In general, I root for technology and technological progress. Overall, I believe it has been a net benefit for humanity. It’s one of the major reasons why we’ve made so much progress as a global society in the past 100 years.
I’ve sung the praises of technology in the past, speculated on its potential, and highlighted individuals who have used it to save millions of lives. For the most part, I focus on the positives and encourage other people to have a less pessimistic view of technology and the change it invites.
That said, there is another side to that coin and I try not to ignore it. Like anything, technology has a dark side. It can be used to harm just as much as it can be used to hurt, if not more so. You could argue that we couldn’t have killed each other at such a staggering rate in World War II without technology.
It’s not hyperbole to say that certain technology could be the death of us all. In fact, we’ve come distressingly close to destroying ourselves before, namely with nuclear weapons. There’s no question that kind of technology is dangerous.
However, artificial intelligence could be far more dangerous than any nuclear bomb. I’ve talked about it before and I’ll likely bring it up again. This technology just has too much potential, for better and for worse.
That’s why when people who are actually researching it have concerns, we should take notice. One such individual spoke out recently, specifically someone who worked for Google, an organization with deep pockets and a keen interest in Artificial Intelligence.
According to a report from the Associated Press, a scholar named Timnit Gebru expressed serious concerns about Google’s AI research, specifically in how their operating ethics. For a company as big and powerful as Google, that’s not a trivial comment. This is what she had to say.
Prominent artificial intelligence scholar Timnit Gebru helped improve Google’s public image as a company that elevates Black computer scientists and questions harmful uses of AI technology.
But internally, Gebru, a leader in the field of AI ethics, was not shy about voicing doubts about those commitments — until she was pushed out of the company this week in a dispute over a research paper examining the societal dangers of an emerging branch of AI.
Gebru announced on Twitter she was fired. Google told employees she resigned. More than 1,200 Google employees have signed on to an open letter calling the incident “unprecedented research censorship” and faulting the company for racism and defensiveness.
The furor over Gebru’s abrupt departure is the latest incident raising questions about whether Google has strayed so far away from its original “Don’t Be Evil” motto that the company now routinely ousts employees who dare to challenge management. The exit of Gebru, who is Black, also raised further doubts about diversity and inclusion at a company where Black women account for just 1.6% of the workforce.
And it’s exposed concerns beyond Google about whether showy efforts at ethical AI — ranging from a White House executive order this week to ethics review teams set up throughout the tech industry — are of little use when their conclusions might threaten profits or national interests.
I bolded that last sentence because I think it’s the most relevant. It’s also the greatest cause for concern. I suspect Ms. Gebru is more concerned than most because the implications are clear.
When a tool as powerful as advanced AI is developed, who gets to determine how it’s used? Who gets to program the ethical framework by which it operates? Who gets to decide how the benefits are conferred and the harms are reduced?
Moreover, how do you even go about programming an AI with the right kind of ethics?
That’s a very relative question and one we can’t avoid if we’re going to keep developing this technology. I’ve tried to answer it, but I’m hardly an expert. Ms. Gebru was definitely in a better position than me or most other people with a passing interest in this field.
Then, she gets fired and starts expressing concerns publicly. The fact that she can and Google isn’t facing much in terms of repercussions should be concerning. It may also be a sign of the larger challenges we’re facing.
Google, like many other organizations researching advanced AI, is a profit-seeking tech company. They’re not some utopian technocrats. They’re a business who is obligated to make their investors happy. Advanced AI will help them do that, but what kind of consequences will that invite?
If profit is the primary motivation of an advanced AI, then what happens when it encounters a situation where profit comes at the cost of lives? There are already human-run companies that make those decision and people die because of them. An advanced AI will only make it many times worse.
Once an artificial intelligence system is as smart as a human, it’s going to be capable in ways we don’t expect and can’t control. If it’s ethics and goals aren’t aligned with us, then what’s to stop it from wiping humanity out in the name of profit?
It’s a distressing thought. It’s probably a thought that has crossed Ms. Gebru’s mind more than once. She may know how close or far we are to that point, but the fact that this is already a conflict should worry us all.
We’ve already become so numb to the greed and excesses of big business. Tech companies may conduct themselves as this team of future-building visionaries intent on making the world a better place, but the profit motive is still there. Like it or not, profit is still a hell of a motive.
Eventually, artificial intelligence will get to a point where it will either adopt our ethics or choose to formulate its own, which may or may not align with ours. When that happens, no amount of profit may be worth the risk.
Now, we’re still a ways off from an artificial intelligence system on that level, but it’s still quite possible that there are people alive today who will grow up to see it. When that time comes, we need to be damn sure these systems have solid ethical frameworks in place.
If they don’t, we really don’t stand a chance. We’re a society that still kills each other over what we think happens when we die without seeing the irony. Even a marginally advanced AI will have no issues wiping us out if we make doing so profitable.