Is The Dot-Com Bubble Back To Haunt Us? What The Past Can Teach Us About The AI Boom

In the late 1990s, the internet promised to change everything. Investors poured billions into any company with a “.com” in its name, believing the web would reshape the world – which, of course, it did.

But, not before a spectacular collapse. The dot-com bubble became one of the most infamous financial manias in modern history, fuelled by hype, easy money and the belief that traditional business logic no longer applied. Everything went out the window and the world would never be the same again.

Fast forward to today, and the same whispers are resurfacing, but this time, the chatter is about artificial intelligence.

Giants like Microsoft, Alphabet, Meta and OpenAI are leading a global race to dominate AI, pouring unimaginable sums into infrastructure, chips and startups. But, beneath the glossy announcements and market euphoria, a question lingers, especially among some of the tech industry veterans – haven’t we seen this story before?

The parallels between the dot-com boom and the current AI frenzy are hard to ignore. Both are driven by revolutionary technology and unbridled optimism, and both risk being crushed under the weight of their own expectations.

When the Bubble Burst

The crash was swift and brutal.

Within two years, the NASDAQ had lost nearly 80% of its value. Companies that had once seemed unstoppable went bankrupt almost overnight. Fortunes were wiped out, and many tech workers found themselves unemployed. Yet, amid the wreckage, a few survivors, like Amazon and Google, would go on to shape the next two decades of technology. Not only did they survive, but they thrived in the midst of one of the biggest industry crashes the world has ever seen.

The fallout also forced investors, regulators and entrepreneurs to face hard truths. The internet wasn’t a magic ticket to success – it was a tool that still required sound business fundamentals, realistic timelines and clear paths to profitability.

While the technology itself continued to evolve, the speculative mania around it became a cautionary tale, and now it’s one that’s being revisited as the world watches the AI boom unfold.

So, are we about to fall into the same trap we did before? Should we know better than we did then? Or, is it totally different? Have we learnt the important lessons that’ll allow us to survive the AI transistion?

Why People Are Comparing It to the AI Boom

Much like the late ’90s, today’s AI industry is being driven by breathtaking optimism.

Every week brings new headlines about breakthroughs in generative AI, record-breaking investment rounds and predictions that the technology will transform every aspect of work and life. Companies are racing to adopt AI tools, even when they don’t fully understand how they’ll integrate or monetise them. It’s all AI – everywhere and all the time.

Indeed, tech giants are spending at a scale that would have seemed unthinkable even a few years ago. Microsoft has reportedly invested billions in OpenAI, while Alphabet and Meta are funnelling massive sums into their own AI research.

Chipmaker Nvidia has had its valuation soaring to knew hights over the last few years, with its most notable recent achievment was the fact that it became the first company to achieve $5 trillion – albeit briefly.

But for all the excitement, the profits haven’t yet caught up. Many AI startups are running at a loss, and even large players are struggling to show how their enormous AI investments will pay off in the short term.

It’s this imbalance – the misalignment between hype and revenue, investment and tangible value – that has many economists and analysts warning of an AI bubble.

The pattern feels eerily familiar – a groundbreaking technology, an avalanche of investment and sky-high expectations that may not align with the current reality. Is the value real or is it being created artificially by the hype that’s being created about AI within the market itself?

What People Anticipated Back Then And What They Expect Now

During the dot-com era, people imagined a world where online businesses would dominate every aspect of daily life. While that vision was ultimately correct, the timing was wildly off. Infrastructure, consumer trust, and digital adoption weren’t ready to support such rapid transformation. Those things took a long time to come, and it was only the industry giants that were able to survive until it did.

Today’s AI hype carries similar expectations. Proponents envision a world where AI handles everything – from automating workplaces and creating art to making life-or-death decisions in healthcare and transport. There’s little doubt that AI will reshape society – in many ways, it already has – but just like the early internet, its potential is being oversold in the short term.

The technology is still expensive, power-hungry and often unreliable. The gap between promise and performance could easily widen before it narrows, and for many companies, especially the small ones, this could be the difference between life and death.

Lessons from the Past

If history offers any guidance, it’s that bubbles aren’t entirely bad.

The dot-com crash, painful as it was, also paved the way for genuine innovation. It cleared the market of unsustainable ventures, leaving behind the companies that could adapt and endure. The infrastructure built during that period – things from fibre-optic networks to broadband connections – laid the foundation for today’s digital economy.

And the same could happen with AI.

Even if some current projects fail or valuations fall, the core technology is here to stay. The challenge for investors and companies is to temper ambition with realism. Growth takes time, adoption takes patience and hype, no matter how exciting, isn’t a substitute for a solid business model.

Another key lesson lies in managing expectations. During the dot-com bubble, optimism blinded many to the fact that technology adoption happens in waves, not overnight. Those who learned to ride those waves, like Amazon, which weathered years of losses before becoming profitable, eventually emerged stronger. The AI industry may face a similar reckoning. When the initial euphoria fades, the survivors will likely be those with clear, scalable applications that solve real problems. And, of course, money. Those who survive will have deep pockets.

The Ghosts of Bubbles Past

So, is the dot-com bubble truly back to haunt us? Perhaps not in the same form, but its spirit certainly lingers. The AI boom carries echoes of that earlier era – that is, the race to dominate a new digital frontier, the torrent of capital and the collective belief that we’re standing on the edge of a technological revolution.

But this time, the industry has a chance to do things differently.

The lessons of the early 2000s – things like caution, sustainability and the very real danger of unrestrained hype, are still fresh in the minds of many in Silicon Valley. The companies that balance innovation with discipline may not only avoid a crash but build the foundations for a genuine new era of progress.

The Forward Requires Innovation Without Illusion

Every technological leap brings both opportunity and risk. The internet survived its bubble and went on to transform the world in ways no one could have imagined in 1999.

Artificial intelligence will likely do the same, but only after a period of correction and consolidation.

The dot-com era taught us that hype can ignite revolutions, but only discipline sustains them. If the AI industry can remember that, then perhaps history won’t repeat itself – it will ring, softly, as a reminder of how far we’ve come and how much further we can go when we balance excitement with reality.