Algorithms, machine learning, transhumanism: Terms like these prompt apocalyptic warnings from distinguished leaders such as Elon Musk and Pope Francis. And yet, the same technology that can create a dystopian government also powers a robotic cat that kindly reminds its users to take their medications. The difficulty of artificial intelligence, also known as AI, is in the myriad of applications; scientists and philosophers alike can’t decide where this life-altering technology may lead. One thing they do know is that once the robotic cat is out of the bag, there’s no going back. Our ethics can’t keep up with the (literally) superhuman speed of progress in the field of AI reprogramming our world for better and for worse. Oxford mathematician and Christian philosopher John Lennox delves into the wired world of AI with a Christian worldview, distinguishing dystopian fiction from fact, and illuminating how AI might reprogram our perception of ourselves, humanity, and God.

A simple reference to AI conjures an array of nightmarish images, from robots towering over skyscrapers to governments implanting digital chips into citizens’ brains. How founded are these fears, though? The term “AI” goes back to 1956, when computer scientist John McCarthy sought a name for this emerging field of technology. Though it’s often used as a blanket statement to refer to anything even mildly robotic or dystopian, most modern developments in AI involve a machine tasked with a single purpose only (and that’s usually not to take over the world). Though AI technology can complete a task with ultrahuman speed and efficiency, that’s about all it can do.

While there are various kinds of AI, the one we’re most familiar with today—the voice behind our Alexas and the tech that sifts through job applications—is called narrow AI. Narrow AI operates according to a predetermined algorithm, which is a human-made manual that instructs the technology in how to tackle a particular task. Algorithms have existed since the shifting, sand-covered times of ancient Babylonia in 1800-1600 BC. Stanford computer scientist Donald Knuth has studied ancient algorithms extensively and found those formulas addressed an array of problems, just like their contemporary algorithmic descendents. The author notes that early AI functioned by way of human-operated algorithms. The narrow AI of today employs “machine learning” to maximize algorithms’ potential to enact solutions for a wide range of issues. In machine learning, computer scientists build one general algorithm, which then uses information gathered from past data to predict and contrive solutions for the future. Once the general algorithm is formulated, humans play no role in determining how the machine will arrive at its conclusions. 

Amazon uses at least one example of this kind of narrow AI. That yoga mat you bought last week will probably bring up a range of new items you might consider purchasing, like water bottles or leggings. Amazon’s algorithms pick up on your browsing and purchase history, aggregate that data, and then project what else you might like to buy based on the data of other customers like you. Narrow AI’s ability to solve problems in the medical field is also groundbreaking. The Christian inventor and AI leader Rosalind Picard joined with her research lab at MIT and Empatica to use machine learning AI tech to create a smartwatch that monitors wearers’ skin to detect possible seizures. If the watch senses a seizure, it promptly alerts a caregiver who can help if the patient loses consciousness or, more dangerously, the ability to breathe. This technology is already in place in the US and the UK, and it’s projected to save the lives of many people suffering from epilepsy.

Naysayers and soothsayers aside, the potential of narrow AI to solve a particular defined problem or set of problems is, no doubt, beneficial for everyone. Danny Crookes, a professor of computer engineering at Queens University Belfast, observes that AI aren’t so smart after all, they’re just good at what they do. Narrow AI are experts at focusing on a single problem and bringing forth a solution. There’s a small chink in AI’s steely armor, though—the danger lies in how the technology arrives at a projection in the first place, a process that involves a potent four-letter word we know so well: data.

Facebook, Google, Instagram—these platforms know more about you than you think, and they use that data to profit off your habits. Employing Harvard professor Shoshana Zuboff’s  groundbreaking though ominous work entitled The Age of Surveillance Capitalism, the author notes that the 2.5 billion of us tethered to our rectangles of light are falling prey to increasingly exploitive corporations. In the absence of clear rules concerning data ownership, corporations gather huge swaths of otherwise personal information to make more money. While a more personalized Google search or a social media feed geared toward a person’s individual interests may seem harmless and beneficial to its users, the methodical means to these useful ends are deceptive. 

The growth of narrow AI technology and the accumulation of user data as varied as one’s favorite band, facial structure, and credit card information, creates fertile soil for growing a dystopian nightmare. China is one place where AI data has turned catastrophic. In the communist controlled country, narrow AI flourishes; in 2018, China projected that by 2020, more than 400 million CCTV cameras would line its streets and peer into the eyes of passersby through facial recognition technology. The Chinese government employs this data from facial recognition and other AI devices to monitor and dictate desired behavior through its social credit system. In 2018, Channel News Asia issued a statement declaring that nine million people already had been barred from airplane flights or train trips based on information about their behavior picked up through the social credit system. 

