LOTS of spoilers ahead.
So Transcendence is out. It’s the latest movie from Christopher Nolan…’s cinematographer, Wally Pfister. Nolan worked on the movie, too (as executive producer), and it’s pretty easy to see his hand in the film. Another rather intellectual Nolan film was Inception, which a movie about dreams that is somewhat confusing and yet still somehow makes sense (it’s like he planned it that way). It was received quite well…so why isn’t Transcendence?
Basic rundown of the movie: Johnny Depp is a prominent Artifical Intelligence (AI) researcher who is the victim of an assassination attempt by a bunch of neo-luddite anti-technology crazy people. As he is dying from a polonium-laced bullet wound, his fellow researchers (wife Rebecca Hall and best friend Paul Bettany) work on copying (not transferring, an important distinction) his consciousness into a powerful computer. They succeed, with Depp dying and being reborn as a computerized mind. Abandoned by hyper-skeptical Bettany, Hall connects her compu-husband to the Internet just as the luddites come a-knockin’. The first thing Depp does is make a lot of money doing high-volume trading on the stock market (although it’s night, so it must be a different stock market than the Dow? I admit I forget the specifics) in order to secure a lot of money for his wife’s safety. Hall makes her way to the desert, where she acts as her husband’s agent while they build an underground data haven to act as Depp’s base of operations and research lab. Two years pass.
Here’s where things start to go a little strange. Morgan Freeman, former teacher and friend to Depp and Hall, is somehow recruited by the luddites (who are now, the reviewers tell me, supposed to be portrayed as the protagonists), along with Bettany. Their goal: upload a virus into the AI to destroy it, wiping out the Internet in the process (Depp’s everywhere at this point). Meanwhile, Depp has spent two years perfecting nanotechnology, which he uses to repair and enhance people who are taken to the facility to have their illnesses or injuries treated. In return, they are connected to Depp via the Internet—they remain autonomous, and most go back to their regular lives—so they can act on his behalf and help construct new solar panels to power Depp’s mind.
The luddites, along with government agent Cillian Murphy, eventually capture the increasingly-freaked-out Hall (she claims Depp is invading her privacy by reading her mind; he’s reading her hormones, not much different from me reading the expression on your face) who agrees to have the virus injected into her via captured nanomachines, which Depp has sent into the atmosphere to rain across the world in…well, rain. During the attack on the compound, Depp and Hall are reunited, with Depp now controlling a regrown version of his own body.
I’m going to diverge here for a moment to explain a few of the threads that have been running through the film that apparently nobody else but me noticed. First is the phrase “people fear what they don’t understand.” At one point during the climax it is literally repeated over and over again in the background. As an explanation of the illogical actions of some of the characters, it’s a bit thin, but you have to admit it works.
This thread ties in with the others. The question is asked at least twice: can you prove you are self-aware? (No. It’s called the Problem of Other Minds.) Tying into this, the immediate assertion that AIs are incapable of emotion is just…made. No real logic behind it. It bothered me at first, but the movie does a pretty good job of hitting you in the face with it: fear.
Another one: is AI Depp actually Depp, or just a being of vague malevolence whose motivations are utterly beyond comprehension? The luddites’ assertion is that the AI isn’t Depp; Hall’s assertion, at least for a while, is that it is. In fact, the answer is yes, only believed when Depp puts his old human body back on—something that makes him relatable again. Why jump to the assumption that it isn’t? Fear.
You get the idea. Eventually Hall is injured during the attack and Depp takes her underground again. The luddites, fighting off the indestructible superhuman townsfolk (all of whom are nanobot-enhanced now), hold unenhanced Bettany at gunpoint to force AI Depp to take his wife’s nanomachine virus into his software by uploading (here, transferring; remember the distinction) his dying wife into the Internet with him. AI Depp, having too little power (due to the damage wrought by the luddites) to both upload the virus and fix his wife’s body, decides to follow his wife’s begging (by this point she’s begged him several times to kill himself because he’s somehow hurting the people outside) and uploads her into the Internet with him…just before it shuts down completely.
