News Ticker

[GUEST REVIEW] Ernest Lilley on ‘Robopocalypse’ by Daniel Wilson

Ernest Lilley is the former Sr. Editor/Publisher of SFRevu and former radio Co-Host of Sci-Fi Talk with Tony Tellado. He currently publishes TechRevu (, occasionally writes for science and technology publications, and is Director of Technology for a DC based educational association. He regularly blogs about sf, technology, culture, poetry and soup at “being Ernest” and can be found here on Facebook. He currently lives in the Gernsback Continuum with that classic trope of SF, a red headed heroine.

Scientists have a grand tradition of turning from writing academic papers to science fiction to transmit their manifestos from the ivory tower to the tech-savvy populace…and beyond. Daniel Wilson has been easing into sf with a string of books about the future, including Where’s My Jetpack?, a look at what happened to the future of sf’s golden age, and the very well-received How to Survive the Robot Uprising.

Wilson’s current book makes no attempt to hide its identity as the latter title, rebooted with 100% more characters, plot, and action. Well, maybe with 50% more of each. Despite Steven Spielberg’s excitement and affection for the book’s prospect as his next feature roboblockbuster, Robopocalypse is short on story, as well as being something of a disappointment as a work of hard sf.

The robo-tech is respectable, as one would expect from someone who just earned his Ph.D. from the United States’ most prestigious robotics school, Carnegie Mellon. But the scenario is weak compared to the story’s predecessors, including Hal’s rampage in 2001: A Space Odyssey, Colossus’ blackmail of the human race in Colossus: The Forbin Project and the benevolent cybergun held to our head in Rob Sawyer’s recent Webmind series. None of which are robot stories…but really, neither is this.

Instead, it’s an emergent AI story with robotic peripherals strewn about the landscape, which is admittedly a new wrinkle on an old dog, but it undermines the value of the narrative. As the jacket blurb says, “People should know that at first the enemy looked like everyday stuff; cars, buildings, phones, planes…” and robots, industrial, personal, and military. Elsewhere on the jacket the author is compared to Michael Crichton, specifically for Andromeda Strain, which is apt. Both authors’ mission is to breathe life into a new techno-bogyman and scare the masses.

It’s interesting that the author goes with the AI-takes-over-everything theme, when the real danger will be in malware loaded into smart devices, both as pranks and as genuinely malicious acts. Don’t be be too surprised when your Roomba(tm) starts regurgitating dirt in jihadist verse. Asimov’s three laws are as fanciful as any other device in sf, but they’d be really, really useful about now.

There actually is a robot uprising in the book, but ironically, it’s against the AI that’s subverted all of robodom against humankind.

The novel, such as it is, starts “twenty minutes after the end of the war,” as the narrator, Cormac “Bright Boy” Wallace, tells us (we won, btw). The first chapter telegraphs the setup; an AI named Archos holed up in a cavern created by a remote underground nuclear test in Alaska, took control of all the robots and automated devices in the world and systematically wiped out the bulk of humanity, starting in the urban centers and moving out into the countryside. Rooting around in the ruins, Cormac’s squad finds the “black box” of the war: a collection of multimedia records of the mostly-human actions in the war, carefully preserved by Archos. Bright Boy decides that it’s a story worth telling, even though he’s not keen to relive it, so he sets out to capture it in words.

All in all, Robopocalypse reads more like a collection of short stories than a novel. In a way, the setup reminds me of the intro to Bradbury’s The Illustrated Man, where we look into the tattoos and they become animated stories. But here there’s an overarching story, with a recurring set of characters.

Some of those characters make only a single walk-on, but a core group emerges fairly quickly: Cormac; his brother Jack and their squad; Mathilda Perez, daughter of a Senator who became aware of the robot danger earlier than most; and Lurker, a teenage “phone-phreak (hacker)” that accidentally traces Archos to his lair and then turns his talents from taunting people to protecting humanity. There’s also Takeo Namura, a Japanese robot mechanic that loves the machines deeply, so deeply that they wind up returning the affection. And the Osage Nation, a Native American enclave in Oklahoma, becomes a character in its own right, and the core of the human resistance.

So the story isn’t as fragmented as it might have been, but rather, it’s distributed across a wide geographical space, and the actors assemble into a virtual community, communicating through human social networks and cyberspace, each contributing a piece of the puzzle that allows us to defeat Archos.

Archos makes it clear, in an inconsistent sort of clarity, that he doesn’t want to wipe out humanity. He wants robokind and humankind to live together in peace, without robots as second class citizens, or worse. Though he’s vastly brighter than a human, being the author’s idea of a Vingian singularity intelligence, he never seems to consider manipulating public opinion rather than wiping the bulk of humanity off the face of the planet. Archos, possibly inheriting the trait from its creator, author Wilson, suffers from profound social blindness.

