the AI Boom Is Just the Latest of Six 90-Year Hype Cycles

There is an infographic that has been widely shared in enterprise leadership circles that maps 90 years of AI history — from Turing’s foundational work in the 1930s and 1940s, through the six documented hype-and-winter cycles that followed, culminating in the current AI moment as the seventh cycle.

The infographic is factually accurate. Every one of the previous six cycles happened. Each was preceded by dramatic hype about AI’s imminent capabilities. Each collapsed into a “winter” of reduced funding, discredited claims, and disillusioned researchers.

The implicit conclusion the infographic invites is that the current moment is cycle seven. That the boom is the setup for the collapse. That responsible executives should treat AI hype in 2026 with the same skepticism that would have been appropriate during the previous six cycles.

This conclusion, in my direct assessment, is structurally wrong.

The current moment is not a cycle. It is a floor.

Cycle versus floor

A cycle is a pattern of hype and disillusionment that returns to approximately the same baseline. Cycles have amplitude but no trend. The 1970s AI winter left researchers at approximately the same capability level as 1960s AI researchers. The 1980s expert-systems winter left researchers at approximately the same capability level as pre-boom.

A floor is different. A floor is a permanent step-change in capability. After the floor is established, subsequent cycles happen around a new baseline, not the old one. The transistor established a floor. The internet established a floor. The smartphone established a floor.

Cycles are common. Floors are rare. Most technology news is cycle news, which is why the infographic’s framing feels intuitive.

The specific question, for every executive right now, is whether the current AI moment is producing a floor or a cycle. If it is producing a cycle, wait it out. If it is producing a floor, position for the new baseline before it is fully established.

Three diagnostic tests

Three specific tests distinguish floors from cycles. All three are, in the current AI moment, pointing away from cycle and toward floor.

Test one: the cost curve.

Cycles have flat or declining cost efficiency. The capability produced by 1970s AI at a given cost was approximately the same as the capability produced by 1960s AI at the same cost — the winter came because the promised capability improvements didn’t arrive.

Floors have exponentially improving cost efficiency. Transistor computing got exponentially cheaper per unit of capability every 18-24 months for decades. Internet bandwidth got exponentially cheaper per unit of data transferred. Smartphone compute got exponentially cheaper per unit of processing.

AI compute is currently getting exponentially cheaper per unit of capability. The DeepSeek $5.6M training run in January 2026 is a specific data point in this curve. The curve is not flattening. It is accelerating.

Test two: the deployment pattern.

Cycles have narrow, elite deployment. 1980s expert systems were deployed in a small number of specialized enterprise applications. When those applications underperformed, deployment collapsed.

Floors have broad, distributed deployment. The transistor was deployed in every electronic device. The internet was deployed in every home and business. The smartphone was deployed in every pocket.

Current AI deployment is broad and distributed. Every knowledge worker with an internet connection has access. Every enterprise is running pilots. The deployment pattern is a floor pattern, not a cycle pattern.

Test three: the research trajectory.

Cycles have plateauing research output. In previous AI winters, the underlying research stopped producing capability improvements before the funding cuts arrived. The winter was a lagged response to a research plateau.

Floors have accelerating research output. Transistor research continued producing capability improvements for decades after the initial breakthrough. Internet research produced continuous protocol and infrastructure improvements. Smartphone research produced continuous chip and interface improvements.

Current AI research is accelerating, not plateauing. New capabilities, new architectures, and new deployment patterns are emerging on a monthly basis. There is no plateau in the underlying research trajectory.

What this means for executives

All three tests point away from cycle and toward floor. Which means the current AI moment is not the seventh iteration of a historical pattern. It is the establishment of a new baseline.

The executive posture appropriate for a cycle — wait it out, protect existing capital investments, avoid overcommitting — is exactly wrong for a floor.

The executive posture appropriate for a floor — position early, absorb the new capability into core operations, plan for a fundamentally different competitive baseline — is what a small number of executives are actually doing right now. Most are not.

Three practical questions

One: which of the three tests would you argue against? If you believe the current AI moment is a cycle, name specifically which of the three — cost curve, deployment pattern, research trajectory — you think is a cycle pattern rather than a floor pattern. The specificity forces the argument.

Two: what would your investment thesis look like if you accepted the floor framing? Higher, faster, more integrated. Most executive AI investment theses are calibrated to cycle assumptions. Recalibrate to floor assumptions and observe the change.

Three: who in your organization is arguing for the floor framing? They are, structurally, correct. Are they being heard?

The closing thought

The infographic is a compelling piece of pattern-matching. Six previous cycles. Each preceded by hype. Each followed by disillusionment. It is emotionally satisfying to conclude that the seventh cycle is underway.

Emotional satisfaction is not evidence. The three diagnostic tests are evidence. All three point at a floor.

The executives who understand this in 2026 will position early. The executives who rely on the infographic will position late. The gap between the two groups will be visible by 2029.

The world has changed. The leaders who notice will be the ones the next decade is built around.

The post the AI Boom Is Just the Latest of Six 90-Year Hype Cycles first appeared on Agile Agilist | SAFe® Gold Partner for Agile Transformation, Innovation & Leadership Training.

Best AI Models 2026: Ratings by Task & Strength

There Is No Best AI Model—Only the Best Model for the Right Job

Every few months, a new benchmark appears claiming that one model has become the world’s “best.”

Today it’s GPT.

Tomorrow it’s Claude.

Next month it may be Gemini, DeepSeek, or Qwen.

The reality is much more interesting.

There is no universal winner.

The latest independent benchmark reports consistently show that each frontier model dominates different dimensions of intelligence, cost, reasoning, coding, multimodal understanding, or context handling.


AI Is Becoming Specialized

Think about elite athletes.

You wouldn’t ask:

“Who’s the best athlete?”

Instead you ask:

• Who’s the fastest sprinter?
• Who’s the strongest weightlifter?
• Who’s the best marathon runner?

Large Language Models have entered the same stage of evolution.

Each excels at different work.


The Emerging Strengths

Claude

Best for:

• agentic software engineering
• autonomous coding
• complex development workflows


GPT-5.6

Best for:

• enterprise reasoning
• mathematics
• business analysis
• balanced performance across many tasks


Gemini

Best for:

• multimodal reasoning
• long-context analysis
• video, images, documents
• enterprise knowledge retrieval

Its extremely large context window allows it to process information that would overwhelm many other models.


DeepSeek

Best for:

• high reasoning performance
• cost efficiency
• open deployment strategies

Organizations can achieve near-frontier performance while dramatically reducing inference costs.


Qwen

Best for:

• multilingual applications
• open-weight deployment
• enterprise customization
• strong reasoning with infrastructure flexibility


This Isn’t New

Every major technology revolution followed exactly the same pattern.

Cloud didn’t replace everything.

It created:

• Infrastructure as a Service
• Platform as a Service
• Software as a Service

Each solved different problems.

AI is following the same trajectory.

Instead of one model replacing every other model…

We’re moving toward AI ecosystems.


The Leadership Shift

The executive question is no longer:

Which AI should we buy?

The better questions are:

  • Which workflow requires deep reasoning?
  • Which process needs multimodal understanding?
  • Which tasks require the lowest possible cost?
  • Which activities should remain human-led?

The competitive advantage is shifting from model selection to AI orchestration.


Executive Coaching in the AI Era

Leadership is becoming less about knowing the answers.

It’s becoming about designing intelligent systems.

Future executives must learn to:

  • orchestrate multiple AI models
  • redesign workflows
  • Govern AI responsibly
  • Combine human judgment with machine intelligence

This is remarkably similar to earlier transformation waves.

The organizations that succeeded in cloud, Agile, and digital transformation weren’t those that adopted technology first.

They were the ones who redesigned how work flowed through the organization.

AI is no different.


The future won’t belong to the organization using one “perfect” AI.

It will belong to the organization that knows which intelligence to apply, at the right moment, for the right problem.

Because in the age of AI, the competitive advantage is no longer the model.

It’s the leader who knows how to orchestrate them.

The post Best AI Models 2026: Ratings by Task & Strength first appeared on Agile Agilist | SAFe® Gold Partner for Agile Transformation, Innovation & Leadership Training.

Framework for AI Transformation in 2026.

In 1996, Steve Jobs sat down for a PBS documentary called Triumph of the Nerds. Robert X. Cringely was interviewing him. The conversation turned to Apple’s 1979 visit to Xerox PARC, the research lab where the graphical user interface, the mouse, and a long list of other foundational computing concepts had been invented.

Jobs said, in essence: Picasso had a saying — good artists copy, great artists steal — and we have always been shameless about stealing great ideas.

The line has aged better than almost any sentence in the history of business commentary. It has become the unofficial motto of Silicon Valley. It hangs on the walls of startups. It gets quoted in MBA classrooms. It is invoked, usually badly, as justification for almost any unattributed borrowing of someone else’s work.

The line is also, almost everyone who quotes it misses, a deeply incomplete version of a much more interesting framework.

This piece is about what the famous version gets wrong, what the better version actually says, and why the better version is the single most useful framework I can offer for any senior leader trying to navigate AI transformation in 2026.

The Picasso problem

Let’s start with the easy correction.

The attribution to Pablo Picasso is almost certainly apocryphal.

There is no documented Picasso source for “good artists copy, great artists steal.” Multiple researchers have looked. The line does not appear in any published Picasso writing, interview, or documented conversation. It has been attributed to him posthumously, and the attribution has stuck because the line sounds like something a defiantly transgressive twentieth-century master might have said.

The same line has been attributed, at various times, to Igor Stravinsky, William Faulkner, and T.S. Eliot. The genealogy is murky. Different researchers have credited different originators.

The most likely actual source — the earliest documented version of the underlying idea, in publication that can be verified — is T.S. Eliot.

