Cyberpunk Review » Robopocalypse: Fact or Science Fiction?

February 26, 2012

Robopocalypse: Fact or Science Fiction?

Source: Life’s Little Mysteries

The scenario of a mechanized hostile takeover of humanity is a popular theme in sci-fi, but just how plausible is it? LLM’s Adam Hadhazy plays mythbuster.

Hollywood Hype. The nightmare of robots usurping humanity has become a staple of science fiction. But have you ever thought if such a scene is possible? Life’s Little Mysteries, the website dedicated to answering some of the more unusual tech questions, gives their opinion about the possibility of “robopocalypse.”

Even the experts seem divided. They believe man and machine will get along, but that relation can turn sour under the right conditions.

“The technology already exists to build a system that will destroy the whole world, intentionally or unintentionally, if it just detects the right conditions,” said Shlomo Zilberstein, a professor of computer science at the University of Massachusetts.


Military (Un)Intelligence. One scenario can be summed up in one word: Skynet. We have the technology to create it, so why not?


Currently, Predator drones in the Middle East have been getting more and more autonomy, more ability to make its own decision to attack a target. Even so, live humans still monitor its operation and can override the drone if needed. When humans tried to shut down Skynet, even the “allies” were determined to be “enemies” and let the nukes fly. The game plan to keep that from happening is to limit what weapons it has access to, and to limit its functionality to specific situations.

Small Eyeborg

“All the systems we’re likely to build in the-near future will have specific abilities,” Zilberstein said. “They will be able to monitor a region and maybe shoot, but they will not replace a [human] general.”

That wouldn’t preclude the possibility of a robotic arms race, leading to both sides loosing control of their machines…


Revolution through evolution. Another scenario of mechanized takeover is not as violent as nuclear war; Humanity simply replaces itself part-by-part, shedding the meat in favor of metal. There would be some resistors (sic), but they’ll be allowed to die out by themselves.

Of course, someone… or something… needs to build those robots. With robots building cars, planes, etc., it won’t be to hard to imagine that robots can build robots, if they’re not doing it already. But when robots not only build robots, but run the whole factory, and possibly the whole infrastructure that humans also rely on, things can get real sticky real fast.


Busted, Plausible, or Confirmed? It would seem that humanity is teetering on the edge of robopocalypse, yet it is something that is easily avoidable:

Overall, a bit of wisdom would prevent humankind from falling into the traps dreamed up by Hollywood screenwriters. But the profit motive at companies has certainly engendered more automation, and the Cold War’s predication on the threat of mutually assured destruction points out that rationality does not always win.

LMM gives a score of 2 out of 4 “rocketboys.” If it was MythBusters, this would be called “Plausible.”

Military leaders and corporations probably will not be so stupid as to add high levels of programmed autonomy to catastrophically strong weapon systems and critical industrial sectors.

Given the levels of stupidity that military and corporations like to show, I’d say this is more than plausible.

This post has been filed under Internet Find by Mr. Roboto.


February 27, 2012

RedFalcon696 said:

Great article and discussion, but this simply won’t happen. The only reason Skynet took control in the original Terminator (1984) movie was because of it’s processor design and poor system parameters. Skynet and Skynet-based processors have a learning ability, aka, a ZISC processor.

Today’s current computer systems are almost exclusive to CISC processors such as Intel and AMD, and RISC processors such as ARM, POWER, PowerPC, Alpha, etc.

ZISC processors do in fact exist, but their applications are limited, as are the few programs they can work with.

ZISC processors have a true learning ability, but not necessarily like Skynet has. A true ZISC processor has the ability to work on specific tasks better and more efficiently over time, such as calling upon a certain database query. The more this task is accomplished and executed, the the more the ZISC processor “learns” or adapts itself to handle that function more efficiently.

However, RISC and CISC processors have pre-build instructions and no ability to “anticipate” or “learn” new functions and can only “learn” through the use of bulky and inefficient software.

I do find it funny that Skynet’s total processing power equals 90TFlops, where as today’s Jaguar supercomputer is capable of 1400TFlops, quite the difference.

So why isn’t Jaguar taking command of everything it reaches out to and usurps humans? Because it utilizes CISC-based AMD processors, and will soon use GPUs for calculations, but still, those GPUs will require the use of CISC-based processors, or RISC if the code is recompiled and heavily modified.

In essence, Jaguar and every other supercomputer has limited abilities to learn and adapt, and thus without software or manual coding support, will not and cannot break it’s simplistic parameters like Skynet did and will not takeover anything other than what it is simply programmed for.

Now, if we had ZISC-based processors on the rise, then we could be worried.
Until then, this is just a fantasy.

February 29, 2012

Laughing Man said:

The evil computers in science fiction are completely different from real life computers, real life computers are programmed by humans, and don’t suddenly come alive for no reason and turn evil, something like this will never happen unless humans program computers like this on purpose.
The entire concept is pretty ridiculous, it makes for fun science fiction (I love HAL, SHODAN, GLaDOS, and others like them just as much as anybody else.) but in real life it only serves to give people what I call “Computerphobia”

Also this
“Given the levels of stupidity that military and corporations like to show, I’d say this is more than plausible.”

Yeah, but Web Admins never make any mistakes or do anything stupid right? obviously the people on sites like this are infinitely smarter than the BIG BAD military, or those horrible horrible corporations (Without whom we would literally have no internet, computers, or technology whatsoever.) but I digress.
Although I haven’t seen much in the “levels of stupidity” the military or any corporations have shown besides, more of what is plastered throughout science fiction, but considering that most Cyberpunk series are mostly pro communism (Like the web master apparently.) that isn’t surprising.

March 1, 2012

defwheezer said:

Maybe the whole Robopocalypse scenario is just the natural tendency of humans to be dark and violent. Could the future be as mundane as lazy and jealous robot workers nagging their jacked-fat meat masters- driving them to even more extreme neuro pharmacological retreat?


Robots have already enslaved us, writ- the automated trading systems that harvest vast sums of capital with little overhead and even sparser contribution to society as a whole. The economic collapse of 2008 was just another opportunity for those algorithms to reap profit. Soul less machines controlled by soul less humans; is it really the machines who are the real danger in this equation…?

bob said:

I don’t think it will happen. These scenarios often involve machines being given sentience or large degrees of autonomy for no real reason. The Animatrix had sentient construction robots. . .why? Why would you need them to have any higher thought than building what you tell them to build? It would be nice to study them in a lab setting, but for widespread use it’s just pointless.

The same goes for the famous evil spaceship AIs and stuff like skynet. Though with Skynet, putting all of your defenses on a single system just seems like a terrible idea by itself.

March 10, 2012

wizard said:

Doing this kind of stuff virtually is the first step to doing it in real life.

March 31, 2012

break_the_bricks said:

If this is something you’re interested in, two incredibly good short stories to read are ‘…That Thou Art Mindful of Him’ and ‘The Bicentennial Man’ by Isaac Asimov. Both of these stories are about how robots (computers) could end up taking over human civilisation, but in vastly different ways. Both of these stories are in ‘The Complete Robot’.

[…] Cyberpunk Review jQuery.ajax({type:”GET”,async:false,url:””,data:”cache_post_id=1119&cache_view_type=robot&cache_output_type=content”,cache:false}); […]

Leave a comment

~All Related Entries Related This~


Made with WordPress and the Semiologic CMS | Design by Mesoconcepts