IIt was December 1999. Tech investors were convinced that a website and a Super Bowl ad were all it took to get rich quick. Spending was considered an increase; Marketing is mistaken for a business model. In a few months, the dot-com boom will be over: $1.7tn market value disappearsAnd the wider economy suffered a shock of $5tn.
Yet something remarkable emerged from the debris. The post-crash Internet was defined not by speculation, but by creation: the rise of Web 2.0 and open-source software – and the birth of platforms like Firefox and Wikipedia. The lesson is simple: When the bubble bursts, what comes next can be better, if we build it differently.
Today, history is repeating itself – this time with AI.
The AI boom sounds extremely familiar. Nearly 80% of stock gains in 2025 are concentrated in just seven companies – Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla, all competing for control of the full AI stack that will underpin our shared future – Hardware, Software, Data, Energy and Infrastructure. It’s not just about market share, it’s about who decides how billions of people learn, create and see the world.
That level of concentration should concern us all.
And like the dot-com days, valuations are rising without a clear path to profitability. However, companies are selling the fantasy that AI will replace human workers 95% AI experiment Internal companies fail to reach production. And instead of creating public-interest devices that expand human potential, most industries are producing what Cory Doctorow calls “productive residue“ – A flood of synthetic media, misinformation and deepfakes.
The problem isn’t the AI itself; This is the current economic logic behind this.
It is not inevitable. This is the result of an economic model that treats technology as an extractive industry – accumulating data, consolidating power, and excluding harm. The AI arms race is driven not by innovation but by dominance, prioritizing profit over humanity.
A different economic model already exists
The good news is that an alternative model already exists. Around the world, open-source developers and mission-driven companies are building shared infrastructure for trustworthy AI – transparent, auditable, and locally adaptable. They are proving that innovation does not need to rely on monopoly control of data.
This is evident in leading companies, founders are creating tools that are both value-driven and competitive. Companies like Hugging Face, which runs the world’s most widely used open-source machine-learning model and dataset hub; Flower AI, which enables decentralized, federated learning to challenge the dominance of centralized large models; and Oumi, which offers a completely open-source platform for building and deploying custom AI models on local infrastructure rather than closed clouds. and many more.
These are not speculative bets; They are the seeds for a more sustainable, pluralistic tech ecosystem. It’s part of what we see as a double-bottom-line economic model for tech — an approach that values mission. And Wealth.
slope is not destiny
If history is any guide, the current mania will end the same way the dot-com boom did: with a crash. But this is not the end of the story – it is the beginning of a new story.
In the last bubble, linux stackThe open-source building blocks that now form the basis of almost everything on the Internet rose from the ashes to defeat Windows. Such open-source building blocks have been created astonishing $8.8tn in value over the past two decades, with New research estimates Startups and other businesses would be worth billions of dollars if they switched from closed AI platforms to open-source models.
How much value can we create today? It is very big.
When the AI bubble bursts, we will be faced with a choice. We can rebuild the same monopolistic model, or we can use the moment to design an economy that is pro-human and driven by values. This means open models, transparent governance and equal participation in the value created by AI.
It also means focusing on what people really want from technology: privacy, security, agency, and fun. The promise of AI is not its infinite scale – it is its ability to make our lives easier, richer and more creative without sacrificing choice or dignity.
This is already happening. As we experiment with privacy-protecting, open-source models for things like browsers and email assistants, we see them getting better and better.
Imagine a future in which individuals and communities can host small, local AI models – energy-efficient, privacy-preserving, and tailored to their needs. Where developers create tools not competitively, but collaboratively. Where innovation is measured not by market share, but by public good.
This is not some imaginary fantasy. If we start now – building AI that is open, transparent and rooted in shared values – we can ensure that the next era of technology will expand human freedom rather than constrain it. The dot-com crash gave us the modern web. The next reform may leave us better off – if we have the courage to rethink the economics of innovation.
Ultimately the choice is ours. We can let a handful of companies own the future. Or we can own what we create – together.