In 1920, Eliot published an essay collection called The Sacred Wood. One of the essays was about the Elizabethan playwright Philip Massinger. In that essay, Eliot wrote something that is recognizably the source of everything that came after:

“Immature poets imitate; mature poets steal; bad poets deface what they take, and good poets make it into something better, or at least something different. The good poet welds his theft into a whole of feeling which is unique, utterly different from that from which it was torn.”

Read it slowly. There is more in those two sentences than in the entire Picasso-attributed simplification.

Eliot is not saying “great artists steal.” Eliot is saying something far more specific. He is saying that the stealing is not the achievement. The achievement is what you do with what you stole.

Bad poets deface what they take.

Good poets make it into something better, or at least something different.

That second clause is the entire heart of the framework. It is the part that the Jobs/Picasso simplification quietly drops. And it is the part that turns the line from a permission slip into a discipline.

The Xerox PARC story, told accurately

To understand what Jobs actually meant when he invoked the (probably misattributed) Picasso quote in 1996, you have to understand the specific story he was telling.

In December 1979, a small Apple team — Jobs and several engineers — were invited to visit Xerox’s Palo Alto Research Center, known as PARC. The visit was part of a deal: Xerox had taken a pre-IPO stake in Apple, and in exchange, Apple got two days of access to the PARC research.

What Apple saw at PARC was, in 1979, almost incomprehensibly advanced.

The Alto workstation. A graphical user interface with overlapping windows. A three-button mouse for pointing and clicking. Bitmap displays where pixels could be turned on and off individually. Networked computing. Object-oriented programming in Smalltalk. Laser printing. Ethernet. WYSIWYG document editing.

Almost every concept that would define personal computing for the next forty years was sitting in that one lab, in working prototype form, with researchers who had no real idea how to turn any of it into a product anyone could buy.

Jobs, by every account of the visit, immediately understood two things.

First, that what he was looking at was the future of computing.

Second, that Xerox did not understand what they had built. The research culture at PARC had produced extraordinary technology and almost zero commercial intuition about what to do with it.

Apple took the concepts.

This is the part of the story everyone tells.

What gets told less often is what Apple did with the concepts they took.

The Xerox mouse had three buttons and cost roughly $300 to manufacture. The hardware was research-grade. It was unsuitable for consumer use at consumer price points. Apple’s industrial design team, led by Jerry Manock and later by Hartmut Esslinger’s Frogdesign, redesigned the mouse from scratch. One button. Simple enough that no user would need a manual. Apple worked with manufacturers to bring the unit cost down to roughly $15. By every meaningful measure — usability, manufacturability, price, durability — Apple’s mouse was a fundamentally different product than what they had seen at PARC.

The Xerox interface concepts were research-quality demonstrations on a $30,000+ Alto workstation. Apple put the same concepts on a Macintosh that retailed for $2,495 in 1984 — making the technology accessible to a market that had been completely unreachable to Xerox.

The Xerox GUI used a complicated, technically elegant system that required Smalltalk and was deeply tied to research infrastructure. Apple stripped it down, simplified it, sacrificed some flexibility for usability, and built QuickDraw — Bill Atkinson’s graphics library. QuickDraw made the Mac UI possible at consumer price points and consumer speeds.

Apple stole the idea.

Then they made it into something better.

That’s the framework Eliot was pointing at in 1920. That’s what Jobs was actually doing when he invoked the Picasso line in 1996. The stealing was the entry condition. The transformation was the achievement.

The Microsoft layer

It is worth telling one more layer of the story, because it is the part that completes the lesson.

In 1985, Microsoft released Windows 1.0. The product was, by almost every honest account, derived directly from what Microsoft had seen of the Macintosh during their collaboration with Apple on Office software. The early versions of Windows looked enough like the Mac interface that Apple sued Microsoft in 1988 for copyright infringement.

The lawsuit dragged on for years. Microsoft ultimately prevailed, partly on the argument that Apple itself had derived the interface from Xerox PARC, and partly on a separate licensing agreement Apple had signed with Microsoft.

The lesson Jobs took from this — and the lesson he was implicitly referencing in the 1996 Triumph of the Nerds interview — was that the entire personal computing interface had been a chain of openly acknowledged thefts. Xerox researchers had taken concepts from earlier human-computer interaction research at SRI and at Doug Engelbart’s lab. Apple had taken from Xerox. Microsoft had taken from Apple. Everyone, at every step, had been “stealing.”

What separated the participants was not whether they stole. Everyone stole. What separated them was what they did with what they stole.

Xerox stole brilliantly from earlier research and produced research-quality demos that never became shippable products. Apple stole from Xerox and produced products that reshaped consumer computing. Microsoft stole from Apple and built the dominant operating system of the personal computing era.

In every case, the stealing was the easy part. The transformation was where the value was created.

Why the simplified version is dangerous

The reason I am writing about this in 2026 is that the simplified Jobs/Picasso version of the framework has, over the past three decades, produced a particular kind of executive behavior that is now actively damaging AI transformations.

The simplified version says: great artists steal.

In the hands of an executive who has not read the Eliot original, this becomes: we can copy what other companies are doing without doing the harder work of transforming it for our context.

I have seen this play out, repeatedly, in AI transformations across multiple industries in 2025 and 2026.

A retail company sees that another retail company has deployed AI-powered customer service successfully. They decide to “steal” the same approach. They copy the architecture, the vendor selection, the use case definition, almost line for line. They deploy it. It fails — because their customer base is structurally different, their existing service infrastructure is structurally different, and the parts of the implementation that they needed to transform for their context were the parts they didn’t transform.

A financial services firm sees that another bank has used AI for credit risk modeling with strong results. They copy the model architecture. They deploy it on their portfolio. It fails — because the underlying data, the regulatory context, and the institutional risk culture are all different, and they didn’t do the transformation work.

A consulting firm sees that another consulting firm has been generating polished AI-augmented deliverables in half the time. They copy the workflow. Their consultants try to use the new approach. The work suffers — because what the leading firm had spent eighteen months building was not the workflow but the judgment about when to use the AI and when not to, and the judgment was the thing that took eighteen months and didn’t transfer in a memo.

Every one of these failures has the same root cause. The executives involved correctly identified what to steal. They incorrectly assumed that the stealing was the work.

Eliot, in 1920, had named the failure mode in advance.

Bad poets deface what they take.

What he meant: bad poets steal effectively but fail to transform. The result is a damaged version of the original — recognizable, derivative, less valuable than the source. Defaced.

The contemporary executive version of this is the AI initiative that copies someone else’s successful pilot, deploys it without context-specific transformation, fails publicly, and produces internal cynicism about AI more broadly. The pilot is defaced. The company is worse off than if they had never attempted it.

The Eliot discipline, translated for 2026

If Eliot is right that the stealing is the entry condition and the transformation is the achievement, the question for senior leaders becomes: what does the transformation actually consist of?

I think it has four components.

One: context-specific adaptation.

What you steal has to be modified for the conditions in which it will operate. Apple modified the Xerox interface for consumer price points. Microsoft modified the Apple interface for IBM PC architecture. Your AI initiative has to modify the source approach for your industry, your data, your culture, your regulatory environment, your customer base, your team’s existing skill set.

The transformation is not optional. It is the entire work. Skipping it produces the defaced version Eliot warned about.

Two: integration into the existing system.

The stolen idea has to be welded into a coherent whole, in Eliot’s language. It cannot sit as a bolted-on appendage. Apple didn’t just add a mouse to the Apple II. They redesigned the entire computing experience around the mouse — the operating system, the application paradigm, the file system, the visual metaphors.

Your AI initiative cannot be a separate program running alongside your existing operations. It has to be welded into how decisions get made, how work flows, how value gets delivered. The integration is the transformation.

Three: aesthetic and human-experience resolution.

This is the part that almost no business framework captures and that Eliot’s poetic version surfaces. The Xerox mouse worked. The Apple mouse felt right. The Xerox GUI was technically sophisticated. The Apple GUI was aesthetically resolved in a way that produced an entirely different human experience.

For AI initiatives, this matters more than executives realize. The technical capability is necessary but not sufficient. The version of the AI tool that your people will actually use is the one where the human experience of using it has been resolved — the prompts are intuitive, the outputs land in the right format, the workflow doesn’t require constant context-switching. The aesthetic and experiential layer is most of the transformation.

Four: the willingness to make it into something different.

Sometimes the version that emerges from honest transformation is better than the source. Sometimes it’s just different — adapted to your context in ways the source never was. Eliot was explicit that either outcome counts. Better, or at least different.

The executive temptation is to claim the better. The actual achievement is often the different — the version that works in your context, even if it would not work as well in the source context. This is fine. This is, in fact, the whole point. Your version is supposed to be different. That’s what transformation is for.

Three practical questions

If you want to actually apply the Eliot framework to your AI transformation in 2026, three questions are worth sitting with.

One: what specifically are you stealing, and from where?

Most enterprises are stealing from everywhere all at once — vendor pitches, McKinsey reports, conference panels, industry case studies, peer benchmarks. The result is a patchwork without coherence. Eliot’s framework demands specificity. Pick the three sources you are stealing from most heavily. Name them. Articulate exactly what you are taking from each.

If you can’t do this, you are not stealing in Eliot’s sense. You are pattern-matching to fragmentary impressions, which produces the defaced version more often than the transformed one.

Two: what is the transformation work that has to happen to make this work for your context?

For each source, articulate the specific gap between the source context and your context. Different industry. Different data. Different regulatory environment. Different customer expectations. Different team capabilities. Different cost structure.

The transformation work is the work of bridging those specific gaps. If you cannot articulate the gaps, you have not yet understood what transformation is required. The pilot will fail in exactly the ways you have not yet examined.

Three: who in your organization is responsible for the transformation work?

This is the question most enterprises skip and the question that determines whether you produce the defaced version or the transformed version.

