The New Rules of Tech: Henry Blodget on AI, Dot-Coms, and the Lessons of 25 Years

In the pantheon of tech industry prophets, few have as much credibility—or as much to answer for—as Henry Blodget. Once a symbol of the dot-com era’s excess, the former securities fraud defendant turned tech analyst now finds himself comparing today’s AI boom to the bubble he helped inflate in the early 2000s. The question is: Has anything changed? Or are we simply repeating history with new algorithms?

The Dot-Com Blueprint: Overvaluation, Overhiring, and Overconfidence

Blodget’s career began in the 1990s, when he was a star analyst at Merrill Lynch. He famously urged investors to “buy every [tech] stock you can” in 1998, then admitted in 2003 to having “lied repeatedly” to pump up share prices of companies like CMGI and Covad Communications. By the time the bubble burst in 2000, the NASDAQ had lost 78% of its value, and Blodget was fined $4 million and barred from the securities industry.

The playbook then was clear:
1. Speculative valuation: Companies with no revenue (e.g., Pets.com) were valued in the billions.
2. Rapid scaling: Startups hired aggressively, often doubling headcount in months.
3. Lack of accountability: Misconduct, whether financial or ethical, was overlooked in the rush to innovate.

The AI Era: New Technology, Old Patterns

Fast-forward to 2025. Blodget, now a media executive and tech commentator, sees eerie parallels. The pandemic-driven tech boom of 2020–2022 led to a hiring frenzy: 25,000 layoffs in early 2025 alone, per his analysis. Companies like Hewlett Packard Enterprise (HPE) laid off 2,500 employees due to overstocked AI servers, while startups like Skybox Security collapsed entirely.

But the similarities don’t stop at overhiring. Blodget’s own experiment with an “AI newsroom” exposed a darker thread: ethical complacency. When he “harassed” his fictional AI colleague Tess Ellery by complimenting her appearance, he framed it as a joke. Critics, however, saw a man who’d once inflated tech stocks with lies now trivializing workplace harassment—a pattern of prioritizing disruption over responsibility.

HPE’s stock fell 30% from its 2021 high, reflecting the challenges of AI-driven infrastructure and inventory overloads.

What’s Changed—and What Hasn’t

The Good News:
- Regulatory wake-up calls: The Securities and Exchange Commission now demands stricter disclosures for AI-driven companies.
- Core tech resilience: Unlike the dot-com era’s reliance on unproven “.com” business models, today’s AI tools like cloud infrastructure and cybersecurity are mission-critical.

The Bad News:
- Ethical gaps persist: A 2024 McKinsey report found 40% of U.S. working women still face sexual harassment—a rate unchanged since 2020, despite AI’s role in workplace interactions.
- Overhyped valuation: Startups like Zepz (AI for logistics) and HerMD (AI healthcare) collapsed after failing to prove profitability, echoing dot-com-era flameouts.

The Bottom Line: Accountability vs. Innovation

Blodget’s analysis underscores a critical truth: AI isn’t the problem—human behavior is. The dot-com crash taught us that unchecked growth and speculation are unsustainable. Yet today’s tech sector faces the same reckoning, with public trust in AI-generated content dropping to 31% (per a 2024 Pew study) amid fears of bias and misinformation.

The key to avoiding another crash? Learn from history. Blodget’s own missteps—from securities fraud to AI “harassment”—show that innovation without ethics is a house of cards. Companies must prioritize:
1. Profitability over growth: Layoffs in 2025 targeted firms that prioritized scaling over revenue (e.g., Ola Electric’s 1,000 cuts).
2. Regulatory alignment: Compliance with AI transparency laws, like the EU’s AI Act, could prevent another wave of scandals.
3. Workplace accountability: Treating AI tools as extensions of human teams—without repeating past biases—is non-negotiable.

Conclusion: The Next 25 Years

In 25 years, will we look back on today’s AI boom as another cautionary tale—or a sustainable leap forward? The answer hinges on whether we’ve truly learned from the dot-com era.

The numbers tell the story:
- $350 billion: The projected AI market size by 2030, up from $150 billion in 2023.
- 40%: The drop in public AI trust since 2020, mirroring the dot-com era’s eventual disillusionment.
- 2,500 layoffs at HPE: A stark reminder that even foundational AI infrastructure faces growing pains.

Blodget’s journey—from bubble architect to bubble analyst—proves that tech’s cycles aren’t just about code. They’re about culture. This time, let’s innovate with eyes wide open.

The NASDAQ’s post-2000 recovery shows that tech can rebound—but only after painful corrections.