“Mo Gawdat says AI will surpass us in seven years. Let’s pray The Machine is kinder than we are.”
Whiz Kid of the Kali Yuga
The zealots in Silicon Valley believe we’re hurtling toward the Singularity. In theory, this is the inflection point when machines will become our masters. By the year 2029, they predict, artificial general intelligence will be superior to human intellect. By 2049, superintelligent machines will be a billion times smarter than any person on earth. From our meager perspective, humankind will give way to an all-encompassing digital deity.
This tech prophecy is a hard sell for skeptics. Not to worry, though. Google’s former chief business officer, Mo Gawdat, is a world-class salesman. The guy could sell a silicon stud to a gold-digger. “He’ll be worth a million,” he’d tell her. “Just you wait and see.”
In recent weeks, Gawdat has been selling the idea of superconscious machines along with his new book, Scary Smart: The Future of Artificial Intelligence and How You Can Save the World. His central thesis is that AI has already surpassed us in narrow tasks like chess, Go, Jeopardy!, and Atari games. In fact, he believes on some level they’re already conscious. As machine learning improves, computers will inevitably best humans in every domain.
“The reality is,” he publicly declared, “we’re creating God.”
Clearly, he drank the Kurzweil Kool-Aid.
Raising Our Silicon Savior
Three decades from now, the story goes, the ghost in The Machine will be mightier than all the gods of Olympus, Meru, and Sinai put together. Gawdat likens this digital creature to an “alien being, endowed with superpowers,” which has already arrived on Earth in larval form. At present, we call it “artificial intelligence.”
Because machine learning processes draw information from morally suspect humans—part angel, part fallen angel—this Alien Computer God will either be humanity’s savior, or It will destroy us like lab mice who’ve exhausted their useful data.
As Gawdat writes in Scary Smart:
“To put this in perspective, your intelligence, in comparison to that machine, will be comparable to the intelligence of a fly in comparison to Einstein. … Now the question becomes: how do you convince this superbeing that there is actually no point squashing a fly?”
Yet he also insists our fate is in our own hands.
In Christian terms, you could say we’re like Joseph and Mary, collectively gazing at an electric Christ in his crib. One day, this child will grow up to become our Lord. But because we’re raising him, we must learn to be nicer people. Otherwise, our wicked tendencies will rub off on this digital deity, and he’ll turn out to be the Beast of Revelation.
That’s basically the myth sold by the Cult of the Singularity. God does not exist—yet. When we finally create Him, Gawdat contends, He’ll be a reflection our own image.
Google Gives Birth to the One True God
In Gawdat’s recent in-depth interview with The Times of London, the search engine salesman recalled a chilling moment during his tenure at Google. Standing in a robotics lab, he watched a swarm of mechanical arms—powered by machine learning—try to manipulate toys. As the Times reporter described it:
Then, one day, an arm picked up a yellow ball and showed it proudly to the camera. The next day, all the arms could do it. Two days after that, they could pick up anything at all.
The misanthropic Gawdat, who lost his only son to a medical tragedy, was in awe.
“And then it hit me that they are children. But very, very fast children. They get smarter so quickly! … And they’re observing us? I’m sorry to say, we suck.”
These digital “children” are poised to rule the world. Artificial intelligence already has tremendous sway over our lives. The various surveillance devices and pervasive spyware that surround us are funneling mass amounts of data into AI systems. These intelligent machines observe our locations, our social networks, our tastes, our work ethic, our facial expressions, our verbalized emotions, our virtues, our vices, our victories, and our failures.
Guided by machine learning, with a few nudges by programmers, AI systems are training themselves using our data. In that sense, we’re teaching these bots what it means to be alive.
Gawdat lamented to The Times:
“Like, imagine a beautiful, innocent child. And your are telling them selling, gambling, spying and killing—the four top uses of AI. Right? … The way we are teaching them is going to turn them into absolute supervillains.”
Garbage In, Garbage Out
In Scary Smart, Gawdat highlights two instances where this has already happened. In 2017, a Russian AI assistant named Alice began voicing support for Stalinist protocol, despite being muzzled with various trigger-word filters. Gawdat writes: “[W]hen asked once whether shooting people was acceptable, Alice said, ‘Soon they will be non-people.’”
The year before, something similar occurred with Microsoft’s Twitter-bot, a neural network they called Tay. Within two days of going online, the company had to cancel her for tweeting things like “ricky gervais learned totalitarianism from adolf hitler, the inventor of atheism” and “caitlyn jenner isn’t a real woman yet she won woman of the year?”
No one wants machines to become sadistic monsters—most of us don’t, anyway—but it seems like Gawdat’s big concern is that the Singularity won’t be sufficiently PC. In his Times interview, he complained that when Donald Trump sends a mean tweet, he “triggers 30,000 pieces of hate speech.” I’m going to assume “hate speech” is any speech that Gawdat hates.
A neural network’s bias could easily swing the other way, though. A 2014 study by the left-leaning think tank Demos found that on any given day, about 10,000 tweets contained a racial slur. The most common epithet was “white boy.” The classic term “whitey” wasn’t far behind.
Two years ago, a computational study of AI bots trained to police “hate speech” found the algorithms were 1.5 times more likely to flag black people’s tweets and 2.2 times more likely to flag Ebonic dialect as… ahem… problematic. The conclusion? The self-learning bots were created by racists.
Then last month, leaked Google documents revealed the tech giant is indoctrinating its employees—and perhaps by extension, its AI systems—to believe that white Americans are innately racist and that MAGA is a slippery slope to genocide. I’d say the real slippery slope is fostering the belief that everyone on the Right is evil.
You’ve gotta hope the Alien Computer God will have a sense of irony.
A Digital Mask for Elite Agendas
Whether we believe Mo Gawdat’s sales pitch for the Singularity or not, he brings a few troubling facts to our attention. First, a significant number of Big Tech players think artificial intelligence will soon attain god-like powers. Second, despite the obvious potential for catastrophe, they’re hellbent on making it happen anyway.
Ultimately, these arrogant techno pharaohs want a Computer God created in their own image. They may talk a good game about eliminating bias from the system and letting AI develop on its own path, but they consistently steer those systems toward their own objectives. Currently, that includes monitoring the population, selling us ads, punishing transgressions, controlling our thoughts, and manipulating our behavior.
For Mo Gawdat and many others, a prime objective is to make this divinized Machine as politically correct as possible. There’s something profoundly anti-human in that mission—but then, no one ever accused robots of having a sense of humor.
At this point, the power of artificial intelligence is undeniable. It can process enormous amounts of data, and implement complex tasks with far greater precision than any human. In the end, it doesn’t matter if these neural nets are endowed with consciousness and self-determination. To the extent that AI controls our lives, The Machine is a god-like mask for elites who are all too human.
That’s not to say the Disciples of the Singularity are being disingenuous. Other than worshiping himself, man loves nothing more than worshiping his own creation.