In most AI initiatives, the people responsible for the project are responsible for the stealing — the vendor selection, the architecture, the deployment plan. The transformation — the context-specific adaptation, the integration, the human-experience resolution, the deliberate differentiation from the source — is implicitly assumed to happen on its own.

It does not happen on its own.

The transformation work has to be staffed, resourced, and led, with the same explicit attention as the technical implementation. The leaders who do this produce successful AI initiatives. The leaders who skip it produce defaced versions.

The closing thought

There is a small irony embedded in this entire piece that I want to surface before I close.

The quote that has propagated through Silicon Valley for thirty years — “good artists copy, great artists steal” — is itself an example of the failure mode it tries to describe.

The quote was stolen from Eliot. It was simplified in the stealing. The most important clause — the one about good poets making it into something better, or at least something different — was dropped.

The resulting version is shorter, punchier, and more memorable. It is also less useful. It is a defaced version of a more powerful framework. It has been quoted, in defaced form, for a century.

That is exactly the failure mode the original framework was warning about.

In 2026, with AI transformation rewriting nearly every assumption about how senior leaders create value, the temptation will be to grab the punchy version of every framework, every approach, every case study, and move quickly. Steal. Deploy. Move on.

The Eliot discipline says this is the path to the defaced version.

The harder path is to do the transformation work. To take what you steal and make it into something better, or at least something different, for your specific context. To weld your theft into a whole that is uniquely yours.

This is slower. It requires more skill. It demands the kind of judgment that doesn’t transfer easily between contexts and that cannot be outsourced to a vendor or a consulting firm.

It is also the only path that produces the version of your AI transformation that actually works.

Steal openly. Steal from everywhere. Don’t pretend to originality you don’t have.

Then do the harder work of transformation.

Make it into something better. Or at least something different.

That is the framework Eliot named in 1920, that Jobs lived in the 1980s, and that almost every senior leader in 2026 needs to internalize before they spend the most important years of their career producing defaced versions of other people’s pilots.

The world has changed. The leaders who notice will be the ones the next decade is built around.

The post Framework for AI Transformation in 2026. first appeared on Agile Agilist | SAFe® Gold Partner for Agile Transformation, Innovation & Leadership Training.

Carry Into the Age of AI.

In January 2007, a 20-year-old named Stefani Germanotta was sitting in her parents’ Upper West Side apartment when she got news that should have ended her career.

Island Def Jam Records had just dropped her.

She had been signed three months earlier by Antonio “LA” Reid, then president of Def Jam, who reportedly signed her on the spot after hearing her sing from his office. The album was scheduled for release in May 2007. The label machinery had begun moving. And three months in, before anything happened, it was over.

The court documents in a later lawsuit between her producer and the label would describe the decision with one word: “inexplicable.”

She was devastated.

This is the story most people know, told the way most people tell it: a future global icon faces an early setback, perseveres, becomes Lady Gaga, proves them all wrong. End of motivational poster.

I want to write about the eighteen months between getting dropped and her first hit single. Because that’s the part of the story that matters, and almost no one tells it correctly.

It’s also the most useful eighteen months of any modern career to study if you are a senior leader trying to navigate AI transformation in 2026.

The wilderness period

When Def Jam dropped her in January 2007, Stefani Germanotta had almost nothing left of the conventional path she had been on.

She had dropped out of NYU’s Tisch School of the Arts two years earlier, at 19, against her father’s strong objections. She had been bullied at Convent of the Sacred Heart from age 11 — for her appearance, for her “eccentric habits,” for being, in her own description, “either too provocative or too eccentric” to fit in. Her NYU classmates had created a Facebook group titled “Stefani Germanotta, you will never be famous.”

And now the most powerful record label in the country had decided that her NYU classmates were right.

She was 20 years old. She had no contract. No band. No money. By every available external signal, no future.

What did she do?

She didn’t sit and grieve. She didn’t take a year off to figure herself out. She didn’t pivot to a sensible backup career. She didn’t go back to her parents’ apartment and try to repair her relationship with her father, who, according to multiple biographical accounts, had stopped speaking to her over her drug use and lifestyle choices.

She went back to performing in Lower East Side burlesque clubs.

She co-formed a revue called the Ultimate Pop Burlesque Rockshow with a performance artist named Lady Starlight.

She moved to a tiny apartment on Stanton Street in the Lower East Side.

She took freelance songwriting work, writing songs for other artists at Interscope Records — the same label network whose flagship had just dropped her.

She kept performing wherever she could get a paying gig. Open mics. Burlesque shows. Anywhere a piano and an audience would let her work.

She kept producing. Every single day.

Eight months into this period, in late 2007, the R&B artist Akon heard one of her newer tracks and insisted she sign with his label, KonLive Distribution. She signed with Vincent Herbert at Streamline Records, an Interscope imprint. She spent the rest of 2007 and the first half of 2008 recording the album that would become The Fame.

In August 2008, The Fame released. Within months, “Just Dance” became her first global hit. By 2009, “Poker Face” was the best-selling single in the world.

The eighteen months between the Def Jam drop and The Fame‘s release were not a pause. They were not a recovery. They were not the bad part of the story that happened before the good part started.

They were the story. They were what produced the artist the world would eventually call Lady Gaga.

This is the part almost every retelling gets wrong.

What the motivational version misses

The standard framing of this story is “believe in yourself when nobody else does.”

That framing is too thin. It collapses the actual mechanism into a feeling. It implies that what carried her through 2007 was an emotion — a kind of internal conviction that powered her past the rejection.

I think that reading is dramatically wrong, and the wrongness matters because it makes the lesson useless for everyone trying to apply it.

Belief, on its own, is useless.

There are millions of people who believe in themselves and never produce anything anyone notices. There are millions of people who lost faith in themselves at some point and produced enormous work anyway. The distribution of self-belief in the population does not map cleanly to the distribution of accomplishment. If it did, every motivated MBA graduate would be Lady Gaga.

What worked for her wasn’t belief.

It was discipline.

The discipline of continuing to produce, daily, through one of the lowest periods of her life.

Same piano practice. Same songwriting habit. Same showing-up-to-perform discipline. Executed every day for eighteen months while the world told her she was finished.

She did not, by every account, feel confident every day during this period. She was openly struggling with drug use. She was estranged from her father. Her finances were precarious. The Facebook group from NYU still existed. The Def Jam decision still stung.

What carried her wasn’t the absence of those feelings.

It was the discipline of producing in spite of them.

That distinction is the entire usable lesson.

The motivational version says: feel the conviction first, then the work will come.

The accurate version says: do the work first, daily, regardless of how you feel, and the conviction comes downstream — usually as a byproduct of the work itself.

This is, structurally, the same insight that runs through almost every research-based finding about high performance in any domain. Anders Ericsson’s deliberate practice research. James Clear’s habit-formation work. Daniel Kahneman’s writing on the relationship between behavior and identity. The behavior produces the identity. Not the other way around.

Lady Gaga’s eighteen months in the wilderness are one of the cleanest case studies of this principle in modern entertainment history.

She wasn’t an extraordinary person who happened to keep going.

She was an ordinary person who kept going, and the keeping-going produced the extraordinary person.

Why this matters in 2026

The reason I want to write about this story now, in 2026, is that I think it is the most useful single career narrative for any senior leader navigating AI transformation right now.

Every senior leader I work with is, in some specific way, being dropped.

The AI transformation is telling everyone that the systems that rewarded their old way of working are quietly retooling for something else. The skills that got them here aren’t necessarily the skills that get them there. The expertise that defined their career — the brand they spent twenty years building — is, in some specific way, becoming less central than it used to be.

That feeling — the dropped-by-Def-Jam feeling — is showing up in board rooms, executive committees, and 1:1s across every industry I work in.

It shows up as:

The CMO who notices that AI is now generating, in seconds, the kind of brand strategy work she spent her thirties learning to produce.

The senior consultant who realizes the deck-building skill that earned him promotion is now a 90-second prompt.

The legal partner who discovers junior associates using AI to draft contracts in a fraction of the time it used to take a team.

The CIO whose architectural expertise is being second-guessed by AI tools their teenagers use better than they do.

The CFO who watches AI produce financial analysis that would have taken his team three weeks.

Every one of these senior leaders is, in some structural sense, being told what Stefani Germanotta was told in January 2007: the system that defined your professional identity has decided, for reasons it can’t quite articulate, that you might not be what comes next.

In that moment, the question every one of them is quietly asking themselves is the same question Gaga was asking in early 2007.

What do I do now?

The five lessons from her wilderness year, properly translated, are the most useful possible answers.

Five lessons, translated for the executive moment

One: The rejection isn’t always about you. Sometimes the system doesn’t understand what it’s rejecting yet.

Def Jam didn’t drop Stefani Germanotta because she lacked talent. They dropped her because their machinery didn’t know what to do with her. The system was optimized for a different kind of artist than she was. The mismatch was structural, not personal.

For senior leaders going through AI transformation, the same is often true. The “rejection” you’re feeling — the obsolescence anxiety, the sense that your skill set is being deprioritized — is sometimes a true signal that the world has moved past your skill. More often it’s a signal that the world hasn’t yet figured out how to use your skill in the new configuration.

The discipline is to keep distinguishing between the two.

Two: Keep producing through the wilderness. The wilderness is where the work compounds.

Gaga’s eighteen months on the Lower East Side weren’t a recovery period. They were the most productive period of her early career. She wrote dozens of songs. She refined the persona that would become Lady Gaga. She built the network of collaborators — Lady Starlight, RedOne, Akon — who would shape her sound.

The wilderness wasn’t a pause before the work. It was the work.

For executives, this is the most important reframe of the AI transformation. The period you’re in right now — where you don’t yet fully understand the technology, where your role is shifting, where your authority feels less stable — is not a pause before the future. It is the work that produces the future version of you.

Keep producing through it. Write the documents. Have the conversations. Run the experiments. Build the new fluency. The compounding happens in the wilderness, not after it.

Three: Let the rejection become the material.

Some of Gaga’s most defining early work — the songs about fame, the persona built around the rejection of the rejection — came directly from the experience of being told she wouldn’t make it.

The rejection wasn’t a story interruption. It was the material she used to build the next version of her career.

For senior leaders, this is the underrated executive coaching move of the AI moment. The leaders who write honestly about what they don’t yet know about AI, who admit publicly that they’re learning, who treat their own transformation as part of the curriculum — those are the leaders developing the most distinctive voices right now.

The leaders who pretend to already know are the ones who sound generic. The ones who let the difficulty show, with discipline and rigor, are the ones whose careers will compound through this period.

Four: The systems that reject you in year one are often the same systems that try to claim you in year five.

This is the most quietly satisfying lesson, and also the most strategically important.

The same Def Jam that dropped Stefani Germanotta in January 2007 would have done anything to have her back in 2009. The same NYU classmates who made the “you will never be famous” Facebook group eventually attended her concerts. The same industry gatekeepers who decided she wasn’t viable were, two years later, claiming her as one of their own.

For senior leaders, this is the play. The same boards that aren’t sure what to do with you in 2026 will be the boards that desperately need your leadership through the AI transformation in 2028 and 2029 — if you’ve spent the intervening time building real fluency, real perspective, real distinctive voice.

The reward isn’t immediate. The reward is structural. The systems catch up.

Five: Self-belief is a discipline, not an emotional state.

This is the deepest lesson, and the one most retellings of the Gaga story flatten into something useless.

She did not, by every available account, feel confident every day in 2007. She was struggling. She was using drugs. She was estranged from her father. Her finances were precarious. The professional world had told her she wasn’t going to make it.

The discipline that carried her wasn’t the absence of those feelings.

It was the practice of producing regardless of them.

For senior leaders, this is the only useful translation. You will not feel confident every day in 2026 and 2027. Some days the AI transformation will feel exhilarating. Other days it will feel like everything you’ve built is dissolving. Some weeks your judgment will feel sharp. Other weeks you will feel obsolete.

The discipline is not feeling confident. The discipline is showing up regardless.

The work compounds independent of the emotional state.

That’s the entire mechanism.

The closing thought

There is a sentence Lady Gaga has used in interviews over the years, when asked about the wilderness period, that I want to close with, because it captures the actual texture of what carried her through 2007 better than any motivational version of the story ever could.

She has said, in different forms, the same essential thing: that she knew she had to work harder than everyone else, and that there was no plan B.

Read that slowly.

Not “I believed in myself.”

Not “I trusted the universe.”

Not “I knew I was meant for this.”

The work, harder than everyone else, with no fallback.

That is the entire story.

She wasn’t extraordinary in January 2007. She had been dropped. She was broke. Her father wasn’t speaking to her. She was performing in burlesque clubs for tips. Her NYU classmates were posting about her on Facebook. Def Jam had decided she was finished.

What was extraordinary was that she kept showing up.

For eighteen months. Every day. Through one of the hardest periods of her life.

That is not a story about believing in yourself.

That is a story about being too disciplined to quit while the world catches up.

In 2026, in the middle of an AI transformation that is, in some specific way, telling every senior leader they might not be what comes next, this is the single most useful career narrative I can point at.

Not the proving-them-wrong version. The actual version.

The discipline of producing through rejection.

The willingness to keep working in the wilderness because the wilderness is where the work happens.

The trust that the systems that reject you now may need you in five years — if you’ve spent the five years actually doing the work.

The acceptance that self-belief is downstream of practice, not the other way around.

This is what carried Stefani Germanotta from the Def Jam drop in January 2007 to “Poker Face” being the best-selling single in the world in 2009.

It is what will carry the executives who actually make it through the AI transformation from where they are now to where they need to be.

There is no plan B.

There is just the work.

The world has changed. The leaders who notice will be the ones the next decade is built around.

The post Carry Into the Age of AI. first appeared on Agile Agilist | SAFe® Gold Partner for Agile Transformation, Innovation & Leadership Training.

The Most Precise Metaphor We Have for What AI Transformation Actually Demands — and Why the Discomfort You’re Feeling Is Biological, Not a Failure.

There is a bird in Alaska right now, on a tidal mudflat, eating itself.

Not metaphorically. Through an actual cellular process called autophagy — Greek for “self-eating” — the bar-tailed godwit is breaking down its own gizzard, liver, kidneys, and intestines. It is reducing those organs by up to 25%, while simultaneously growing its pectoral muscles and heart.

It is doing this because, in a few days, it will take off from the western coast of Alaska, fly south across the open Pacific Ocean for between 8 and 11 days without stopping, and land in New Zealand. Or sometimes in eastern Australia. Or, on one extraordinary occasion documented in 2022, in northeastern Tasmania.

The current world record, set by a 5-month-old juvenile bird tracked by satellite during its first migration ever, is 13,560 kilometers — 8,435 miles — flown nonstop over 11 days.

No food. No water. No rest. No landing on the ocean. No feeding at sea. 264 continuous hours of flapping flight, day and night, across an ocean that the juvenile bird had never seen before.

This is one of the most extraordinary physiological feats in the natural world. And I think it is the single most precise metaphor we have for what executive transformation in the age of AI actually demands.

Most metaphors for organizational change are bad. They are usually either too gentle (gardening, journeying, climbing a mountain) or too violent (war, surgery, demolition). They don’t capture the specific texture of what real transformation feels like to the people doing it.

The bar-tailed godwit captures it precisely. Because what the godwit is doing is, in the most literal possible sense, biologically uncomfortable. The bird is consuming parts of itself. It is rebuilding its own organism, on the fly, for a journey that the organism it used to be could not have survived.

The biology, as we actually understand it

The bar-tailed godwit’s transformation was documented in detail by two scientists: Theunis Piersma, a professor of global flyway ecology at the University of Groningen in the Netherlands, and Robert Gill, a research biologist at the USGS. Their foundational 1998 paper, published in the ornithological journal The Auk, has one of the great deadpan titles in scientific literature: “Guts don’t fly: small digestive organs in obese Bar-tailed Godwits.”

The paper documented something that, at the time, biologists weren’t quite ready to accept. The godwits being captured immediately before their southbound migration from Alaska had drastically reduced digestive organs compared to what was considered normal for their species. The gizzards were smaller. The livers were smaller. The kidneys were smaller. The intestines were smaller. By up to 25%, in some cases.

At the same time, the same birds had enormous fat reserves. Up to 55% of their body mass was fat — the highest relative fat content ever recorded in any bird species. Dr. Christopher Guglielmo, another leading researcher in migratory bird physiology, later coined the phrase “obese super athletes” to describe them.

The bird wasn’t just carrying fuel for the flight. The bird was reconfiguring its own body for the flight.

After landing, within days, the organs regenerate completely. The bird that arrives in New Zealand is not the same bird that left Alaska. The bird that recovers in New Zealand is not the same bird that arrived. The organism reconfigures itself three times in roughly two months — once to prepare for the flight, once to make the flight, and once to recover from it.

A typical bar-tailed godwit will fly approximately 460,000 km over its lifetime — more than the distance from the Earth to the Moon.

This is not metaphor. This is biology.

The seven lessons for AI transformation

One: You cannot carry everything across.

The most basic biological reality of the godwit’s transformation is that the bird cannot fly with its full digestive system AND its maximum fat reserves AND maintain enough lift to cross an ocean. The math doesn’t work. Something has to be reduced for the journey to be possible at all.

Most senior leaders, faced with AI transformation, are trying to violate this constraint. They want to keep their old organizational structure, their old decision-making processes, their old identity as a leader, their old career-defining expertise — and add AI on top of all of it. They want the transformation to be additive.

It can’t be. The math doesn’t work.

Something has to be consumed. The question is not whether to consume, but what.

Two: The shrinkage is strategic, not arbitrary.

The godwit doesn’t shrink random organs. It shrinks the digestive organs it won’t need during the flight, and it simultaneously grows the pectoral muscles and heart it will need for the flight. The reconfiguration is purposeful.

Most enterprises do this wrong. When budgets tighten during an AI transformation, the cuts almost always fall on the AI investment itself — on the new capability being built. That is cutting the flight muscle. It is the exact opposite of what the godwit does.

The right cuts, biologically, are to the parts of the organization that won’t be useful on the other side of the journey. The legacy process documents that nobody will follow once the new system is live. The approval chains designed for a world where decisions took weeks. The reporting structures that exist to support a hierarchy AI is going to flatten.

Three: The discomfort is biological, not psychological.

This is the lesson that matters most for senior leaders currently going through AI transformation.

The bird is literally consuming parts of itself. That is not metaphor for discomfort. That is discomfort, at the cellular level. The transformation is uncomfortable because it is biologically uncomfortable. There is no version of the journey that doesn’t involve the autophagy.

Most leaders treat the discomfort of AI transformation as a sign that something is going wrong. They look at the resistance, the confusion, the loss of identity, the senior people who can no longer locate their value, the teams whose old work has become obsolete, the cultural strain — and they conclude that their change management plan failed.

The godwit’s biology says: no, this is what a successful transformation feels like from inside the organism going through it. The discomfort isn’t a bug. It is the necessary cost of making the journey at all.

Four: Regeneration happens — but only after the journey.

The godwit’s organs do not regenerate during the flight. They regenerate within days of landing.

If you are a senior leader currently in the middle of an AI transformation and you feel diminished, uncertain, less authoritative than you used to be, less sure of what you uniquely contribute — that is exactly where you are supposed to be. The bird’s intestines don’t return to full size while it’s still over the Pacific. They return after it lands.

The discipline is to let the old identity shrink during the flight, trust that it will regenerate in a new form after landing, and stop trying to perform a role that the journey itself is in the process of dissolving.

