SHIT! (screencap, annotated by Betabeat)

Does the prospect of Skynet keep you awake at night? Well, now there's something you can do to fight your fears, besides chucking your Sarah Connor Chronicles DVDs. The BBC reports that yesterday, the Campaign to Stop Killer Robots officially launched in London.

The group is exactly what it sounds like. Founded by reps from nine different NGOs, including the Human Rights Watch, International Committee for Robot Arms Control and Nobel Women’s Initiative, the campaign wants a "pre-emptive and comprehensive ban" on fully autonomous weapons.

Building on a report released last fall, leaders are calling for an international treaty, plus national laws as needed.

The group's leader Jody Williams was upbeat about their prospects in an interview with the BBC, predicting that people will "flock" to the cause. She said:

"The public conscience is horrified to learn about this possible advance in weapons systems. People don't want killer robots out there. Normal human beings find it repulsive."

Bear in mind that the U.K. government has said it's got no intention of developing such weapons. Of course, that's what the government would say, isn't it?

Smile, Sarah! (Photo: Terminator Wikia)

Good news for those of you kept awake at night by terrible apocalyptic visions and/or whose name is Sarah Connor: The BBC reports that the University of Cambridge is taking up the study of the existential risks posed by new technologies. Think killer AIs with indestructible robot bodies.

One of organizers is Skype cofounder Jaan Tallinn, along with philosopher Huw Price and astrophysicist Martin Rees. What, the English professors were all busy?

The BBC helpfully adds that, "Fears that machines may take over have been central to the plot of some of the most popular science fiction films" and proceeds to explain the example of Skynet which "gained self-awareness and fought back after first being developed by the US military."

But the Cambridge Project for Existential Risk won't be limiting its sights to scenarios suggested by the Terminator franchise. The founders are also worried about a wide range of technological developments that could "soon pose new, extinction-level risks to our species as a whole." Think everything from massive climate change to nanotechnology.

The project's mission statement argues that "these issues require a great deal more scientific investigation than they presently receive," adding:

The seriousness of these risks is difficult to assess, but that in itself seems a cause for concern, given how much is at stake.

Q.E.D, sir.

The project should open its doors sometime early next year, as the BBC drolly notes, "survival of the human race permitting." That leaves plenty of time for all the Singularitarians to buy plane tickets.

More like ActionHeroBot, right? (Photo: Georgia Tech)

They might be able to make burritos and rescue the drowning, but robots are still lacking in some basic functionality. Namely: The ability to do very much with tools. Those of you who've seen Planet of the Apes and/or ever attempted to jimmy the cap off a beer bottle will surely recognize that this is an important part of our special sauce as a species, and one that our mechanical brethren can't quite yet replicate. Hence, as per the MAKE blog, a team of researchers at Georgia Tech are working on that.

Specifically, they would like to build a robot MacGyver.

As this Georgia Tech announcement points out, we're increasingly deploying robots in dangerous situations and hard-to-get-to places (hello, Mars rover!), but they lack human abilities to interact with their environment. If they lose their keys, they can't root through their purse and find something to pick the lock: 

If today’s most sophisticated robot was trapped in a burning room by a jammed door, it would probably not know how to locate and use objects in the room to climb over any debris, pry open the door, and escape the building.

Those of us on self-appointed Skynet Watch might call that a feature, rather than a bug, but the research team headed up by professor Mike Stilman begs to differ. They've gotten a $900,000 grant from the Office of Naval Research to build a robot that can "use objects in their environments to accomplish high-level tasks."

In other words:

“Our goal is to develop a robot that behaves like MacGyver, the television character from the 1980s who solved complex problems and escaped dangerous situations by using everyday objects and materials he found at hand."

The solution will involve algorithms, naturally, which is exactly how we get Cylons, so thanks in advance, guys.

(Photo: ihasaflavor.com)

No big deal or anything. Don't be alarmed. But Google's secretive computer network simulating the human brain has learned to recognize cats. On its own. With no hints from its mortal creators. AH-HA! So this is how Skynet will begin.