Canceled flights and missed business trips aren’t the height of China’s AI-fueled oppression, though. The government weaponizes AI technology to control its minority Muslim population, the Uighur people in the region of Xinjiang. Several entities, including British newspaper The Times, non-governmental organization Human Rights Watch, and The Wall Street Journal document the brutally monitored lives of the Uighur population. Uighurs who stray from China’s cultural norm are forced to hand in their phones so that their personal information can be added to a database for AI sifting. QR codes line the doorways of homes too, making it easier for police to monitor the occupants. Presumably, these QR codes allow police officers to track Uighur people who pose potential threats to their regime in order to create a record of their behavior. If a Uighur person fails to live up to China’s technologically mandated expectations, she is sent to a “re-education” camp where she is effectively brainwashed. She is forced to let go of her culture all because a device read danger in the contours of her face.

Maya Wang, a Human Rights Watch researcher for the area of China, sees the country’s use of AI technology as a stark warning, not only for the future of China, but also for the rest of the world. The author notes that most developed countries hold the technological power to operate surveillance regimes akin to China’s. In 2019, The Times reported that the UK purchased 1.2 million CCTV cameras from a Chinese company. Currently, use of this technology outside of China is supposedly confined to safety and security, but it doesn’t take much for this usage to turn. The data is there and the AI is in place—steely and forbidding, it offers various different futures, intermingling tones of divine omniscience with the dystopian.

The evolutionary arrow of narrow AI diverges into two equally frightening directions: those of Artificial General Intelligence (AGI) and transhumanism. Narrow AI is the only operable form of AI today, and even its relatively harmless content can be twisted for tyrannical ends. AGI, on the other hand, strives for a more ambitious goal. Instead of confining itself to addressing a single problem, it seeks to address all manner of problems, just as a human would. AGI is a form of “superintelligence,” and experts expect it could exceed the capabilities of human intelligence by 2084. The apocalyptic predictions of Google’s director of engineering, Ray Kurzweil, hold that AGI will surpass humans in the near future, with AGI machines wrenching themselves from their creators and dictating their future progress on their own.

While AGI projects a foreboding future, with a superintelligent technological species that possesses the power to usurp humanity’s position of supremacy, this isn’t the worst of the vision. Transhumanism offers an even more disturbing future. The author draws upon the highly influential work of historian Yuval Noah Harari and his book Homo Deus to paint a portrait of transhumanism, a frightening project to build a better human. “Homo deus” translates tellingly (and chillingly) to “human god,” and it posits that human biological evolution, aided by technology, will resurrect a new, better form of humanity—one that’s immune to death, living its days in idyllic, everlasting pleasure. Max Tegmark, a physicist and professor at MIT, sees the evolution of AI as drawing toward a new kind of life and a new way of being, a transformation transposed as our flesh falls and scales of metal burst in their place.

Don’t panic just yet—developments like these assume several factors in order to become a reality. Consciousness is a big issue here. If AGI can build robots as superior replicas to humans, then they must be conscious. In order for those robots to attain human consciousness, scientists must know exactly what it means to be a conscious being, which of course, is the glaring problem. Even the most ardent transhumanists have a long way to go before narrow AI gives way to AGI and their momentous Transhumanist Project. 

Still, many researchers proceed with caution, wading the silent, ebbing waters of AI developments slowly and with question. Thought leaders like Elon Musk and Stephen Hawking, along with 8,000 others, signed a cautionary letter in 2015, acknowledging the potential of AI to go awry. C. S. Lewis’s prophetic book The Abolition of Man posited as early as 1943 that humanity’s desire to conquer nature with technology will result in powerful, technologically equipped humans conquering other humans. The silver glow of technology is tempting to humans who constantly strive, seek, and stretch beyond themselves and into the next, whatever that may look like. 

But where do we look for guidance as to whether or not we should replace our flesh for steel and our minds for machinery? When the world turns gray, where do we look for color? We should look to ethics, worldview, and God.

Ethical concerns stand at the forefront of the public’s AI imagination. The most prominent voices in AI are those of atheists like Max Tegmark, a physicist who constructed various scenarios in which AGI overtakes the earth. While the atheist voice appears to dominate, Christian scientists like John Polkinghorne and other atheists like John Gray see the blatant error in scientific progress grounded in a Godless worldview. Even Charles Darwin acknowledged the flaw in his thinking: If we’re all mere creatures driven by evolutionary appetites, the practice of science is nothing more than another mechanistic drive for meaning. We are clocks that tick to the rhythm of ourselves, and science is simply another servant to our animalistic rhythm.