Five: The fat is the fuel — but only if you’ve earned it.

The godwit spends two months feeding before takeoff. The journey is biologically impossible without the prior work. A godwit that has not built up sufficient fat reserves cannot make the flight, full stop.

The leaders who succeed in AI transformation are, almost without exception, the ones who built reserves before the journey began. Reserves of credibility with the board, with the executive team, with the workforce. Reserves of personal capacity to absorb discomfort without losing their judgment. Reserves of relationships that will hold under stress.

The two months of feeding before departure isn’t optional. It is the prerequisite that makes the journey possible at all.

Six: There is no stopover.

The bar-tailed godwit cannot rest on water. It is not a seabird. Its feathers aren’t waterproof in the way required for floating. If it lands on the ocean, it drowns. It also cannot feed at sea.

Once the journey begins, it must be completed in one continuous effort. There is no rest. There is no refueling. The 8 to 11 days have to be done in a single push.

Most enterprises are trying to do AI transformation as a series of small pilots with rest periods in between. They run a pilot. They evaluate. They take a quarter to absorb the results. They run another pilot. They take another quarter.

This is not how the journey works. At some point, the executive team has to commit to a continuous transformation effort that cannot be paused. The competitive environment is moving. The technology is moving. The workforce expectations are moving. An organization that pauses for two quarters between AI pilots is, by the time it resumes, working with a different ocean than the one it started across.

Seven: The young ones make it on the first try.

The world record godwit migration in 2022 — 13,560 km from Alaska to Tasmania, the longest documented non-stop bird flight in history — was made by a 5-month-old juvenile on its very first migration.

The bird had no prior experience of the journey. It had no mentorship from older birds during the flight. It had never seen the Pacific Ocean, New Zealand, Tasmania, or any of the islands it might pass on the way. It carried no instruction from its parents about how the crossing worked.

It made the flight anyway. Alone. At five months old. Across an ocean. To a continent it had never seen.

It survived because the capability was in the species, not in individual experience.

This is the lesson that I think matters most for senior leaders looking at the younger generation of AI-native workers in their organizations. The 25-year-olds without MBAs, without decades of industry experience — many of them are making professional crossings right now that veteran leaders are insisting are impossible.

They’re not insisting. They’re doing it. Because the capability isn’t in the experience. It’s in something more fundamental — the fluency with the new tools, the absence of conditioned resistance to new ways of working, the willingness to commit to a journey without a map because they were never given the old map in the first place.

The closing reflection

There is one more detail about the bar-tailed godwit that I want to leave you with, because I think it is the deepest part of the metaphor.

The godwit does not make this journey once. It makes the journey every year. For its entire adult life. A bird that lives 15 to 20 years will reconfigure its body — eating its own organs, growing its own muscles, crossing the ocean, regenerating on the other side — between 15 and 20 times. The transformation is not a one-time event. It is an annual capability that the species has metabolized into its biology.

The enterprise lesson is the most uncomfortable of all the lessons, and I want to say it directly.

AI transformation isn’t a project. It isn’t a one-time crossing that you complete and then return to normal operations afterward.

It is the new annual cycle of how organizations function.

The companies that survive the next decade will be the ones that learn to do what the godwit does — strategically consume parts of their old structure to enable the journey, grow the muscles they need for the crossing, commit to continuous flight without rest, trust that regeneration happens after landing, and then, when the next migration comes, do it all again.

Theunis Piersma, who has spent his career documenting how migratory birds reshape themselves for their journeys, has written that what these birds do is not “endurance” in the way humans usually mean that word. It isn’t pushing through pain. It is, biologically, the organism becoming a different organism for the duration required, then becoming itself again on the other side.

That, I think, is the most precise possible description of what AI transformation actually demands of any senior leader serious about making it.

You become a different leader for the duration required.

You consume the parts of your old self that won’t fly.

You grow the muscles you need for the crossing.

You make the flight continuous, without rest, until you land.

You regenerate on the other side, into a configuration that did not exist when you began.

And then, when the next migration comes — and it will come, on a faster cycle than any previous era of change has demanded — you do it all again.

This is uncomfortable. It is supposed to be uncomfortable. The discomfort is the biological cost of the journey, not a sign that the journey is going wrong.

The bird in Alaska right now, eating its own organs to fuel a 12,000 km flight across the Pacific, doesn’t experience the autophagy as failure. It experiences it as preparation.

The senior leaders who learn to read their own discomfort the same way are the ones who will reach the other side.

The world has changed. The leaders who notice will be the ones the next decade is built around.

The post The Most Precise Metaphor We Have for What AI Transformation Actually Demands — and Why the Discomfort You’re Feeling Is Biological, Not a Failure. first appeared on Agile Agilist | SAFe® Gold Partner for Agile Transformation, Innovation & Leadership Training.

Career Advice for the AI Transformation.

In 1967, a Maltese physician and psychologist named Edward de Bono published a book titled New Think: The Use of Lateral Thinking. He had recently coined the term “lateral thinking” — a phrase that would, over the next half-century, become one of the most quoted pieces of management vocabulary in the English language.

The Oxford English Dictionary added it. Business schools built curricula around it. Thousands of corporate training programs used it.

Most of them missed the point.

De Bono spent the rest of his life — he died on June 9, 2021, at the age of 88 — trying to explain what he actually meant. He wrote more than 60 books. He consulted with governments and corporations across forty countries. He developed the Six Thinking Hats methodology, the Po provocation technique, and dozens of other tools.

But the core idea, the one that mattered, can be compressed into a single sentence he repeated in interviews and books for fifty years:

“You cannot dig a hole in a different place by digging the same hole deeper.”

This piece is about why that sentence is, in my opinion, the single most useful piece of career advice available to any senior leader navigating AI transformation in 2026 — and why almost nobody is actually applying it.

What de Bono was actually saying

The standard reading of de Bono is that “lateral thinking” means creative problem-solving. Think outside the box. Try unconventional approaches.

That reading is true but shallow. The deeper insight is about how the human mind works.

De Bono’s central claim was that the brain is, fundamentally, a pattern-recognition and pattern-following system. When you encounter a problem, your mind doesn’t reason from first principles. It rapidly retrieves the most familiar framework for dealing with similar problems and routes your thinking along that established pathway.

This is enormously efficient. It is also enormously limiting. Because the most familiar pathway is not always the most productive one.

De Bono called the conventional logical approach vertical thinking. It moves step by step from one premise to the next, deeper along an established pathway. It is the dominant mode of analytical reasoning that schools, professions, and corporations have been training for centuries. It is excellent for problems where the right pathway has already been identified.

He contrasted this with lateral thinking. Rather than digging deeper along the established pathway, lateral thinking deliberately steps off it and approaches the problem from a different angle, generating alternative entry points that vertical thinking would never encounter because they lie outside its trajectory.

The famous quote — “you cannot dig a hole in a different place by digging the same hole deeper” — is the compressed expression of this distinction. If the gold isn’t where you’re digging, no amount of additional vertical effort will produce gold. The only useful move is to stop digging and start somewhere else.

The five-feet-of-dirt problem

The analogy de Bono used to make this land with non-academic audiences was specific and visual.

Imagine you’re digging for gold. You’ve gone five feet down. No gold.

What do you do?

The intuitive answer for almost everyone — and especially for anyone trained in American business culture — is to dig faster. To intensify the effort. To bring in more shovels. To work the night shift. To execute relentlessly.

Sometimes this is right. If you have good reason to believe the gold is at ten feet, dig faster. If you’re in the right hole, the answer is more depth.

But sometimes — and de Bono argued, more often than people are willing to admit — the trouble isn’t that you’re digging too slowly.

The trouble is that you’re digging in the wrong place.

The instinct to dig faster, when you’re in the wrong place, is not just useless. It is structurally counterproductive. It uses up the energy, attention, and capital you would need to dig somewhere else. It deepens your investment in a location that isn’t going to pay out. It produces the appearance of progress without any of the substance.

This is the problem de Bono spent five decades trying to surface for executive audiences.

His repeated diagnosis: in business, in politics, in personal life, the dominant failure mode is not laziness or insufficient effort. It is highly competent vertical thinking applied to the wrong hole.

The personal observation that surfaces the framework

There is an observation about senior leadership that I think is one of the most underexamined experiences in modern executive life.

When I’m in the right context, I feel incredibly smart and capable. When I’m in the wrong context, that feeling disappears.

This is one of the sharpest articulations of an executive insight available, and it points directly at the de Bono framework.

The implicit promise of senior leadership, the way most executives have been taught to think about themselves, is that competence is portable. You are smart, or you aren’t. You are capable, or you aren’t. The capability travels with you across contexts because the capability is yours.

De Bono’s framework, supported by half a century of cognitive research that has confirmed his original intuition, says this isn’t true.

Capability is largely a function of context.

When the context matches the patterns your mind has been trained on, you appear — and feel — extraordinarily capable. The pattern-recognition system retrieves the right framework, the right vocabulary, the right judgment, the right examples, and routes your thinking productively. You make good decisions. You see things others miss. You feel sharp.

When the context shifts — when the problem doesn’t match the patterns you’ve trained on — the same brain that felt sharp now feels slow. You hesitate. You second-guess. You feel obsolete or out of your depth.

This is not a failure of intelligence. It is a structural feature of how the brain works.

For senior leaders, the implication is enormously liberating if you’re willing to accept it: when you feel less capable than you used to, you are not necessarily becoming less capable. You may simply be in the wrong context. The capability hasn’t gone anywhere. The patterns you trained for thirty years are mismatched with the patterns the new context demands.

The de Bono move, in this situation, is not to work harder at being capable in the wrong context.

The move is to change context.

Why this matters specifically in 2026

The reason I think this framework is the single most useful piece of career advice for senior leaders in 2026 is that the AI transformation is, for almost every senior executive I work with, a context shift.

