Artificial intelligence seems to be all around us. It’s a constant in the news, it’s a focus for founders and investors are piling in fast.
And whilst it seems like the AI revolution is speeding up in front of our eyes, many have taken a step back and wondered, ‘Is the bubble going to pop?’.
With flashbacks to the dot com bubble, which saw online companies’ valuations soar before collapsing in 2000, many are looking at the latest rise in AI with caution.
After all, with NVIDIA stocks up 46% in the past year alone, it’s easy to see why investors are feeling that things at the moment seem a little too good to be true.
What Does A Bubble Burst Mean When It Comes To Stocks?
When it comes to stocks, a “bubble burst” is used to talk about the sudden drop of a market that was growing quickly.
In a normal bubble burst, investors would get excited about a new sector, pile in on stocks and drive them upwards. After some time, investors then realise that the stock is massively overpriced, triggering a mass selling event. Prices then plummet, wiping out billions for those still invested.
In short, a bubble burst is like a correction, but much much quicker!
So, with stocks climbing and hype around AI growing, the real question is: Is the AI bubble about to burst? We asked the experts…
Our Experts
- Leyland Clowsley, Managing Director at Apex Manufacturing
- James Fisher, Chief Strategy Officer at Qlik
- Marc Fernandez, Chief Strategy Officer at Neurologyca
- Dan Lawyer, Chief Product Officer at Lucid Software
- Julia Prostak-Bazan, IT Learning & Development Specialist at eSky and Thomas Cook
- George Sweeney, DipFA, Investment Expert at Finder
- Alexandre de Vigan, Founder & CEO at Nfinite
- Dylan Dewdney, Co-Founder and CEO at Kuvi.ai
- Marco Luciano, Managing Director at BBN AI Studio
- Tobias Robinson, CEO at Broker Listings
- Deepak Shukla, CEO at Pearl Lemon AI
- Professor Filip Bialy, Professor of Computer Science and AI Ethics at Open Institute of Technology (OPIT)
- Toby Coulthard, Chief Product & Growth Officer at Jacquard
- Ray Eitel-Porter, AI Governance Expert
- Jonathan Horn, CEO and Co-Founder at Treefera
For any questions, comments or features, please contact us directly.
Leyland Clowsley, Managing Director at Apex Manufacturing
“I believe the AI bubble is inflated and it won’t burst overnight, but it’s already showing signs of strain.
“AI software firms are pulling in extraordinary sums. In the first quarter of 2025 alone, AI startups attracted over 73 billion dollars, representing nearly 60 percent of all global venture capital investment. Yet many of these firms have little more than a proof of concept and a slide deck. The reality is that most do not survive beyond a few years. Studies suggest that more than 80 percent of AI startups will fail within three years, which is in line with general startup failure rates, but here the stakes and valuations are far higher.
“At the same time, hardware businesses like mine face longer lead times, high capital costs, strict regulation, and greater scaling challenges. These companies rarely receive the same hype or support. Too often, AI is attached as a buzzword to products with limited genuine intelligence simply to secure funding.
“The adjustment will come. What remains will be the companies providing tangible value, rather than those that only promised it.”
James Fisher, Chief Strategy Officer at Qlik
“No, the AI hype cycle isn’t bursting, but how businesses look at AI is changing. Like every major technology, adoption comes in waves. What we are seeing now is the initial phase of experimentation giving way to business leaders asking the hard questions about return on investment from AI projects that are underway.
“Our own research has shown how this challenge plays out. Only 11% of UK businesses say their AI initiatives are delivering measurable gains, and nearly a quarter (23%) admit most projects remain stuck in pilot mode.
“This isn’t because AI is all hype, it’s because many businesses don’t have the data ready to power these projects to their full potential. As we all know, the quality of the data being used for AI projects determines whether it sinks or swims.
“Businesses are now realising that without a strong data foundation, including trusted, governed and contextualised data, even the most advanced models won’t deliver at scale. Just as important is the analytics foundation that gets the right data and signals to the right people at the right time. That means governed access, in-workflow delivery, clear KPIs, and the ability to trigger actions, alerts and automations where decisions are actually made.
“So rather than write AI off as hype that doesn’t bring business benefit, we need to take a step back and get our data and analytics foundations in order, set clear use cases, and scale AI adoption as projects show their value.”
Marc Fernandez, Chief Strategy Officer at Neurologyca
“The idea that the AI bubble is about to pop doesn’t match what’s happening in enterprises. Adoption is still moving forward, but the focus has changed. Many of the early projects were launched quickly to capture attention and prove that a company was “doing something with AI.” Those efforts often came with inflated timelines and ROI expectations. The current focus is on use cases where AI clearly improves workflows and produces measurable outcomes.
“A big lesson from that first wave is that most AI systems can process data but fail to capture the human side like context, emotion and intent. When it is clear from the task or the prompt that human context adds value, the difference in results is significant. We see that across wellness, where AI can distinguish between stress and focus; soft skills development, where it can track confidence or communication improvements; AI agents and robotics, where adapting to human cues is essential; education, where personalization goes beyond right and wrong answers; security, where intent recognition can improve threat detection; and personalized recommendations, where AI can adjust not only to preferences but also to someone’s state of mind.
“This shift makes adoption look slower on the surface, but it shows companies are building more sustainable strategies for how AI fits into their business. The conversation has moved from experiments to impact and from hype to results. The companies that succeed will be the ones that use AI not just to process information but to support how people actually work and make decisions.”
For any questions, comments or features, please contact us directly.
Dan Lawyer, Chief Product Officer at Lucid Software
“Businesses are eager to adopt AI solutions, but often lack the necessary infrastructure to do so effectively. In fact, 59% of knowledge workers at businesses under 1,000 employees agree with that sentiment and 25% also state a lack of transparency around best practices is a key blocker to its successful adoption. While employees report a disconnect, 59% of executives actually believe their AI strategy is well aligned. This gap could stall efforts to prepare businesses to effectively implement AI solutions like agentic AI in 2026.
“Instead of trying to layer AI onto legacy processes, which adds more complexity, organisations need to evaluate current workflows, processes, and systems to assess how well ready they are for AI enhancement. From there, they can build a foundation that allows AI to integrate seamlessly and responsibly into current workflows. This starts by documenting and optimising processes, data flows and collaboration practices. Having this clarity on how people, systems, and processes connect will enable businesses to adopt AI agents that genuinely improve the workplace.”
Julia Prostak-Bazan, IT Learning & Development Specialist at eSky and Thomas Cook
“Talk of an “AI bubble” misunderstands what is happening. A bubble implies speculation without substance, but AI has already embedded itself into daily life and business operations. It is not the future – it is the present.
“We see this most clearly in how ordinary people now use AI tools. Travellers can ask ChatGPT or Gemini to plan realistic budgets, highlight hidden costs such as baggage charges or currency conversion fees, and weigh up trade-offs between short city breaks versus long-haul adventures. These are practical, tangible uses that make decision-making clearer and faster – not hype.
“AI is also transforming work across industries, including travel. Marketing teams can generate campaign assets in hours instead of days, while customer service departments can analyse every interaction rather than relying on random samples. These are structural shifts in how companies operate and how people experience services.
“As with any powerful technology, there are challenges – from accuracy to responsible use – but dismissing AI as a passing trend ignores the scale of change already underway. The reality is that AI has moved past experimentation and into everyday infrastructure. It is not a bubble: it is the new normal.”
George Sweeney DipFA, Investment Expert at Finder
“The potential “AI bubble” we’re in may not pop, but I do see a scenario where we see what’s more like an almighty unravelling. Lately we’ve been seeing the stock prices of the likes of Oracle, Microsoft, and Nvidia all moving up due to deals and contracts linked to OpenAI.
“What seems to be being overlooked by lots of investors and analysts is that currently, OpenAI generates a relatively small level of revenue while spending piles of cash, and it’s making lots of future promises that will be difficult to fulfill at scale.
“The finances of some of the world’s largest tech stocks are becoming increasingly intertwined because of deals around compute power, semiconductors, chips, and other tech and infrastructure linked to the AI ecosystem.
“If one link breaks in this back and forth chain, it could lead to a knock-on effect that leads to a serious unwinding of valuations for tech stocks that are currently heavily based on future AI revenue and profit projections that may or may not materialise.
“It’s definitely starting to give echoes of the dotcom bubble where the tech will undoubtedly revolutionise the world as we know it; but there’s likely to be a lot of financial collateral damage on this road to new horizons.”
For any questions, comments or features, please contact us directly.
Alexandre de Vigan, Founder & CEO at Nfinite
“There’s a lot of talk about an “AI bubble,” but what we’re living through looks less like a bubble and more like a shift between eras. Language models have brought astonishing progress, yet they are already running up against the limits of text-only training. The next leap, what Google, Meta, Nvidia and others are calling “world models”, requires machines to perceive and act in the physical world. That shift won’t collapse the market; it will deepen it.
“The real risk isn’t a pop, it’s a bottleneck. AI systems can’t learn to navigate homes, hospitals, or store aisles without the right infrastructure: structured, simulation-ready 3D data about products, environments, and human contexts. Without it, models remain brilliant in the abstract but unreliable in practice.
“At Nfinite, we call this missing layer Physical AI. Digital twins and high-fidelity 3D assets are already delivering ROI in commerce: boosting conversion rates, reducing returns, and enabling automation. The same infrastructure underpins robotics, AR/VR, and autonomous systems.
“So no, the AI bubble isn’t bursting. But for AI to sustain its trajectory, it needs to root itself in the physical world. The winners will be those building the data foundations that make intelligence useful beyond the screen.”
Dylan Dewdney, Co-Founder and CEO at Kuvi.ai
“I don’t see the AI ‘bubble’ bursting, at least not in the way many commentators frame it. What we’re experiencing is the classic pattern of a breakthrough technology: initial hype, some cooling off, and then steady, compounding integration into every layer of our lives. The AI market has already corrected somewhat from its fever pitch a year ago, but that’s not collapse, it’s just normalization.
“The reality is that masses of people around the world are only just waking up to the practical utility of agentic frameworks, systems where AI doesn’t just respond, but acts on intent and orchestrates outcomes. Over the next five years, these tools will not only improve rapidly in quality, but also embed themselves across stacks, from enterprise workflows to consumer apps to personal assistants that shape how we interface with the physical world.
“We should expect AI adoption to mirror the internet’s trajectory. Yes, there was a dotcom bubble. But within a decade came e-commerce, social platforms, and the digital economy we now take for granted. Corrections are inevitable, but the long-term productivity lift from AI is only just beginning.”
Marco Luciano, MD at BBN AI Studio
“Yes, but we’re all watching the wrong one. There’s the obvious bubble of inflated tech valuations, which may well burst.
Then there’s the invisible bubble: the massive underestimation of AI’s impact on B2B. Most companies are drowning in AI noise, not solutions. They’re collecting tools, creating expensive chaos, and mistaking it for transformation.
“The real value isn’t in consumer-grade toys; it’s in rebuilding entire business operations. This is the tectonic shift happening below the surface.
“So, let the hype bubble pop. The real danger isn’t being caught when it bursts, but being left outside the invisible one as it becomes the new foundation of business.”
For any questions, comments or features, please contact us directly.
Tobias Robinson, CEO at Broker Listings
“I do not believe AI is a full-on bubble about to go pop. However, parts of the market are definitely overheating.
“We’ve seen firms pour over $500 billion into AI over the last couple of years, yet the revenue generated from this investment sits at less than $50 billion – not even 10%. That worries me.
“A lot of AI stocks, with Nvidia as a case in point, are also trading at valuations way above their current earnings or sales, banking on massive future profits that frankly may not come to fruition.
“Many of the newer players in the AI space are also built on venture capital or debt, which makes them vulnerable if funding conditions start to tighten.
“Additionally, investor sentiment has largely been one-sided – extremely bullish. But many investors, retail especially, are following hype rather than solid fundamentals, something that reminds me of the dot-com boom.
“Yet it’s not all doom, gloom, and bubble bursting. Unlike the dot-com era, AI is already powering products across finance and logistics, amongst other industries. And crucially, the big technology firms, such as Microsoft, are sitting on massive amounts of cash, so they are somewhat insulated.
“So, in my view, AI isn’t a single bubble about to go pop – it’s a legitimate technology boom, just with some dangerously overhyped pockets that I expect will see a sharp correction.”
Deepak Shukla, CEO at Pearl Lemon AI
“People keep asking if we’re heading for an AI bubble burst, and honestly, I get why. It’s been wild lately — the hype, the headlines, the sudden flood of “AI-powered” everything. Everyone wants a piece of it. But a total crash? I don’t see it. What’s happening feels more like a reset. The early noise is quieting down, and people are finally starting to separate the useful from the useless.
“Sure, a bunch of companies will vanish, especially the ones that jumped in without a real plan or product. But that’s not a disaster; it’s just how any young industry grows up. The real ones — the ones using AI to actually solve problems and save people time or money — they’re the ones that’ll stick around.
“We’re seeing it firsthand. A year ago, clients mostly wanted to “try AI.” Now they’re asking how to make it actually work for their business. That shift says everything.
“So, no, the AI bubble isn’t popping. It’s just thinning out the fluff. What’s left is a steadier, smarter space built on results, not buzzwords.”
Professor Filip Bialy, Professor of Computer Science and AI Ethics at Open Institute of Technology (OPIT)
“It is important to differentiate between the end of AI hype and the possible burst of the AI bubble. The former does not necessarily mean the latter. AI hype – an overly optimistic view of the technological and economic potential of the current paradigm of AI – contributes to the growth of the bubble. However, the hype may end not with the burst of the bubble but rather with a more mature understanding of the technology.
“While concerns that existing infrastructure is unable to cope with the growth of AI are certainly justified, the promised returns from investment in AI over the next five to ten years seem to convince decision-makers across the world to remove most environmental restraints that would slow down the construction of new data centres that are essential part of that infrastructure.
Several important factors may not only sustain but even increase the current levels of investment in AI:
“First, the fierce competition that is taking place not only between technology companies. The number of chips purchased and data centres built has become the most visible marker of progress for venture capitalists, who may otherwise be ignorant about the technology. Paradoxically, this makes companies less willing to research more cost- and power-efficient models, as such work might reduce the need for infrastructural expansion and thus confuse investors.
“Second, AI has become part of geopolitics, with the United States and Chinese governments both supporting their domestic AI companies in the hope of using the technology as a means of achieving global domination, including in military contexts.
“Third, AI has become a focal point of the global economic system. It is precisely in this sense that it has become “the new oil”: it builds vast fortunes for technology companies and their chief executives, just as actual oil did during the first Gilded Age and for most of the twentieth century.
“Fourth, we should not underestimate the determination of contemporary “robber barons”, such as Elon Musk or Sam Altman, who are motivated not only by economic reasons. They appear to genuinely subscribe to the tenets of transhumanist philosophy, which perceives the creation of artificial general intelligence (AGI) as the ultimate solution to human problems and as a step towards the space expansion of the human race. It is important to stress that many AI experts – such as Yann LeCun – do not believe that large language models (LLMs) will lead to AGI, citing hard limitations such as insufficient new, high-quality data and the fact that LLMs do not possess an understanding of the physical world.
“At the same time, we are already witnessing a reality check regarding some of the most exaggerated promises of AI. A recent, widely publicised MIT report claims that the vast majority of generative AI pilot programmes fail to implement AI in delivering services. The report shows that while the technology itself may not be to blame, the way in which it is being deployed leads to a high rate of failure. One major reason is that companies invest resources in visible use cases, while less spectacular but potentially more transformative implementations remain underinvested.
“An AI enthusiast would argue that this is a necessary learning curve for business leaders and technologists alike, leading towards an AI landscape in which a proper understanding of AI’s potential produces significant improvements in business processes. In this context, the overblown role of generative AI may diminish, as it is often the now “old-fashioned” discriminative AI models that hold the key to real transformation.
“In sum, what we may be looking at over the next few years is a more mature understanding of AI, which hopefully will not require the AI bubble to burst. This does not mean, however, that we will not continue to be subjected to AI hype from technologists and their followers. Yet we may eventually become better at telling the difference between their claims and reality.”
For any questions, comments or features, please contact us directly.
Toby Coulthard, Chief Product & Growth Officer at Jacquard
“There is clear value in AI that investors are rightfully identifying, however as the 2025 MIT study showed, we are witnessing Roy Amara’s law manifest. That is to say that people tend to overestimate the value of a technology in the short term, and underestimate it in the long term. This mismatch in the short term expectations of the market and the value that AI can bring today has resulted in a good amount of froth in the market. A clear litmus test of value is looking beyond just ‘productivity and efficiency’ and seeing whether the technology solves defined problem sets – it’s these companies that will persevere when the bubble pops.”
Ray Eitel-Porter, AI Governance Expert
“Opinions differ sharply among industry leaders, investors and economists as to whether the AI bubble is about to burst. What matters more is the potential impact on the public and companies were this to happen. While a crash would impact investor returns, from venture capital firms to pension funds, and new funding for start-ups would slow to a trickle, unlike a purely financial crash, much substance would remain.
“Most of the huge investments in AI are for data infrastructure and power generation – these assets would remain after a crash. Importantly, the capabilities of frontier AI models themselves already far outstrip the ability of companies to turn them into tangible products and services. Implementation is held back by technical and change management challenges but also by a lack of trust. Business leaders and consumers recognise that generative AI often delivers unreliable results and is prone to other risks such as lack of explainability and security vulnerabilities. A pause in the breakneck speed of frontier AI development would allow time for companies to establish robust AI governance and build confidence in using AI. This suggests there would still be plenty of scope for AI-fuelled innovation leading to beneficial consumer outcomes.”
Jonathan Horn, CEO and Co-Founder at Treefera
“The idea that generative AI has ‘run its course’ is misguided. The real frontier lies in AI’s ability to abstract scientific and technical complexity into something decision-makers can actually use. That transformation of hard-to-interpret data into reliable, actionable insight is what will make AI essential infrastructure for business.”
“We’re only beginning to see how this applies across industries. In global supply chains, for example, data is fragmented, inconsistent, and often inaccessible. AI can synthesize that complexity into foresight — enabling sourcing decisions, risk mitigation, and regulatory compliance in the first mile. That’s not a bubble, that’s progress.”
“Agentic approaches are part of this evolution, but the bigger story is AI’s growing capacity to turn complexity into clarity at scale. That’s where its enduring value lies.”
For any questions, comments or features, please contact us directly.
