Isn't it funny how smart people sometimes fall prey to dumb ideas?
Bill Gates, for all his panicky tendencies over things like climate change, isn't a dumb guy. In his younger years, he was a key player in the beginnings of the Information Revolution; in part due to his efforts, pretty much every home in the developed world has a computer, and a majority of them are running Microsoft Windows, which was originally Bill Gates' brainchild. But that was a long time ago now, and these days Bill is more into shouting about the sky falling.
He's been into climate scoldery for some time, and has grown somewhat quieter on that front of late because fewer and fewer people are worried about it, mostly due to the fact that the panicked predictions of the panic-mongers never come true. But now Gates has found a new booger-man to be afraid of: Artificial Intelligence (AI) and, of course, he has a government solution in mind to assuage his panic.
Tech billionaire Bill Gates warned that artificial intelligence had the potential to kill a billion people in the wrong hands — as he joined the call to put safety restrictions on the technology.
The Microsoft co-founder called for greater government oversight on AI during a talk with NBC’s “Meet the Press” on Sunday, where he suggested AI could be abused and lead to humanity’s near extinction.
“AI is certainly powerful enough to drive events that cause a billion deaths,” he warned.
“So even though it’s pretty hard to get to 100% [of humanity], there’s never been a weapon as powerful as the combination of people with ill intent using the latest AI tools,” he added.
Gates explained that without the proper guardrails in place, it would be simple for potential terrorists or anyone with ill intent to use the tech to develop weapons or biological horrors.
Here's the problem: Just like people who screech for gun control, Gates is blaming the sword for the hand that wields it. AI does indeed pose certain dangers, but "proper (government) guardrails" won't make a lick of difference. Why? Here's why, and I'm fixing to tell you.
First: Bad people will always do bad things, no matter what tools are available to them. AI is a tool; nothing more. And there's another catch: Unlike a rifle, or a rocket launcher, or even a nuclear warhead, you can't seize possession of an AI program. You can't lock it away in a safe. It's ephemeral. It's out there, in the internet. It is and always will be; the internet, as the saying goes, is forever, and countries like Russia, China, and, yes, Iran aren't liable to be too concerned about any "guardrails."
Second: Because AI is ephemeral, it has physical limits. You might use an AI program to tell you how to make an atomic bomb, but the AI doesn't come with the enrichment facilities and labs that will enable you to actually put together a device. An AI may tell a goblin how to genetically engineer a virus, but it won't provide the laboratories and equipment with which to do so. It is those things, those physical devices and facilities, around which we should be worried about guardrails or the lack thereof.
No, in this, Gates is simply wrong. There are evil people in the world who will do evil things. But instead of spending time and resources trying to put "guardrails" around software in our modern connected world, a task that is a lot like shoveling flies across a barn, we should spend that time and those resources dealing with the evil people who would put this software to evil use. Any other course of action is ultimately futile.