I can fire the other nine. She argues this is entirely the wrong plane of thinking. The better question is: if we can 10x ten people, how do we get 100x value? What new capabilities and services become possible?
She highlights Connect Up, a call-center company, as a case study. Instead of using AI to cut agents, they reduced error rates by 85 to 90 percent with co-pilots, shortened time-to-proficiency for new hires by 30 to 40 percent, freed agents from routine tasks to focus on higher-value engagement, and built entirely new services for clients — like analyzing non-performing loans for law firms in minutes instead of months. Importantly, they committed to increasing headcount by 5% over two years, not shrinking it.
Afshar reinforces this with his own language of value reclamation: AI can reclaim value that used to be ignored for economic or capacity reasons. By offloading low-impact, deterministic work to machines, leaders can finally invest human talent in activities they previously had no bandwidth to pursue. The mindset shift is from subtraction — efficiency only — to addition: new revenue, new experiences, new capabilities.
Building an AI-Fluent Workforce
Li outlines four building blocks for AI fluency. Mindset — curiosity, openness to change, willingness to experiment, and comfort with never being on top of everything as the space evolves. Skill set — knowing what AI can and cannot do, and the ability to frame problems in ways AI can help solve. Tool set — access to AI capabilities embedded in day-to-day workflows, with practical training on the tools employees actually use. And decision set, or governance — clear rules and processes for responsible, ethical, and sustainable usage, with guardrails that are understood and actionable, not just policy documents.
She compares AI fluency to learning chopsticks: at first it feels awkward and unnatural, but over time you stop noticing the tool and simply flow. The inflection point arrives when you start using AI to ask how you should use AI to help you do this task better.
Most organizations, she notes, provide some tools, minimal training, and almost no time to practice. Without deliberate time and psychological safety to experiment, true fluency never emerges.
The Case for an AI Value Owner
Afshar raises a structural question: should companies appoint a Chief AI Officer, or embed AI value owners within each business line?
Li’s concept of an AI value owner is less about title and more about function. Everyone in the organization owns a piece of the AI puzzle — HR owns the future workforce, GTM owns customer experience, Finance owns token and compute economics. But when everyone owns AI, no one truly does.
The AI value owner wakes up every day ensuring there is a coherent, enterprise-level AI roadmap, that functions are aligned and moving at a similar tempo — no jackrabbits sprinting ahead and no turtles dragging behind — and that the roadmap is updated frequently, at least quarterly and often continuously.
This person also helps every employee answer three questions: what future are we heading toward, what is our strategy to get from today to that future, and what is my role in making that strategy a success? In Li’s view, until leaders across the organization can clearly articulate these answers, the AI value owner has more work to do.
Part 2: From Data Frustration to Data Inspiration
In the second half of the episode, Sebastian Wernicke shifts the lens from AI back to its foundation: data and culture.
For years, leaders have been told that data is the new oil, and that big data, analytics, machine learning, and digitalization will transform everything. Yet survey results remain stubbornly consistent: nearly all leaders say data and AI are a top priority, and about two-thirds admit they are not satisfied with the value they’re actually getting.
This persistent gap breeds data frustration — we’re spending a lot but not seeing transformative impact — and data cynicism — we tried that, it never really worked for us. Wernicke argues this is not primarily a tooling issue. It is a cultural and decision-making issue.
Data's True Purpose: Change, Not Dashboards
Wernicke urges leaders to ask a simple but profound question: what is the purpose of data? The conventional answer is to generate insights we can use. His answer is more direct: the purpose of data is change — change in decisions, change in behaviors, change in how the business is run.
“If data isn't changing how you operate, you don't need the expensive tech stack, data cleaning, or analytics infrastructure you've built.”
He highlights a core misconception he calls the data deficit theory: the belief that if we just bring the right data to the right people at the right time, good decisions will automatically follow. Decades of psychology research say otherwise. When we’re deeply invested in a belief or direction, we do not naturally turn to data to disconfirm it — instead, we question the data’s quality, criticize the analyst, or rationalize the result. Without a culture of inquiry and safety, data rarely overturns strongly held assumptions.
Three Myths of Data That Undermine Decision-Making
Wernicke identifies three common myths that sit behind the dream of being data driven.
Myth one: data is objective. In reality, data is a compression of reality. Every metric — even something as simple as active customers — reflects choices about who counts as a customer, what active means, and over what time window. Those design decisions matter enormously when interpreting and acting on data.
Myth two: data speaks for itself. Even a simple question like which show is best, based on user ratings, has multiple defensible answers — the highest average rating, the most ratings overall for reliability, or the fewest worst ratings to minimize bad experiences. Each definition yields a different winner, all justified logically. Without clear definitions upfront, data-driven decisions can be steered toward any preferred outcome.
Myth three: data provides clear, definitive answers. In real analytics, you quickly leave the world of simple counts and enter statistics — probabilities, confidence intervals, likelihoods. Executives often want a binary answer, left or right, but data typically says something closer to option B is 70% likely to be better. If you’re not prepared to operate in shades of probability, data-informed decision-making becomes frustrating and underused.