Ultimately, what bothers me about the story isn’t the fragmentation, technology, or tone, all of which are moderately well handled. Nor is it that the author is covering ground that others have pretty well trampled over the last century, essentially retelling the story which introduced the term Robot to us: R.U.R. (Rossum’s Universal Robots), written by Czech author Karel Capek and published in 1920. Stories need to be made fresh for new generations, and made relevant to new times, and that’s clearly what the author is doing here.

What bothers me is the ineptitude of the robot hordes and how easily they’re defeated by the ad hoc army of survivors. When our future robot masters do rise up against us, don’t expect it to be so easy.

11 Comments on [GUEST REVIEW] Ernest Lilley on ‘Robopocalypse’ by Daniel Wilson

  1. Making things too easy on the heroes is the fatal flaw of a lot of recent science fiction. It’s the sort of feel-good, crowd-pleasing moment in movies that makes me groan. The future will be a mess. Dealing with it won’t be easy. There’s a lot of great story material in that.

  2. Great review, Ern. Sounds like Wilson was going for a WORLD WAR Z approach, telling the story through vignettes and diverse points of view assembled by one overarching narrator (in this case, Bright Boy). The difficulty in setting up a story like this is that we’ve seen this story played out in various forms in myriad ways since ENIAC went live ages ago. If an author can’t bring something new, then the story looks and feels like reheated elements of other stories– a “remember this?” travelogue, if you will.

    Hopefully Wilson has a strong follow up in his laptop, ’cause it sounds like Robopocalypse is not a home run.

  3. Dan Kimmel // June 9, 2011 at 8:24 pm //

    Thanks for the head’s up.  The scariest notion is that Spielberg is thinking of turning this into a movie. After the disappointments of “A.I.” and “War of the Worlds” this is exactly the sort of material he should be avoiding.

  4. What bothers me is the ineptitude of the robot hordes and how easily they’re defeated by the ad hoc army of survivors. When our future robot masters do rise up against us, don’t expect it to be so easy.”

    I cannot believe–nay, I will not accept–that Wilson has a doctorate in robotics, because he clearly has no friggin’ clue how commercial robotics will evolve, and he has even less than no friggin’ clue about Artificial Intelligence. Yes, an Artificial General Intelligence like Archos that is unfriendly toward humans due to misunderstanding, incomprehension, or even good old-fashioned belligerence could very well pose an extinction-level threat to human life…but would it act like Archos, taking command of toys and Roombas to destroy the Human Menace? The everyday bots that appear in the early chapters of this worthless clutter of words are meant to be thoroughly-average domestic appliances…so why are they apparently built of military-grade plastics and ceramics and are strong enough to rip a car in half?

    I threw this book away from me in disgust after reading less than fifty pages. Wilson needs to be stripped of his tenure at Carnegie Mellon and sent packing for producing such a juvenile, Luddite wank-job like Robopocalypse.

  5. Interesting post, Ernest, and thank for pointing this out to me.

    In a world where the mantra is “everything’s connected”, infective malware Von Neumann-style is a real problem – and worse still if there is a controlling Mind ready to exploit the virus, although MMS’s _One of Us_ describes self-aware domestic appliances with humour and grace.

    It sounds like my own robot horde, the New Machine Jihad, could pwn Wilson’s Robopocalypse in the space of an afternoon! Prepare…

  6. Wonderful review, Ern.  Thanks for doing it.  There’s definitely an AI summer going on, at least in science fiction:  a lot of us rethinking old ideas about thinking machines.  As Moore’s Law marches on, I guess we’re all getting ready — steeling ourselves, if I may — for their advent. 😉

  7. Speilberg should make Allen Steele’s Hex into a movie.  Check it out if you want some more traditional scifi.


  8. Great to see comments here from two authors whose AI trilogies I really liked, and mentioned in the piece. Granted I may have mentioned the review to them. Still, I strongly recommend both, though they are totally different:


  9. I can’t tell you how delighted I was to see a review that cites Colossus: The Forbin Project. 

    This is a great review–very thoughtfully done. Of course a good review of a not-so-good book is better for the readers than the author. At least you refrained from the appelation “wank job.”



  10. Tam’s Law: newer versions of software will take up 150% more resources, see Windows and Office as canonical examples



  11. @Bob: Considering the hype, I don’t think it’s going to hurt the book all that much. How you feel about it whill probably have some correlation to how much sf you’ve read. Thx, btw.

    @Tam: Tam’s Law – the unholy child of Parkinson’s and Moore’s Laws


Comments are closed.

%d bloggers like this: