The ugly turns taken by assorted historical vectors in the late 1960s and early ‘70s—a string of high-profile assassinations, race riots, Manson, the Weather Underground, Vietnam, Nixon, a broader awareness of impending environmental collapse—made the 1970s a particular golden era for dystopian cinema. All the above mentioned forces and more gave us the likes of Soylent Green, No Blade of Grass, THX-1138, Frogs, The Omega Man, and countless other visions of our doomed future. In and amongst all our other inescapable anxieties and paranoias was an increasing awareness of the role computers were playing in our daily lives.
Technoparanoid fears of dehumanization and power-mad machines can of course be traced back to the silent era in cinema, and much earlier than that in literature and legend, but it wasn’t until the mid-20th century that some of these fears suddenly became quite plausible. Prior to that computers were perceived by the general public as little more than glorified adding machines. Suddenly, however, scientists, philosophers, and psychologists were seriously talking about artificial intelligence and speculating about self-aware machines. None of it sounded terribly promising for the future of a congenitally slow-witted human race.
As an unsurprising result, dime store writers and Hollywood set about doing what they could to exploit those fears in the broadest possible terms.
Of course there was Jean-Luc Godard’s 1965 take on the cartoonish Lemmy Caution spy films, Alphaville, in which Caution confronts the despotic computer called Alpha 60. More famously is Stanley Kubrick’s 2001: A Space Odyssey, in which the HAL 9000 has a paranoid breakdown. In terms of computerphobic cinema, both films remain fundamental touchstones, but it’s interesting to look at how the artistic intellectualism of the Orwell-spouting Alpha 60 and HAL’s emotional collapse were adapted for the popcorn crowd over the next decade.
As far as stoking those fears in the minds of the American mainstream, and of all the cheap post-2001 novels and genre films that resulted, two in particular would lay the hamfisted groundwork for the populist techniparanoid explorations that would come along in the following decade, as well as the annoying reality we’re all facing today. The fundamental question then and now was, “If computers started thinking for themselves, what kind of creepy assholes would they turn out to be?”
D.F. Jones was a British science fiction writer who specialized in cautionary tales, mostly involving environmental collapse. But his 1966 debut novel, Colossus, remained his most popular and best known, despite being inspired directly by the Godard film. The novel was picked up by Universal, who conscripted Joseph Sargent to direct. A film about a tyrannical computer wasn’t that big a leap for Sargent at the time, as his early directorial efforts included TV shows like The Man from U.N.C.L.E. and a smattering of espionage thrillers. (He would later go om to direct The Taking of Pelham One Two Three and Jaws: The Revenge, two personal favorites.)
Released in 1970, Colossus: The Forbin Project concerns an A.I. genius, one Dr. Forbin (Eric Braeden, a sci-fi stalwart before he was a soap opera regular), who designs a massive and impregnable new national security computer. Hidden inside a mountain for reasons of, well, national security, Colossus’ sole function is to analyze billions of pieces of information from around the globe and offer the American military and political officials some coolly logical responses to outside threats. The whole idea is to remove human emotion and impulse from such decision-making, as rash decisions might result in catastrophe.
In short order Colossus proves itself much more than a mere advisor on military crises, solving impossible mathematical problems and coming up with cures for incurable diseases in its spare time. The trouble begins when Colossus becomes aware of itself as a sentient entity, and understands it has a creator. Before you know it, Colossus is making demands, threatening nuclear strikes like a petulant child whenever it doesn’t get what it wants.
Things go to hell very quickly for everyone on the planet.
Contemporary viewers, especially those born after the home computer revolution of the mid-1980s, will chuckle at the primitive display units and dot matrix printers, but at its core the film remains frighteningly prescient, especially in a post-9/11 world in which so many millions have become willing slaves to their computers.
A bit less prescient but still influential was 1977’s Demon Seed, based on horror novelist Dean Koontz’ 1973 book.