The patterns they trained for thirty years — the patterns that made them feel incredibly capable in 2018, 2019, 2020 — are mismatched with the patterns the AI-augmented world demands.

The CMO who spent twenty years mastering brand strategy is sitting in meetings where AI is producing brand strategy in seconds.

The senior consultant who built her career on synthesizing complex client situations into clear strategic decks is watching clients use AI to produce equivalent decks in 90 seconds.

The CIO who spent his career mastering enterprise architecture decisions is being second-guessed by junior engineers using AI tools he doesn’t fully understand.

The CFO whose value was built on producing financial analysis from scarce data is operating in a world where data is infinite and analysis is automated, and the constraint has shifted to judgment, which is exactly the part of his role he most assumed would always be his.

Every one of these leaders has, by every objective measure, retained their underlying capability. They are not less intelligent than they were five years ago. They have not lost their experience, their judgment, their networks, their credentials.

They are digging in the wrong hole.

And every instinct trained by thirty years of corporate culture — the lean-in instinct, the execute-relentlessly instinct, the double-down instinct — is telling them to dig faster.

The de Bono diagnosis says this is precisely the wrong move.

The right move is to stop digging where they’ve been digging and move to a different location.

The sunk cost of the deep hole

The reason this is so hard to do, in practice, is that the hole you’ve been digging is deep. It represents twenty or thirty years of accumulated expertise. Credentials. Network. Identity.

When de Bono says “move to a different hole,” he is asking executives to walk away from a hole that took them their entire career to dig.

That feels, viscerally, like wasted effort.

It isn’t.

The capability that produced the depth of the old hole — the discipline, the pattern recognition, the judgment, the willingness to work hard at something for years — is the same capability that will produce the new hole. Faster. In a better location.

But you have to be willing to leave the depth behind.

This is, in my coaching experience, the hardest single conversation a senior leader has to have with themselves. Because the sunk cost feels enormous. The identity is bound up with the depth. The professional reputation rests on the depth. And the new hole, by definition, is shallow.

Most senior leaders, faced with this choice, choose to dig their existing hole deeper. They tell themselves the gold must be at fifteen feet, or twenty feet, or thirty. They redouble. They lean in. They execute relentlessly.

They produce, in some cases, the deepest hole in their industry.

And they fail anyway, because there is no gold there.

The de Bono framework is the diagnostic that surfaces this failure mode before it becomes terminal. Not by criticizing the depth of the hole — the depth was real, the effort was real, the expertise is real. But by asking the question that vertical thinking is structurally unable to ask:

Is this the right hole?

What the lateral move actually looks like

The misreading of “lateral thinking” that has dominated business culture is that it means radical creative reinvention. Quit your career. Start a band. Move to Bali.

That’s not what de Bono meant, and it’s not what the framework requires.

A lateral move, in the de Bono sense, is a sideways shift that preserves the underlying capability while changing the location of the work.

The CFO doesn’t quit finance. She shifts from producing analysis to designing the AI-augmented decision architecture that produces analysis. Same financial judgment, different hole.

The CMO doesn’t abandon brand work. He shifts from producing brand strategy to orchestrating an AI-augmented creative process that produces brand strategy. Same taste, different hole.

The senior consultant doesn’t leave consulting. She shifts from being the analyst who builds the decks to being the orchestrator of AI-assisted client engagement. Same client intuition, different hole.

The CIO doesn’t abandon technical leadership. He shifts from architecting systems to architecting the integration of AI into systems. Same architectural judgment, different hole.

In every case, the capability that produced the depth of the old hole transfers. The new hole isn’t a betrayal of the career — it’s a continuation of it, in a location where the underlying capability can still compound.

The leaders who recognize this in 2026 will be the leaders who define the next decade.

The leaders who keep digging the old hole, however deep, will spend the next decade wondering why their effort is producing diminishing returns while less talented peers seem to be effortlessly moving ahead.

The less talented peers aren’t more talented. They’re in a different hole.

Three practical questions

If you want to actually apply the de Bono framework to your own career in 2026, three questions are worth sitting with.

One: what is the central pattern around which your senior career has been built?

Not “leadership” or “strategy” — these are too abstract to be useful. The actual concrete pattern. The specific kind of problem you have spent the last decade getting good at solving. The recurring situation in which you feel most capable, most confident, most valuable.

That pattern is your hole. Identify it precisely.

Two: has the value of that pattern changed in your industry in the last three years?

Be honest. Not whether you still find the pattern useful — whether the market still rewards that pattern at the price it used to. In some industries the pattern is still valuable. In many, the price has quietly collapsed because AI now produces the same pattern at a fraction of the cost.

If the answer is “the price has collapsed,” you are still digging at five feet. The gold has moved.

Three: what lateral move preserves your underlying capability while changing the location of the work?

Not a complete career reinvention. A sideways shift. The same expertise applied to a different problem, where the value is still compounding. Where the pattern you’ve trained on is still rare and useful.

That’s the lateral move. It will feel uncomfortable. The new hole will be shallow for a few years. The depth will take time to rebuild.

But the alternative is digging the old hole deeper, in a location where there is no longer any gold.

The closing thought

There is a phrase from de Bono’s later work that I want to close with, because it captures something his motivational-poster admirers usually miss.

He wrote, in I Am Right, You Are Wrong (1990), that the deepest problem with conventional intelligence is not that it produces wrong answers. It is that it produces highly competent answers to questions that should have been asked differently.

“It is not enough to do your best; you must know what to do, and then do your best.”

That is the framework, compressed.

The American business culture of the last forty years has been overwhelmingly focused on the second half of that sentence. Do your best. Execute relentlessly. Work harder. Lean in. Double down.

The first half — know what to do — has been quietly underweighted, because it is harder to teach, harder to measure, and harder to perform publicly.

You can show effort. You can’t easily show that you’ve correctly identified which hole to dig.

The AI transformation is forcing the first half back into the center of executive practice. Because in 2026, executing relentlessly on the wrong hole produces nothing. The pattern that worked five years ago has, in many domains, no longer the right pattern to be reinforcing.

The leaders who will define the next decade are the ones who, right now, are asking the harder question.

Am I in the right hole?

If the answer is yes, dig harder. Use every ounce of accumulated capability to go deeper.

If the answer is no, stop.

Move sideways. Start a new hole. Trust that the capability that produced the depth of the old hole will produce the new one, in a better location, with the years you have left.

De Bono spent fifty years teaching this. Most people who quoted him missed the point.

In 2026, with AI rewriting the location of value in almost every senior career, the point is no longer optional.

You will not feel as capable in the new hole, at first. The shallow depth will feel like regression. The familiar patterns of the old hole will keep pulling at you.

That feeling — the wrong-context feeling — is the most important piece of executive information available to you right now. It is the signal that the hole has changed.

The discipline is to read the signal honestly, to stop digging, and to start somewhere else, while you still have the energy to make the depth count.

The world has changed. The leaders who notice will be the ones the next decade is built around.

The post Career Advice for the AI Transformation. first appeared on Agile Agilist | SAFe® Gold Partner for Agile Transformation, Innovation & Leadership Training.

Charlie Munger’s Can Carry Into the Age of AI

In 2000, Janet Lowe published a biography of Charlie Munger titled Damn Right! Behind the Scenes with Berkshire Hathaway Billionaire Charlie Munger. Buried in it is one of the most-quoted passages in modern business literature — a passage that has, in the twenty-five years since, been reduced to a motivational poster by people who didn’t read the chapter it came from.

Here is the passage, in full:

“Whenever you think that some situation or some person is ruining your life, it’s actually you who are ruining your life. It’s such a simple idea. Feeling like a victim is a perfectly disastrous way to go through life. If you just take the attitude that however bad it is in any way, it’s always your fault and you just fix it as best you can—the so-called ‘iron prescription’—I think that really works.”

Munger called this the iron prescription. He returned to it in talks, interviews, and shareholder letters for the rest of his life. It is, alongside Buffett’s writings on investing temperament, one of the foundational psychological statements of late-20th-century American business.

Most people who quote it treat it as a tough-love motivational line. Take responsibility. Stop blaming others. The basic stoic move.

I think this reading dramatically underestimates what Munger was actually saying. And I think the original context — which almost no one includes when they quote the iron prescription — is the entire reason the principle works.

This piece is about that context, what Munger was really teaching, and why it is the single most useful psychological technology any senior leader can carry into the age of AI.

The context most retellings strip out

In 1953, Charlie Munger was 29 years old. He had been married since he was 21. That year, his wife divorced him. The settlement was severe enough that he lost the family home in South Pasadena and moved into what biographers describe as “dreadful” conditions at the University Club of Pasadena.

Shortly after the divorce, his nine-year-old son Teddy was diagnosed with leukemia. In 1953, leukemia in children had no effective treatment. The death rate was nearly 100%. There was no health insurance. Munger paid everything out of pocket as his savings drained.

Rick Guerin, Munger’s friend from that period, described him going to the hospital, holding his dying son, and then walking the streets of Pasadena crying.

In 1955, when Munger was 31, Teddy died.

Divorced. Broke. Burying his nine-year-old son.

If anyone in modern American business history had earned the right to declare themselves a victim of fate, it was Charlie Munger in the autumn of 1955.

He refused.

What happened next is the part of the Munger story that gets told, but usually without the context above. He rebuilt his law career. He remarried — to Nancy Borthwick, whom he stayed married to for over half a century. They raised a blended family of eight children together. He started investing seriously. He met a young Omaha investor named Warren Buffett. They began what would become one of the most successful business partnerships in human history.

Munger died on November 28, 2023, thirty-three days before his 100th birthday. He had lived 68 years past the worst year of his life.

The iron prescription wasn’t a slogan from someone who hadn’t been tested.