These myths set organizations up for disappointment. The antidote is embracing nuance, defining terms upfront, and using data to support structured inquiry — not to rubber-stamp preconceived answers.
Creating a Culture Where Data Can Actually Change Minds
Afshar and Wernicke dive into the psychological safety and leadership behaviors needed to make data truly impactful.
Wernicke shares a simple three-question survey he gives audiences. Is your organization data driven? Most hands go up. In the last month, have you changed your mind based on data? Fewer hands. In the last month, has your manager publicly changed their stance because of new data? Almost no hands.
“People watch what leaders actually do, not what posters say.”
Several practical leadership moves emerged. First, ask junior people to speak first — Afshar cites Jeff Bezos’s practice of inviting the most junior person in the room to share their view first, reducing anchoring on the most senior voice and signaling that good ideas and strong data matter more than title.
Second, publicly model being corrected by data. Wernicke suggests leaders should actively seek out examples where data contradicts their stance, change their minds openly and visibly, and show that being wrong and adjusting is not punished, but respected.
Third, reward constructive dissent. Promotions, bonuses, and recognition should flow to people who use data to challenge assumptions productively, raise uncomfortable but important truths, and help the organization learn, not just comply. Without this, junior analysts quickly learn which kinds of numbers are career-enhancing and which are career-limiting — and they adjust accordingly.
Radical Data Integrity: Trustworthy, Not “Perfect”
On data quality, Wernicke introduces the idea of radical data integrity. Most organizations think good data means clean data — few missing values, numbers that tie neatly to the balance sheet. But for continuous, high-stakes decision-making, what matters more is trustworthiness.
Trustworthy data means you understand how it was generated, you know what it can and cannot be used for, you know who is accountable for it, and you can ask questions and get credible answers when something looks off. Cleanliness is useful, but integrity and transparency are critical if you want leaders to rely on data even when it challenges their instinct.
Most organizations struggle because they treat data as an annual spring-cleaning project: fund a one-time clean-up, declare victory, and watch entropy return over the next year. Instead, Wernicke calls for systemic and ongoing stewardship, where integrity is designed into processes and roles, not treated as a periodic initiative.
Bringing It Together: Winning with AI Because You're Data Inspired
By the end of the episode, the synergy between the two books is obvious. You can’t win with AI if you’re still frustrated with data. You can’t become data inspired without a culture of inquiry and psychological safety. And you can’t scale either without leaders owning AI as a strategic lever, not a technology project.
Charlene Li’s message: use AI to serve and accelerate your business strategy. Move beyond pilots, embrace speed, and invest in organizational readiness across mindset, skill set, tool set, and decision set.
Sebastian Wernicke’s message: make data’s purpose explicit — to drive change. Debunk the myths of objectivity and easy answers, build radical data integrity, and foster a culture where it’s safe, and expected, to let data challenge your beliefs.
Final Thoughts
DisrupTV Episode 443 makes an argument that cuts against a lot of conventional AI messaging: the hardest part of winning with AI was never the technology. It’s whether leaders are willing to do the harder, messier work of organizational change — owning strategy personally, killing the pilots that feel safe but go nowhere, and building a culture where data is trusted enough to actually change minds.
Charlene Li and Sebastian Wernicke approach the same problem from opposite ends and arrive in the same place. Li starts with AI and works backward to the organizational readiness required to use it well. Wernicke starts with data and works forward to the cultural conditions required to let it matter. Both conclude that the technology was never really the constraint — leadership, trust, and culture are.
For organizations willing to do this hard, human work — clarifying strategy, appointing real ownership, investing in fluency, and building a culture where being wrong in public is safe — AI stops being another wave of tech hype and becomes a genuine engine for transformation.
“Winning with AI starts with becoming data inspired — and that starts with leadership, not technology.”
Related Episodes
If you found Episode 443 valuable, here are a few others that align in theme or extend similar conversations:
Your Hosts
R “Ray” Wang is the CEO of Silicon Valley-based Constellation Research Inc. He co-hosts DisrupTV, a weekly enterprise tech and leadership webcast that averages 50,000 views per episode and blogs at www.raywang.org. His ground-breaking best-selling book on digital transformation, Disrupting Digital Business, was published by Harvard Business Review Press in 2015. Ray's new book about Digital Giants and the future of business, titled, Everybody Wants to Rule The World was released in July 2021. Wang is well-quoted and frequently interviewed by media outlets such as the Wall Street Journal, Fox Business, CNBC, Yahoo Finance, Cheddar, and Bloomberg. Short Bio R “Ray” Wang (pronounced WAHNG) is the Founder, Chairman, and Principal Analyst of Silicon Valley-based Constellation Research Inc. He…...
Read moreElle is the Director of Marketing at Constellation Research and producer of the weekly enterprise tech show, DisrupTV. She leads marketing and communications efforts across all avenues while also managing an array of programs including the SuperNova Awards, Business Transformation 150, AX100, AI150, S50 and ShortLists. She has previous experience in public relations and has a strong passion for the marketing world. Contact Elle at [email protected] ...
Read morePublished
Authors