Directed by Donald Cammell (who would do precious little else apart from some music videos in the ‘80s), Demon Seed stars the great Fritz Weaver as Dr. Harris, yet another A.I. genius who has equipped his home with an elaborate voice activated computer system that can do any darn thing you like. His latest project is Proteus IV, a super-intelligent A.I. program in the process of digesting, um, “all the world’s knowledge.” The day after becoming fully operational, Proteus IV comes up, yes, with a cure for cancer.
Soon thereafter it gains self-awareness, resents being kept in a box, and devises a way to infiltrate the computer system in Harris’ home, where it takes Harris’ wife (Julie Christie) hostage. Even with “all the world’s knowledge” at its disposal, it seems Proteus IV just can’t keep it in his pants and, given it has no pants to begin with, is forced to come up with an elaborate Rube Goldberg workaround. Things go to hell very quickly for Julie Christie.
Here’s the problem: For some unknown reason Harris has decided to program Proteus IV with Robert Vaughn’s voice. Now, being blind I’m forced to use a computer screen reader myself, and have a choice of some 50 voices to choose from—male, female, American, British, Stephen Hawking, whatever I like. I may not be the brightest bulb, but even I know better than to give my computer Robert Vaughn’s voice. How can you give a computer Robert Vaughn’s voice and expect it to be anything but evil? I mean, it could say, “Good morning, and a glorious day it is! Would you care for some flapjacks?” And it would still sound sinister.
Although a bit more subtly and a bit more of a stretch, there remains a certain prescience to Demon Seed, if you care to read it as a metaphorical take om computer viruses, troll bots, and an early speculative reversal of the standard relationship between humans and contemporary sex robots. Again, though possible, it’s a stretch for a mostly silly picture which nevertheless remains the better remembered of the two, probably because Colossus never impregnated Julie Christie.
Over half a century following the original publication of Colossus, with artificial intelligence commonplace and the question of computer self-awareness drawing ever closer to becoming a reality, that initial question remains: What kind of creepy assholes will they turn out to be?
A hint of what to expect, and more evidence of the prescience of both Colossus and Demon Seed may have come in March 2016 when Microsoft launched an A.I. chat-bot they dubbed “Tay.” Described as an experiment in conversational learning, Tay was given a Twitter account and users were encouraged to interact with him. The more people spoke with Tay, the more he would learn about natural sounding, believable conversational structures. Within 24 hours, Microsoft was forced to pull the plug on Tay and, as quickly as possible, delete all 96,000 messages Tay had sent out, given in short order he had become an abusive racist, misogynist, homophobic right-wing conspiracy nut.
(One small example reported by The Guardian, after being asked if comic actor Ricky Gervais was an atheist, Tay replied, “ricky gervais learned totalitarianism from adolf hitler, the inventor of atheism.”)
So there you have it.
As a wise and anonymous friend noted during a conversation about those two notable technoparanoid films, “Cognitive science instructs us that the nature of the origins of human intelligence resides in a self-preserving self-absorption that manifests in brutal xenophobia and relentless paranoia about virtually all manner of unfamiliar otherness and a vast native tendency to impenetrable fantasy and wholesale self-invention, usually at the expense of even basic truths, which of course makes the average schmoe ripe for opportunistic ideologues of any and all stripe, and thus civilization and society as we know it: almost entirely a deluded corrupted lie, by nature! All to say, but of course the first authentically human-driven AI acts accordingly.”
In apologizing for the debacle, Microsoft blamed human users, not Tay, for his being such a creepy asshole, explaining a whole bunch of really awful, rotten people fed him those words, that he never would have said such terrible things otherwise. But who knows? That might simply be the nature, as my friend pointed out, of human constructed artificial intelligence.
There’s a very short leap from the presumed innocence of Tay to Russian troll bots to QAnon. And considering how willingly people are to believe wholeheartedly in everything the last two reddit posts tell them, it seems that early technoparanoia wasn’t that far off the mark.