The author argues that for this reason, science and the naturalist worldview are contradictory; a naturalist silences her own argument with the cry that we are nothing more than stuff of the earth, fading and meaningless. Within this worldview, all ground for universal morality shifts and shatters beneath us, and the value of the human being crumbles. Atheist philosopher Jürgen Habermas cautioned culture about the inevitable ethical destruction that would result from the rupture of the Judeo-Christian worldview. This is what makes the quest for AGI and the Transhumanist Project so dangerous to the morality of society—its progress drives in ethical darkness.

The things we create are not autonomous; rather, they’re reflections of what we as a people believe. If we don’t believe in God, we have no grounds for a transcendent morality. If we have no ethical code to inform our work, what will our creations even look like? Harari notes that the transhumanist dream is another manifestation of the human desire to act like gods; there’s no goodwill in it at all, just another attempt to lord power over others. The Christian worldview must find a spot in the ethical conversations surrounding AI developments. Only then will technology be approached in a way that’s not exploitative of other human beings. A thought leader in AI herself, Rosalind Picard is a Christian innovator who advocates that machines must be built with ethics and morality in mind. Picard developed a new field in AI called “affective computing,” which seeks to engineer AI tech that can read and portray human emotions in order to help other people. Her work has already benefited those with autism who have a difficult time discerning the emotions behind facial responses. AI can be ethical, but first it must be created within a worldview that prizes the human being as an individual with value and meaning.

The biblical account of Genesis expresses that humans are more than material; reality extends beyond what we see. True reality rests in God, who transcends the physical with His spirit. John 1:1 says that “In the beginning was the Word, and the Word was with God, and the Word was God,” using the Greek term logos to denote ‘word.’ The Greek use of logos posits an underlying, fundamental order upon which the world rests, and in Christianity, this logos is Jesus, the ultimate groundwork of reality. In Genesis, He gave our material bodies nephesh, which is Hebrew for “soul” and “life.” The desire to engineer conscious beings is ultimately a desire to replay the Genesis story with humans as gods, but this ignores a stark fact: Only God can give life, and only He can breathe meaning into dust.

The author notes that the real, truest act of Homo Deus exists in God becoming human—an event that speaks to the overwhelming value He places upon each person not as simple bits of material, but as His children. Humanity has clawed after the throne of God for centuries, from the time of Genesis to the contemporary era of AGI and the Transhumanist Project. In the eyes of many AI developers, AGI developers, and transhumanists, humanity is a product that can be improved through its own wit and ability; death and discontent aren’t essential facets of a fallen human nature, they’re simply issues to overcome. This is exactly why all projects that envision a utopia for humanity fall into disillusion and chaos—the real issue at the core of humanity isn’t something we can patch up on our own. Sin that distances us from God isn’t a bridge we can cross with our own feet. Rather, only through God and the death and resurrection of Jesus Christ is our real problem remedied.

The author discusses the inevitable outcome of the world before the Second Coming of Christ, as described in Revelation 13. The apocalyptic beast of Revelation is the same one discussed in Daniel 7 and denoted by Paul as “the man of lawlessness” in 2 Thessalonians 2:3. According to the author, the 10 horns of the beast denote 10 world powers united beneath a singular entity. The author notes that the beast is not a collective evil but a single person, just as Revelation 13:18  says, “Let the person who has insight calculate the number of the beast, for it is the number of a man.” The Bible, Aldous Huxley’s Brave New World, and George Orwell’s 1984 all see the inevitable end of humans craving divinity: ultimate power and tyranny waged by the hands of a single person. Society’s movement toward greater centralization, aided by developments in AI, tills the ground for world power apart from Christ to fall into the hands of a tyrant parading as God. The European Union has already amassed an impressive 27 countries in its collective order, and the author notes that the problems plaguing the world today, including global warming and economic affairs, make a centralized source of global power even more attractive.

Rather than discourage us, these warnings should mobilize us. Whether we’re narrow AI scientists seeking a way to build an ethically aware vehicle or social media users fighting to protect online privacy, we are people created by God with purpose, meaning, and value. Though we should approach this world and developments in AI with caution, Jesus tells us, “Do not be afraid of those who kill the body but cannot kill the soul.” (Matthew 10:28) Instead of completely cutting oneself off from the world, Paul admonishes people to recognize the inherent instability and corruption in humanity and in worldly systems, in order to do one’s best to live as faithful followers of Christ.

It’s freeing to know we don’t have to build our own salvation—redemption isn’t a transhumanist project of temporal steel and relentless human zeal. God is the intelligence within which the earth turns. Living within God is a gift to people, who aren’t mere material creations or man-made mechanisms, but divine inheritors of eternity.