The AI bubble will burst, and it's our responsibility to ensure a responsible replacement. A similar scenario unfolded in December 1999, when tech investors were convinced that a website and a Super Bowl ad were the keys to instant wealth. This led to the dot-com boom, which, within months, collapsed, resulting in a $1.7 trillion market value loss and a $5 trillion economic impact. However, the aftermath brought something remarkable: the emergence of Web 2.0, open-source software, and platforms like Firefox and Wikipedia. This taught us that when bubbles burst, we can create something better by changing our approach. Today, history is repeating itself with the AI boom, which is eerily similar to the dot-com era. Nearly 80% of stock gains in 2025 are concentrated in just seven companies, all vying for control of the AI stack's hardware, software, data, energy, and infrastructure. This level of concentration is concerning. Like the dot-com days, valuations are skyrocketing without clear paths to profitability. Companies are promoting the idea that AI will replace human workers, despite the fact that 95% of AI experiments inside firms fail to reach production. Instead of creating public-interest tools, the industry is generating synthetic media, misinformation, and deepfakes, as described by Cory Doctorow. The issue isn't AI itself but the economic logic behind it, which treats technology as an extractive industry, hoarding data, consolidating power, and externalizing harm. The AI arms race is driven by domination and profit, not innovation. Fortunately, an alternative economic model exists. Open-source developers and mission-driven companies worldwide are building shared infrastructure for trustworthy AI, which is transparent, auditable, and locally adaptable. They prove that innovation doesn't require monopolistic control of data. Companies like Hugging Face, Flower AI, and Oumi are leading the way, offering values-driven and competitive tools. These aren't speculative bets but seeds for a sustainable, pluralistic tech ecosystem. This approach aligns with a double-bottom-line economic model for tech, valuing both mission and money. History teaches us that the current frenzy will end in a crash, similar to the dot-com boom. However, this crash can mark the beginning of something new. The Linux stack, an open-source foundation, rose from the ashes of the dot-com crash and beat Windows. Open-source building blocks have created an astonishing $8.8 trillion in value over two decades, with potential tens of billions in value for startups and businesses switching to open-source AI models. When the AI bubble bursts, we'll choose between rebuilding monopolistic models or designing a pro-human, values-driven economy. This means open models, transparent governance, and equitable participation in AI-generated value. It also involves focusing on privacy, security, agency, and joy in technology. The promise of AI lies in making our lives easier, richer, and more creative without sacrificing choice or dignity. This vision is already becoming a reality. As we experiment with privacy-protecting, open-source models for browser and email assistants, they are continually improving. Imagine a future where individuals and communities host small, local AI models, energy-efficient, privacy-preserving, and tailored to their needs. Developers collaborate, not compete, and innovation is measured by public good, not market share. This isn't a utopian fantasy; it's achievable. By starting now and building open, transparent, and values-based AI, we can ensure that the next era of technology expands human freedom. The dot-com crash led to the modern web, and the next correction could bring a better one if we dare to rethink the economics of innovation. The choice is ours: let a few companies control the future, or own what we build together.
