In 2018, a group of engineers from Corning Incorporated visited one of Meta‘s data centres. What they saw prompted a rethink.
Corning had been making optical fibre for decades. In fact, the 175-year-old glassmaker had spent years investing in a business whose economics often looked less than compelling. But inside increasingly dense data centres, something was changing. Copper connections were becoming a constraint. Fibre could move enormous amounts of data more efficiently, but the cables needed to become thinner, denser and easier to bend.
Corning started adapting its technology. Then generative AI arrived and demand for data-centre connectivity exploded. In January 2026, Meta announced a multi-year agreement worth up to US$6 billion for Corning to supply optical fibre, cable and connectivity products for its AI infrastructure. Corning’s Optical Communications sales rose 36 per cent year-on-year in the first quarter of 2026.
There is an obvious lesson here about patience and long-term investment. But I think there is another one that matters just as much.
Corning was ready for a future that did not yet exist because it kept questioning what fibre might be for.
Every organisation operates on assumptions. Customers want this. People will pay for that. This process takes six weeks. Our industry works this way. That technology sits outside our sector. This is what quality looks like.
Most of those assumptions are sensible when they are formed. The danger is that they can remain embedded in a business long after the conditions that made them true have changed.
And right now, a surprising number of those conditions are moving at once.
The customer may not want what the category has always sold
I was reminded of this recently while speaking with leaders at Vinarchy, the wine company formed from the merger of Accolade Wines and Pernod Ricard Winemakers. One idea in particular caught my attention: a shift from thinking “grape to glass” towards thinking “glass to grape”.
Traditional winemaking naturally starts with the grape: the vineyard, varietal, vintage and craft. But what happens when you reverse the sequence and begin with the person holding the glass? What do they want to drink? When? In what format? What flavours feel relevant? What are they comfortable spending?
That change in vantage point can produce some decidedly untraditional answers. In the UK, Vinarchy’s Echo Falls launched a bright blue raspberry fruit wine in 2026. It was deliberately colourful, sweet and highly visible, aimed at drinking occasions where traditional wine can struggle to stand out. Vinarchy reported rapid sell-through at some wholesalers and more than two million organic social impressions soon after launch.
Purists may recoil. That is almost the point. The experiment is questioning whether conventions that matter deeply to an industry matter equally to the next customer entering it.
Procter & Gamble is running a similar experiment in a very different aisle. In February, Tide launched evo nationally in the US: a waterless fibre-detergent tile. Laundry detergent has already moved from powder to liquid to pods. Now Tide is asking whether the plastic bottle itself is an assumption worth reopening.
None of this means blue wine or detergent tiles will replace what came before. P&G itself expects multiple formats to coexist. The more useful point is that even very mature categories can contain surprisingly immature assumptions about how customers will want to buy and use them next.
When technology gets cheaper, process becomes harder to sell
There is another assumption being tested across professional services: that time and process are reasonable proxies for value.
For decades, complexity created a moat. Information was scarce. Expertise was difficult to access. Producing a polished piece of work required time and teams. A six-week process could signal rigour because, quite often, six weeks of human effort genuinely sat behind it.
AI is changing that equation. Research, analysis, drafting, iteration and production can increasingly happen in hours or minutes. That does not make expertise irrelevant. It changes where the expertise needs to show up.
Strategist Adam Ferrier put this provocatively in one of the research pieces I collected: if your margin depends on how long something takes or how many people touch it, AI puts pressure on the premise of the model. His argument is that judgement, taste, curation, lateral thinking and outcomes become more valuable as execution gets easier.
The legal sector offers an early glimpse of this. Industry research I reviewed shows technology portals and streamlined delivery increasingly being treated as baseline expectations rather than compelling differentiators. At the same time, clients are placing more emphasis on pricing predictability, senior access and cross-disciplinary expertise.
This pattern will spread well beyond law. When the machinery behind an outcome becomes dramatically faster, customers eventually ask why they are still paying for the machinery.
That is an uncomfortable question for any business whose value proposition has quietly become entangled with friction.
Look one step beyond the obvious disruption
Perhaps the most interesting opportunities, though, are the ones that appear outside the industry where a trend begins.
Consider GLP-1 weight-loss drugs. Their obvious implications sit in healthcare, pharmaceuticals and food. Airlines would hardly make the first page of most trend maps.
Yet analysts at Jefferies modelled what widespread weight reduction could mean for aviation. Because passenger weight contributes to total aircraft weight, even modest reductions could lower fuel consumption. Their scenario suggested a 10 per cent reduction in average passenger weight could reduce total aircraft weight by around 2 per cent and fuel costs by up to 1.5 per cent. Those are estimates, not observed savings, but they illustrate something important.
A pharmaceutical innovation can alter airline economics.
AI creates the same effect. One of its unexpected winners is a glassmaker. The demand is not just for better models and chips, but for the physical infrastructure connecting enormous clusters of computing power. Corning is now expanding manufacturing capacity under its Meta agreement, and in 2026 announced other major optical-fibre supply agreements as AI infrastructure demand grew.
This is why I increasingly think “How will this trend affect our industry?” is too narrow a question.
A more useful question is: What does this change make newly possible, newly valuable, newly cheap, newly expensive or newly unnecessary?
That forces us to look beyond the first-order effect. And the second-order effect is often where the less crowded opportunity sits.
Getting closer to reality matters more when prediction gets harder
There is a practical problem with all of this. If assumptions are expiring faster, leaders need a way to notice.
Burger King tried an unusually literal version of this in February 2026. Rather than relying only on surveys and dashboards, the company published the phone number of its US and Canada president, Tom Curtis, and invited customers to call or text him directly. Every message was to be reviewed and answered, and Curtis committed to taking as many calls as he could personally.
Thousands of customers responded. By July, Burger King said that feedback was already shaping operational changes, including a redesigned restaurant-manager role intended to make it easier to fix customer problems in the moment.
The phone number is a stunt, of course. But the principle underneath it is serious: reduce the distance between the people experiencing change and the people making decisions about it.
Starbucks has been doing something related from another direction. After years of pushing convenience, digital ordering and throughput, its “Back to Starbucks” strategy has reinvested in baristas, service standards and the coffeehouse experience. In its second fiscal quarter of 2026, global comparable-store sales rose 6.2 per cent and US comparable sales rose 7.1 per cent. Those results cannot be attributed to any one initiative, but they are a useful reminder that the right response to technological change is not always more technology.
Sometimes the assumption worth questioning is that customers want less human interaction.
Not every old assumption is wrong
There is a trap here. Once leaders become excited about disruption, it is easy to treat every legacy practice as a liability and every new idea as progress.
Corning is a useful corrective.
Part of the reason it can capitalise on today’s fibre boom is that it kept capabilities that, at various points, looked inefficient. During the post-pandemic downturn, CEO Wendell Weeks has acknowledged the company was carrying thousands more employees than its revenue appeared to support. In a conventional efficiency drive, cutting deeper would have been easy to justify. When demand returned, that retained capability suddenly looked very different.
The goal, then, is not to abandon old assumptions. It is to stop treating them as facts.
Run an assumption audit
Most strategic planning begins by looking outward: competitors, technology, regulation, demographics, economics. All of that matters. But there is value in occasionally turning the lens inward and asking what your strategy assumes will remain true.
I think of it as an assumption audit. Start with four areas.
- Customer: What are we assuming people still want?
- Value: What are we assuming customers will continue paying for?
- Constraint: What are we assuming will remain difficult, expensive or scarce?
- Boundary: What are we assuming sits outside our industry and therefore does not concern us?
The point is not to manufacture doubt about everything. It is to make invisible assumptions visible again, because once an assumption becomes visible, it can be tested.
This matters even more in an era of AI. Producing answers is getting easier. Spotting that the question itself has changed may become one of the more valuable capabilities a leader can develop.
Every business has an expiry date on its assumptions. The organisations that stay relevant will not be the ones that chase every new trend. They will be the ones curious enough to notice when an old truth has stopped being true.
Michael McQueen is a globally recognised trend forecaster, change strategist and keynote speaker.
A bestselling author of 10 books, Michael’s latest release was named by Malcolm Gladwell and Adam Grant as one of the top five must-read new leadership books. He is a sought-after media commentator, with his insights regularly featured in Forbes, The Guardian, and CNN.
To find out more about Michael and his work, click here.
NOTES
Meta. “Meta Announces Up to $6 Billion Agreement With Corning to Support US Manufacturing.” January 27, 2026. https://about.fb.com/news/2026/01/meta-6-billion-agreement-corning-support-us-manufacturing/.
Corning Incorporated. “Corning and Meta Announce Multiyear, up to $6 Billion Agreement to Accelerate US Data Center Buildout.” January 27, 2026. https://investor.corning.com/news-and-events/news/news-details/2026/Corning-and-Meta-Announce-Multiyear-up-to-6-Billion-Agreement-to-Accelerate-US-Data-Center-Buildout/.
Corning Incorporated. “Corning Announces Strong First-Quarter 2026 Financial Results.” 2026. https://investor.corning.com/financials/quarterly-results/default.aspx.
Vinarchy. “Wholesale Gets First Pour as Echo Falls Taps Blue Raspberry Demand.” March 18, 2026. https://vinarchy.com/blogs/news-updates/wholesale-gets-first-pour-as-echo-falls-taps-blue-raspberry-demand.
Procter & Gamble. “Tide Unveils World-First Laundry Breakthrough: Introducing Tide evo, the Fiber-Detergent Tile.” February 17, 2026. https://pgn2020news.q4web.com/news-releases/news-details/2026/Tide-Unveils-World-First-Laundry-Breakthrough-Introducing-Tide-evo-the-Fiber-Detergent-Tile/default.aspx.
Ferrier, Adam. “If Your Business Model Relies on Friction … You May Be Fucked.” 2026. Research note supplied by author.
Li, Yun. “Airlines to Save Big Money on Fuel as New Weight Loss Pills Gain Popularity, Wall Street Says.” CNBC, January 14, 2026. https://www.cnbc.com/2026/01/14/airlines-to-save-on-fuel-as-weight-loss-pills-grow-popular-wall-street-says.html.
Burger King. “Burger King Puts Its President on the Line, Giving Guests Direct Access to Share Feedback.” February 17, 2026. https://news.bk.com/blog-posts/burger-king-r-puts-its-president-on-the-line-giving-guests-direct-access-to-share-feedback.
Burger King. “Burger King Continues Turning Guest Feedback into Action with the ‘Your Way Champion’ and Whopper Guarantee.” July 20, 2026. https://news.bk.com/blog-posts/burger-king-continues-turning-guest-feedback-into-action-with-the-your-way-champion-and-whopper-guarantee.
Starbucks Corporation. “Starbucks Reports Q2 Fiscal Year 2026 Results.” April 28, 2026. https://investor.starbucks.com/news/financial-releases/news-details/2026/Starbucks-Reports-Q2-Fiscal-Year-2026-Results/default.aspx.
McKinsey & Company, McKinsey Center for Future Mobility. “How Consumers Are Reshaping the Future of Mobility.” July 14, 2026.
