Opinion | The Luddite and the engineer
- Democrat Senate hopeful wants permanent ban on AI smarter than humans, ‘kill switches’ mandatory, and prison for tech executives.
- Plan echoes Bernie Sanders, demands new Cabinet agency to freeze frontier models and nationalize top AI labs — treaty with China included.
- Microsoft's Nadella counters with cybersecurity 'zero-trust' approach: treat AI as potential insider threat, limit access via software, not corporate promises.
- Luddite nostalgia fails; engineers say regulation of AI use, not intelligence itself, is the only viable speed bump on the road to progress.
Brief Summary
Senate hopeful Abdul El-Sayed wants to hit the pause button on human evolution, proposing a permanent ban on any AI that could outthink a college freshman. His plan reads like a wishlist for a tech police state: freeze model releases, nationalize the labs, and slap prison terms on execs who dare ship a clever chatbot. He's even dreaming of a US-China treaty that outlaws superintelligence — good luck getting Beijing to sign off on that. This is fearmongering dressed as policy, and it's as likely to stop AI as King Canute was to stop the tide. He and Bernie might as well be smashing looms in a 19th-century cotton mill. In contrast, tech's top engineer, Satya Nadella, proposes a saner path: forget banning brains, and instead build them into secure systems with strict, auditable limits on what they can do, enforced by software not self-governed promises. Treat AI like a potentially rogue employee on your company network — always verify, never trust the machine's intentions. It's the only approach that doesn't involve smashing all our shiny new toys.
Why This Matters
Beyond the obvious political theater, this fight is about whether progress gets to keep its current brisk pace or gets strangled in red tape. A 'safe AI' certification hall, ala the El-Sayed vision, would hand Washington enormous power over every piece of software you touch — and likely stall innovation for a generation. For the average reader, it's a battle between a future where your tools are strictly monitored under strict liability and one where vague regulatory threats just make your software shoddier out of fear. Nadella's 'zero-trust' perspective is the only option that keeps your data, your job, and your gadgets actual usable. If El-Sayed wins, expect more bots named 'Compliance Assistant,' and less actual help. The choice is between pragmatism and a bureaucratic doomsday for your childrens' gadgets.