Doom or Bloom
问题 1
你认为AI对我们的未来意味着什么？为什么？
AI could give us an extraordinary future—helping cure diseases, improve education and accelerate science—but only if we build powerful tools that remain under human control. A medical assistant is a tool. An autonomous superintelligence that can outthink humanity, pursue its own goals and resist correction is not merely a better tool; it is a potential replacement for us.
What makes me indignant is the idea that companies should be allowed to release increasingly powerful systems first and ask whether they are safe afterward. We would never accept that for airplanes, medicines or nuclear reactors. Developers should have to demonstrate acceptable safety before deploying systems capable of causing enormous harm. Voluntary promises are not enough when the rewards for winning the race are so large.
This race is a political choice, not a law of nature. Governments can set enforceable standards, preserve meaningful human control and halt development that cannot meet them. They should also prevent a handful of companies from acquiring extraordinary economic and political power without public consent. I am deeply optimistic about AI’s benefits on a different path. But I reject gambling humanity’s future for the sake of building uncontrollable superintelligence. Let’s build tools, not beings, and keep humans in charge.