So fast-forward three years to the beginning of the movie. It’s a post-Internet age, where people are going hungry and electricity is scarce (it makes sense, just not a whole lot). Bettany is watching water droplets fall from the plants in his old friends’ garden…when he sees the nanomachines Depp rained into the water are alive and well. Other shots from just before this scene show the Earth being revitalized by nanomachines: trees growing, water purifying, that sort of thing. It’s heavily implied that Depp and Hall are still alive in the nanomachine network, now part of the Earth.
So, to beat you over the head once again: why do the luddites destroy someone who is only trying to help them, who has been in fact their friend’s mind the entire time? Fear.
Why do the luddites behave so strangely irrational? Fear. Why do Freeman and Bettany immediately assume AI Depp is a mere facsimile of his human self and immediately plot to destroy him? Fear. Why does long-time AI researcher Bettany assume immediately that AIs are not capable of emotion, even though Depp has been doing everything he could to make Hall (his wife, remember) happy by fulfilling her dreams (at one point he even says that Human Depp only ever cared about her)? FEAR.
It’s FEAR! What’s scary about a godlike posthuman intelligence that is so radically incredibly fantastically mind-bogglingly superlatively intelligent that it can create nanomachines from practically nothing in just a couple of years? EVERYTHING. It’s damn frightening to be in the presence of something so far above you. That’s why the augmented townsfolk weren’t afraid: they had been augmented to be at Depp’s level. Why? Because stupid workers were of no use or interest to him! He was still human, and a smart human. He wanted smart people around him but he wasn’t crazy. He left them their autonomy because playing with puppets gets boring after a while.
Depp’s original incarnation uses the word “transcendence”, but it’s really (as he states) the Singularity. What is the Singularity? Imagine a black hole. The point at which no light can escape the black hole is called the Event Horizon. Beyond that is the black hole’s Singularity. It is impossible for us to see within it: light (that is, information) cannot escape. Similarly, we cannot see past the Technological Singularity, because change will be happening so quickly that our unaugmented intelligences cannot imagine it. And what we cannot imagine, we cannot understand, and what we cannot understand we fear.
The majority of the moviegoing audience may have thought the movie was trying to get them to sympathize with the (murderous) neo-luddites, but that’s not what I felt at all. I was convinced the entire time that the protagonist was AI Depp. Was I creeped out by the uncertainty of his motivations? Absolutely, as I should have been; this is deep shit we’re considering here. It’s healthy to be disturbed a little by this sort of thing. Was I confused by the luddites’ arguments? Sure, but Human Depp himself notes that a group of anti-technology terrorists who want to stop AI research to save humanity yet does this by murdering AI research teams is somewhat lacking in logical ability. Is it strange that Bettany and Freeman, two AI researchers themselves, are so immediately swayed by the luddites? Sure, but they’re clearly not as personally inured to the concepts of AI research as they had thought. They’re afraid.
Also, they must not read Frank Herbert:
“I must not fear. Fear is the mind-killer. Fear is the little-death that brings total obliteration. I will face my fear. I will permit it to pass over me and through me. And when it has gone past I will turn the inner eye to see its path. Where the fear has gone there will be nothing. Only I will remain.” Frank Herbert, Dune
Transcendence is a fantastic movie, balancing—with, admittedly, some parts succeeding better than others—the creepiness of the unknown with the hope of technology, the anger of fear with the sadness of love, and how they all drive us to lengths of which we did not know we were capable. It’s natural to fear. Humankind has been afraid its entire history, and the future will be no different. The wrong reaction to fear is to shun what makes you afraid. The correct reaction is to attempt to understand it, as it is only through that can we conquer our fears—and make them work for us.
For more on AI, look up Eliezer Yudkowsky. He talks a lot about the inevitability of AI, and the need therefore to understand AI and to make a friendly one before someone makes an unfriendly one. On the other hand, a truly intelligent being would likely understand that in the long term—the only term that really counts—friendliness is better than wrath. I’m not worried about Skynet happening; I’m worried about Transcendence—the good parts, at least—not.