The project, of course, comes out of Google's clandestine X Labs, the same futuristic outfit responsible for augmented reality on your face and cars that drive themselves. Geeze, they just can't make humans obsolete fast enough, can they?

Operation First Lets Start with the Kitties has been several years in the making, reports John Markoff at The New York Times. Scientists built one of the largest "neural networks," mimicking the human brain, by connecting 16,000 computer processors with more than one billion connections. Then, they proceeded to feed it random thumbnails of images from 10 million YouTube videos.

The results in the cat test were about twice as accurate as prior efforts. But what's perhaps more remarkable, is that unlike most commercial vision technology, which has humans "supervise" (scare quotes his!) the process and label features, GOOG gave its brain no help.

The Google brain assembled a dreamlike digital image of a cat by employing a hierarchy of memory locations to successively cull out general features after being exposed to millions of images. The scientists said, however, that it appeared they had developed a cybernetic cousin to what takes place in the brain’s visual cortex.

Neuroscientists have discussed the possibility of what they call the “grandmother neuron,” specialized cells in the brain that fire when they are exposed repeatedly or “trained” to recognize a particular face of an individual.

Wait, if cats are the first faces the machine learned to the recognize, does that mean they'll be spared . . . or targeted?

TARGET ACQUIRED.

In the Venn diagram between techies and Terminator 2 fans, we're guessing the overlap is oh, let's say 100.00 percent. But a new iPhone app built by Silicon Alley's Rich Cameron and Haris Amin might be the first time the two have officially merged.

Cyborg Vision, which made its App Store debut for the iPhone and iPad 2 this morning, uses the Face.com API and Facebook to let see the world around you like a cyborg assassin sent from the future. Its best trick is actually recognizing your Facebook friends and returning a red-screen in real time with their Facebook data that would make Skynet proud. Mr. Cameron and Mr. Amin, who both work at the health and fitness app DailyBurn, first developed the concept at the video hack day, where the app took first prize. They then spent weekends getting it Apple-ready.

On the phone with Betabeat last week, Mr. Amin explained how it works: When a user opens Cyborg Vision, the app requests permission to sign in with Facebook, after which it downloads info about your friends into its database. Through the phone's camera function, you scan a friend's picture or, if they're sitting across from you, the actual friend. A request is sent to Facebook to "so it trains against your friends." The image is then run through Face.com's API, and info from the database is pushed out in real time onto your screen.

Mr. Cameron came up with the idea after the two collaborate on a slide-synching app at TechCrunch Disrupt. "I just wanted to do something that was more fun than presentation sharing—something a little less enterprise-y! We were just kind of throwing out ideas, and I threw out the idea that what if we made Terminator-vision, like it can look at somebody and recognize them and it will look like the movie and everything," Mr. Cameron said on the phone. "We love the movies. I obviously didn’t care for the third one, I don’t think anyone does. [Ed. Agreed!] That was the event of my childhood, I feel like. It was the biggest movie of any summer."

The two developers embedded some surprises in the app. "We hid some Easter eggs in the app, which I think you’re the only person who knows about besides me and Haris," said Mr. Cameron. "It recognizes Arnold Schwarzenegger, even if you’re not signed into Facebook. It will recognize certain people, maybe tied to the movies, or technology. It recognizes Steve Jobs."

Neither developer plans on quitting their day jobs, though they figure $0.99 is a fair price to pay for impressing your friends. "We’re viewing it as a parlor trick or this is a fun thing you can pull out on your friends at the bar. At the same time, it feels like the first time where facial recognition has been handed off to the consumer world where—for a dollar—you can get this on your phone and it will actually do facial recognition and identity people. It’s kind of neat," said Mr. Cameron, whose app already has a couple 5-star reviews.

"If we make millions, that’s another story."

Cyborg Vision from Rich Cameron on Vimeo.