Communications technology keeps us in touch. Medical technology saves our lives. Other forms of technology do our math and accounting for us, support our welfare on nearly every front.

But technology also puts us at risk, and never more than when we extrapolate from the technology we do have to the technology we don’t have. When we think of technology as the magic bullet.

Take the problem of high-level nuclear waste. Ever since the Manhattan Project we’ve assumed that the problem of disposing of radioactive waste, or somehow altering it so its toxicity wouldn’t threaten us for millennia, would be solved.

Now, after decades of research, weapons development and nuclear power plant operation, the problem isn’t solved. At the Vermont Yankee power plant near Brattleboro and elsewhere in the country, caches of radwaste must be carefully secured to avoid disaster.

We’re sure that science has the potential to clean up spills of toxic substances such as oil. And it may—but now we have an oil spill in the Gulf of Mexico that’s so catastrophic that the full effects aren’t even known yet, and we don’t have the technology to clean it up. Once again we see that we’re stuck—that a hypothetical state-of-the-art solution doesn’t solve today’s problem.

With arable land disappearing all over the globe, surely technology will help us shrink the amount of space it takes to grow enough to feed the world; that’s what our chemical manufacturers were happy to help us believe. But when costs (in water as well as money), environmental pollution and diminishing yields in relation to energy put in are taken into account, the “green revolution” has had mixed results. And according to a Union of Concerned Scientists report, Failure to Yield, genetically modified seeds have not produced significantly increased yields of basic commodities like corn and soybeans.

It’s easy for people who have seen dramatic technological changes to believe that technology can do anything. No one is saying that we have reached the end of our ability to make such advances. But we invite disasters when we fail to distinguish between the capabilities we have and those we don’t have, and gamble on the latter.