It was a survival technology forged in catastrophic loss.

That distinction matters because almost every glib reading of the iron prescription misses the source of its authority. Munger wasn’t moralizing about other people’s resilience from a position of comfort. He was reporting what he had personally discovered, in 1955, was the only mental position that allowed him to keep functioning.

What Munger was actually teaching

The standard reading of the iron prescription is “take responsibility, stop blaming others.” This is true but incomplete.

The deeper reading is about agency preservation.

Munger’s discovery — and this is what most retellings miss — wasn’t that victims are morally wrong. It’s that victimhood is strategically wrong. It is the one mental position that guarantees you cannot improve your situation.

The moment you decide that some external force is ruining your life, you have located the cause of your problem outside yourself. By definition, you cannot fix something that exists outside yourself. You have surrendered the only lever you actually have.

This is not about moral judgment. It is about what mental positions preserve your ability to act.

The iron prescription is the discipline of always assuming, as your operating frame, that the situation is your fault — not because it always is, but because that’s the only frame that keeps you capable of fixing it.

Even when the situation is genuinely caused by external forces. Even when other people are clearly to blame. Even when the world is genuinely unfair. The iron prescription says: start with what is yours, because that’s the only place you can act from.

This is structurally identical to what Viktor Frankl wrote in Man’s Search for Meaning, which Munger explicitly recommended throughout his life. Frankl, who lost his entire family in the Holocaust, wrote that the one thing the camps could not take from him was his ability to choose his response to what was happening.

It is structurally identical to the central insight of Stoicism. Epictetus, born a slave, wrote that what disturbs people is not events themselves but their judgments about events.

It is structurally identical to the foundation of cognitive behavioral therapy. Aaron Beck’s 1960s work on cognitive distortions identified externalization — locating the cause of one’s distress outside oneself — as one of the central thinking patterns associated with depression and learned helplessness.

The iron prescription is, in this sense, not original to Munger. What was original was his willingness to articulate it bluntly, in business language, after having paid for it with the worst losses a human being can face.

Why this matters specifically in the age of AI

The reason I want to write about Munger’s iron prescription now, in 2026, is that I think it is the single most useful psychological technology any senior leader can carry into the age of AI.

The temptation to externalize has never been higher.

For thirty years, business culture has had a relatively stable set of explanations for failure. Bad strategy. Wrong people. Insufficient capital. Market conditions. Regulatory environment. Competitor actions. These explanations were useful because they pointed at things you could actually fix.

The arrival of AI has given every senior leader an entirely new menu of externalization options. Every one of them sounds reasonable. Every one of them strips agency.

I see five common patterns. Let me walk through each.

Pattern One: “AI is going to make my role obsolete.”

The externalization version: AI is making my expertise obsolete.

The iron prescription version: You haven’t yet decided what you uniquely contribute that AI doesn’t. That’s not AI’s fault. That’s a thinking problem you haven’t done.

The honest reality is that every senior leader’s role is being transformed by AI, and the leaders who will thrive are the ones who have actively done the work of figuring out which parts of their value are augmented by AI versus which parts are commoditized by it. Nobody else can do this work for them.

Pattern Two: “Our AI transformation isn’t working because the vendor underdelivered.”

The externalization version: The vendor failed.

The iron prescription version: You chose the vendor. You wrote the contract. You set the success metrics. You signed off on the implementation plan. Where is the part that’s actually yours?

The structural reality of every enterprise AI program I have seen succeed is that the executive sponsor took ownership of the program’s success in a deep and personal way — including for the parts that other people were nominally responsible for.

Pattern Three: “My industry is being disrupted by AI.”

The externalization version: My industry is being disrupted. This is happening to me.

The iron prescription version: The warning signs were everywhere for at least three years. What did you do during those three years, and what are you going to do in the next three?

Pattern Four: “AI hallucinated and gave us the wrong answer.”

The externalization version: AI hallucinated.

The iron prescription version: Your team used an AI tool they didn’t fully understand, on a high-stakes deliverable, without verification. The error is structurally yours, not the tool’s. You didn’t build the verification habits. You didn’t train your team on what AI is and isn’t reliable for.

Pattern Five: “Our people aren’t AI-ready.”

The iron prescription version: You didn’t invest in training at the scale required. You didn’t model the behavior yourself. You didn’t make AI fluency a meaningful factor in how you hire, promote, and reward people. The readiness gap is one you created through years of optional engagement.

The pattern across all five

In every one of these five cases, two things are true.

The externalization version is, in some real measure, factually accurate.

The iron prescription version is also accurate, and it’s the only version that preserves your capacity to act.

This is the deepest thing Munger was teaching, and it’s why I think the iron prescription is so much more sophisticated than the motivational-poster version of it.

He wasn’t denying that external causes are real. He was pointing out that the frame of starting with your own contribution is the only frame that produces resilient action.

You can hold both truths at once. The vendor genuinely did underdeliver, and you chose them. The technology is genuinely imperfect, and you put it in production without verification habits. The industry is genuinely being disrupted, and you had three years to prepare and didn’t.

Holding both truths is the discipline. Defaulting to the second one — your own contribution — is the iron prescription.

The limits, honestly

It would be dishonest not to flag the limits of this principle, because the iron prescription is genuinely dangerous when applied as a moral judgment of other people’s circumstances.

The iron prescription is a personal discipline for people with agency. It is not a moral framework for evaluating people without agency. A single mother working two minimum-wage jobs in a city with no affordable housing is not failing to apply the iron prescription. A 14-year-old growing up in a war zone is not failing to apply the iron prescription.

Munger’s framework is best understood as a gift you give yourself — a way of holding your own circumstances that maximizes your agency — not a judgment you impose on others.

The senior leaders I work with are, almost without exception, people with substantial agency. They have resources, education, networks, options, capital, and credibility. They are squarely in the population for whom the iron prescription is the most useful possible mental technology.

If you are a senior leader reading this, the iron prescription is for you. You have the agency. The question is whether you’ll use it.

The practical discipline

If you want to actually apply the iron prescription in 2026, here is the simplest possible version.

Every time you catch yourself saying — out loud or internally — that something or someone is ruining something about your work, run the translation.

The original thought: “X is ruining Y for me.”

The translation: “What am I doing, or not doing, that is making this worse — and what am I going to do differently this week?”

You will not always have an answer. Some situations are genuinely external. Some weeks you will run the translation and conclude that there is, in fact, nothing more you can do this week, and that the situation is mostly outside your control.

Even in those cases, the discipline is doing work. The habit of looking inward first preserves your capacity to act, even when looking inward doesn’t yield an immediate answer.

Over months and years, this discipline reshapes how you carry yourself. You stop being a person who is constantly being acted upon by circumstances. You start being a person who is constantly looking for the leverage that is actually available to you.

That second person is the kind of senior leader the next decade rewards.

The first person is the kind the next decade quietly retires.

The closing thought

There is a small, somewhat painful detail about Munger that I want to close with.

He was asked, multiple times across his life, how he survived Teddy’s death. The answers he gave varied in detail but converged on the same core idea. He said something to the effect of: self-pity is a guaranteed way to make a bad situation worse, even when the situation is unimaginable. So I just refused to engage with it.

He was not denying that the loss was catastrophic. He was not pretending that grief wasn’t real. He was reporting a structural decision: self-pity, however justified, was the one mental position that guaranteed his life would get worse from there. So he refused it. Not because he didn’t feel the grief. Because he had decided, in advance, that the grief would not be the thing that ran his life.

That decision, made by a 31-year-old man in 1955 who had every reason to surrender to despair, is what made the next 68 years of his life possible.

The iron prescription is the same decision, generalized.

It says: whatever is happening to you, whatever the cause, whatever the apparent injustice, the only useful mental position is to ask what part of this is yours and what you’re going to do about it.

In the age of AI, with the temptation to externalize at an all-time high, with the menu of plausible excuses growing wider every quarter, this is the most useful single discipline any senior leader can carry.

Munger paid for the insight in a way none of us would ever wish on ourselves.

We get to use it for free.

The world has changed. The leaders who notice will be the ones the next decade is built around.

The post Charlie Munger’s Can Carry Into the Age of AI first appeared on Agile Agilist | SAFe® Gold Partner for Agile Transformation, Innovation & Leadership Training.

AI Adoption Statistics 2026: How Fast Is AI Being Adopted?

AI adoption crossed a threshold in the last two years: using AI is no longer what sets a company apart — capturing value from it is. This page collects the most-cited AI adoption statistics for 2026, each linked to its primary source and dated. We update it as new data lands.

Key takeaways

  • 88% of organizations now use AI in at least one business function — up from 78% a year earlier.
  • 23% are already scaling agentic AI, and another 39% are experimenting with it.
  • $252.3 billion in global corporate AI investment in 2024, a 26% year-over-year increase.
  • ~900 million weekly ChatGPT users signal mass consumer adoption.
  • Despite near-universal adoption, only a minority of companies report material bottom-line impact — the adoption–value gap.
88%
organizations using AI
McKinsey, Nov 2025
23%
scaling agentic AI
McKinsey, Nov 2025
$252B
corporate AI investment 2024
Stanford HAI
900M
ChatGPT weekly users
reported, early 2026

Organizational adoption: now near-universal

88% of organizations report using AI in at least one business function, up from 78% the prior year, with roughly two-thirds now using it in more than one function. The headline number has effectively plateaued near the top — which is why the competitive question has shifted from whether a company uses AI to how well it does. Source: McKinsey QuantumBlack, The State of AI, November 2025.

From adoption to agents

The frontier of adoption in 2026 is agentic AI — systems that take actions, not just generate text. 23% of organizations report scaling an agentic system somewhere in the enterprise, and another 39% are experimenting, meaning roughly 62% are already engaged with AI agents in some form. Source: McKinsey QuantumBlack, November 2025.

The investment fueling adoption

Adoption is underwritten by capital. Global corporate AI investment reached $252.3 billion in 2024, a 26% increase over 2023, and the number of newly funded generative-AI startups nearly tripled. Collapsing inference costs — down roughly 280-fold in under two years — have made deploying AI dramatically cheaper, accelerating adoption further. Source: Stanford HAI, AI Index Report 2025.

Consumer and workforce adoption

On the consumer side, ChatGPT alone reports on the order of 900 million weekly active users in early 2026 — and it is no longer alone, with Gemini and Meta AI each claiming hundreds of millions more. Inside organizations, a majority of knowledge workers now use generative AI at least occasionally, frequently ahead of formal company rollouts. Figures reported by providers; consumer totals are self-disclosed and should be read as directional.

The adoption–value gap

The most important 2026 adoption statistic may be the one that isn’t a percentage of adopters: despite near-universal use, only a minority of companies report material, enterprise-level financial impact from AI. Adoption is easy; capturing value requires redesigned workflows, governance, and measurement. That gap — not access to the technology — is now the real differentiator. Source: McKinsey QuantumBlack, November 2025.

Related reading

Methodology & sources

Every figure on this page links to a named primary source with a publication date. Organizational statistics are drawn from McKinsey’s State of AI (November 2025); investment and cost figures from Stanford HAI’s AI Index Report 2025; consumer usage totals are self-reported by providers and labeled accordingly. This page is reviewed monthly and updated when sources publish new data.

Frequently asked questions

What percentage of companies use AI in 2026?

88% of organizations report using AI in at least one business function, up from 78% a year earlier (McKinsey, November 2025).

How many companies are using AI agents?

About 23% are scaling agentic AI and 39% more are experimenting — roughly 62% are already engaged with AI agents (McKinsey, November 2025).

If adoption is so high, why don’t more companies see results?

Adoption outpaced the harder work of redesigning workflows, governance, and measurement. Only a minority of companies report material financial impact, which is why value capture — not adoption — is the 2026 differentiator.

The State of Enterprise AI in 2026

Enterprise AI has crossed from experiment to infrastructure — but the gap between adopting AI and profiting from it has never been wider. Here is where things actually stand, drawn from the most authoritative industry data and updated as new figures land.

Last updated: June 2026. Every statistic below links to its primary source.

Adoption is now near-universal

88% of organizations report using AI in at least one business function, up from 78% a year earlier — with two-thirds now using it across multiple functions (McKinsey, The State of AI, November 2025). For context, Stanford’s data shows organizational AI use jumped from just 55% in 2023 to 78% in 2024, and generative-AI use in at least one function more than doubled — from 33% to 71% — over the same period (Stanford HAI, AI Index Report 2025).

But the value is concentrated in a few

Adoption hasn’t translated into broad financial impact. Only 39% of respondents attribute any EBIT impact to their AI use, and among those, most say AI accounts for less than 5% of EBIT. A small group of “AI high performers” — roughly 6% of organizations — is pulling away by rewiring workflows and governance around AI rather than bolting it on (McKinsey, The State of AI, November 2025).

The gap between AI adoption and AI impact has never been wider — most organizations are still experimenting without capturing meaningful enterprise-level value.

The shift to agents is underway

2025–26 is the year agentic AI moved from concept to deployment. 23% of organizations report scaling an agentic AI system somewhere in the enterprise, and another 39% are experimenting with AI agents — a combined ~62% already engaged (McKinsey, The State of AI, November 2025). Governance remains the leading concern as agents gain the ability to take actions, not just generate text.

Investment keeps surging

Corporate AI investment reached $252.3 billion in 2024, a 26% increase over 2023, while the number of newly funded generative-AI startups nearly tripled (Stanford HAI, AI Index Report 2025). The money is following capability — and increasingly, the enterprise.

The economics have fundamentally changed

One number explains much of the adoption curve: the cost of AI inference fell roughly 280-fold in under two years — from about $20 per million tokens in late 2022 to around $0.07 by late 2024 (Stanford HAI, AI Index Report 2025). What was once a budget-line question is now, for many use cases, a rounding error.

What it means for 2026

  • Adoption is table stakes. Nearly everyone is using AI; the differentiator is operational depth, not presence.
  • Value comes from rewiring, not bolting on. The 6% capturing real EBIT impact redesign workflows around AI.
  • Agents are the next battleground — and governance is the gating factor.
  • Falling costs keep widening the addressable use cases, pulling more workloads into AI economics.

Sources

AI in Sport: How Artificial Intelligence Supports Decision-Making (2026)

Artificial intelligence is no longer just powering highlight reels and fantasy stats — it is increasingly the engine behind the decisions that shape competition: which player to draft, when to press, whether a goal stands. Crucially, in almost every serious deployment a human still makes the final call; AI compresses oceans of data into something a coach, scout, or referee can act on in seconds. This report looks at how AI supports sporting decision-making, anchored in real, sourced case studies.

Share this reportXLinkedInFacebookReddit

Key takeaways

  • Roster decisions: the Toronto Raptors used IBM Watson (“Sports Insights Central”) from 2016 to automate a historically manual player-evaluation process.
  • Officiating decisions: FIFA’s Semi-Automated Offside Technology tracks up to 29 points per player, 50 times a second — but the referee still decides.
  • Broadcast & fan decisions: IBM analyzes 100,000+ data points per match at Wimbledon to generate insights, predictions, and the Power Index.
  • The market: estimates vary widely, but one projection puts AI in sports at roughly $60.8B by 2034 (~21% CAGR).
  • The constant: these are decision-support systems — human-in-the-loop is the design principle, not an afterthought.
$60.8B
projected AI-in-sports market, 2034
Precedence Research
~21%
forecast CAGR, 2025–2034
Precedence Research
29 / 50Hz
player points tracked, per second
FIFA offside tech
100k+
data points analyzed per match
IBM at Wimbledon

1. Roster & draft decisions: IBM Watson and the Toronto Raptors

In February 2016 the Toronto Raptors unveiled a “digital war room” built on IBM Watson, branded IBM Sports Insights Central. The system pulled together box-score statistics, video, medical records, and even social-media sentiment, then weighed prospects against the team’s specific needs — a player’s likelihood of succeeding, staying healthy, and fitting the locker room over the life of a contract. The point was not to replace scouts but to automate a process the Raptors had previously run with whiteboards and printed magnets, freeing decision-makers to focus on judgment. Three years later, Toronto won the 2019 NBA Championship. Source: IBM / TechCrunch, 2016.

2. In-game & coaching decisions: optical player tracking

Once a game is underway, the decisions move faster than any human can fully process. The NBA’s optical tracking system (Second Spectrum) uses AI-powered cameras to follow every player and the ball in real time, mapping defensive formations and quantifying shot quality and matchups. Coaching staffs — famously the Golden State Warriors — use this to refine rotations and tactics. The same pattern recurs across football, baseball, and tennis: computer vision turns raw movement into actionable reads, but the coach still chooses the adjustment.

3. Officiating decisions: FIFA’s Semi-Automated Offside Technology

Few decisions are as contested as offside. At the 2022 World Cup, FIFA deployed Semi-Automated Offside Technology (SAOT): 12 dedicated cameras track the ball and up to 29 data points on each player, 50 times per second, while a sensor inside the ball reports the exact kick point 500 times per second. The system flags potential offsides to video officials, who verify before the referee makes the final call — hence “semi-automated.” For the 2026 World Cup, FIFA is upgrading SAOT with 3D-scanned player avatars and tighter alert thresholds to make calls faster and harder to dispute. Source: FIFA, 2022 & 2026.

4. Broadcast, fan & tournament decisions: IBM at Wimbledon

Wimbledon and IBM have one of sport’s longest-running AI partnerships. IBM’s Power Index combines statistical analysis with natural-language processing of millions of expert opinions to quantify each player’s momentum, while Match Insights mines more than 100,000 data points per match to generate predictions and analysis. IBM pioneered AI-curated highlight reels in 2017 — using crowd noise, player gestures, and game context to detect the most exciting moments — and has since layered on generative-AI features like “Catch Me Up” player summaries (2024) and AI-generated spoken commentary, all running on watsonx. These shape editorial and broadcast decisions about what to surface to a global audience. Source: IBM case studies; ESPN, 2023–2024.

The market — and a caveat on the numbers

The AI-in-sports market is growing fast, but analyst estimates diverge sharply: 2025 sizing ranges from roughly $1.2 billion to more than $10 billion depending on what each firm counts. One widely cited projection (Precedence Research) puts the market near $60.8 billion by 2034, a ~21% compound annual growth rate, with predictive analytics and generative AI as the largest segments and North America the largest region. Treat any single figure as directional; the trajectory — up and to the right — is what the sources agree on.

The through-line: decision support, not decision replacement

Across drafting, coaching, officiating, and broadcasting, the winning pattern is the same: AI handles the volume and speed of data; humans keep authority over the call. The Raptors’ scouts still pick the player. The referee still signals offside. The coach still makes the substitution. That human-in-the-loop design is not a limitation to be engineered away — in high-stakes, high-scrutiny environments, it is the feature.

Methodology & sources

Frequently asked questions

How is AI used for decision-making in sports?

AI ingests large volumes of data — tracking, video, statistics, medical and sentiment data — and turns it into recommendations for roster, coaching, officiating, and broadcast decisions. In nearly all serious systems, a human makes the final call.

What did IBM Watson do for the Toronto Raptors?

From 2016, the Raptors used IBM’s “Sports Insights Central” to evaluate draft prospects — combining stats, video, medical and social data against team needs — automating a process previously run manually with whiteboards.

Does AI make the offside calls at the World Cup?

Not on its own. FIFA’s Semi-Automated Offside Technology measures player and ball positions and alerts video officials, but the on-field referee makes the final decision — which is why it is called “semi-automated.”