MIT – BLDGBLOG

Every Reflection A Leak

[Image: “Two images of the same room, one reconstructed from video footage of a bag of chips within the room (top) and the other photographed directly (bottom),” as described by Scientific American. Images courtesy Jeong Joon Park.]

“Researchers have now found that by filming a brief video clip of a shiny item, they can use the light flashing off it to construct a rough picture of the room around it,” Scientific American reports. “The results are surprisingly accurate, whether the reflections come from a bowl, a cylinder or a crinkly bag of potato chips.”

It comes down to mathematically modeling “what a known object will look like—how light will reflect off it—when it is placed in new surroundings,” such that you can then reconstruct the proper orientation of what it reflects.

There’s a lot more in the original article, but what immediately struck me about this was how this technology could be used for crime or espionage, both.

You send an unsuspecting group of school kids into a target building, carrying highly reflective silver balloons, or you wear a slyly reflective and precisely designed item of clothing into a business meeting: in both cases, a photographer on a roof across the street or hidden in a park nearby snaps away through a telephoto lens. The reflections spilling off in all directions are like a 360º spherical photograph of the building interior—the art on the walls, the position of furniture. The location of a safe.

Think of the Japanese pop star who was tracked by a stalker after he deduced her location by analyzing the reflection in her eye in a selfie. Every mirrored surface becomes a security leak—“Las Meninas” as burglary tool.

[Image: “Las Meninas” (1656) by Diego Velázquez; if my reference to this painting makes absolutely no sense in the present context, it’s because I’m being pretentious and indirectly referring to Michel Foucault’s The Order of Things, where he discusses the painting’s use of internal reflection.]

Of course, you may also recall that sounds can be reconstructed from the vibrations of distant objects: “Researchers at MIT, Microsoft, and Adobe have developed an algorithm that can reconstruct an audio signal by analyzing minute vibrations of objects depicted in video. In one set of experiments, they were able to recover intelligible speech from the vibrations of a potato-chip bag photographed from 15 feet away through soundproof glass… In other experiments, they extracted useful audio signals from videos of aluminum foil, the surface of a glass of water, and even the leaves of a potted plant.”

It’s worth noting here how potato chip bags pop up in each example. Ocean’s 14 will open with a surreptitious potato chip delivery…

In any case, political dissidents, high-value corporate CEOs, and adversarial diplomatic attachés will never be safe again. Just a brief reflection from a cigarette lighter or a piece of silverware, just a tiny ripple of sound across the leaves of an exotic orchid in the center of a dinner table, and someone across the city with a telescope has your bank passcode, the location of your home safe, and a complete 3D map of your building interior, even down to where your security guards are sitting.

[This is only somewhat related, but recall that an engineer at Carnegie Mellon has developed “a long-range iris scanner that can identify someone as they glance at their rear-view mirror” in a moving vehicle, Rob Meyer reported for The Atlantic back in 2015.]

Pleased to meet you. Hope you guess my name.

There was an interesting sequence of otherwise unrelated articles published over the last few days.

Over at Aeon, Murray Shanahan, a professor of “cognitive robotics,” asked: “Beyond humans, what other kinds of minds might be out there? From algorithms to aliens, could humans ever understand minds that are radically unlike our own?” He goes on to discuss, and even graph out, “the space of possible minds.” Briefly, I’m reminded of one of my favorite quotations of all time, from author William S. Burroughs, who, in his book The Ticket That Exploded, described “a vast mineral consciousness near absolute zero thinking in slow formations of crystal,” hidden somewhere inside the surface of the Earth. Try understanding—and conversing with—that.

As an aside, I generally find these sorts of discussions—including, most of all, the Turing Test—to be oddly fixated not on consciousness at all, but specifically on the social mores and recognizable etiquette of a well-educated middle class Western consciousness capable of rational conversation, something that is by no means synonymous even with human self-awareness, let alone with sentience itself. Engaging in conversation with your own coworkers can already be unnervingly impossible, let alone recognizing the potential intelligence of a sea urchin, a virus, a geomagnetic field, or a pulsar. Or, for that matter, a “time crystal.”

In any case, while some of us are contemplating the existence of other types of minds, those other types of minds might simply be trying to rip us off—or so the New York Times suggested in an article called, “As Artificial Intelligence Evolves, So Does Its Criminal Potential.”

In a scenario that sounds like something from Rivka Galchen’s recent book, Atmospheric Disturbances, we’re told to “imagine receiving a phone call from your aging mother seeking your help because she has forgotten her banking password. Except it’s not your mother. The voice on the other end of the phone call just sounds deceptively like her. It is actually a computer-synthesized voice, a tour-de-force of artificial intelligence technology that has been crafted to make it possible for someone to masquerade via the telephone.”

You can read the rest of the article, but there’s something oddly hilarious in the fear that we might finally encounter another form of radically inhuman intelligence—only for it to prank call us, spam us, and con us out of our life savings.

And then it gets worse. According to Quartz reports, researchers at MIT are using Artificial Intelligence “to create pure horror.” “A series of algorithms dubbed the Nightmare Machine is an effort to find the root of horror by generating ghoulish faces, and then relying on user feedback to see which approach makes the freakiest images,” we read.

To be completely honest, the resulting images are disappointing and stupid—a Target Halloween costume aisle is more frightening—but the notion, not that we will encounter an alien intelligence intent on terrifying us, but that we will deliberately create one specifically for this purpose is excellent evidence for anyone wondering how humans have made it this far.

Meshworm

The last few years have seen the rise of “soft robots,” squirming, biomorphic, and highly flexible little machines that can be used to slip through cracks, infiltrate tight spaces, even explore architectural ruins in the wake of earthquakes and warfare.

But soft robots are also getting closer to becoming what are, in effect, mechanically agile medical devices that can “monitor your insides,” in the words of Sangbae Kim, assistant professor of mechanical engineering at MIT, as reprinted by Popular Science, sneaking around inside your body like an earthworm.

The so-called “meshworm” is exactly that: a robotic “worm” made from layered wire mesh that uses “nickel-titanium alloy for muscles.” The application of a high temperature “shortens the wire, tightens the spring’s coil, and squeezes that body segment.” Thus, “when a segment contracts, the one behind it stretches out, and the robot inches forward. The tendon also has muscles attached so the robot can turn left or right.”

The result is the oddly grotesque and somewhat phallic creeping machine you see in the short video, above. The idea is that this could be used for medical diagnosis or vascular surgery.

However, the architectural or broadly spatial uses of this technology are also worth considering, including the potential for monumentally scaled-up versions of the meshworm, capable of assisting human or material transport through the built environment—a kind of peristaltic package-delivery tube that could replace the much-discussed pneumatic tubes of an earlier urban era. Like something out of a David Cronenberg film, the city would have a kind of giant bowel-infrastructure distributing waste material from point to point.

More interestingly, though, this new class of soft robots and meshworms could quickly assume their roles as architectural explorers in their own right, burrowing through collapsed buildings, passing beneath or around doors, even being taken up by the more ambitious burglars and tactical operations teams of the world.

Or, for example, earlier this month in the cave state of Kentucky, the annual “Cave City Hamfest” explored how to bring radio transmission deep underground. This was “accomplished by placing handheld (relay capable) walkie-talkies or relay boxes along a cave passage.” “After the inital debugging phase, we demonstrated the ability to simply walk the cave, until data was lost and then backing up a few feet for a solid link. Then placing a radio on a convenient rock and continuing.” Taking this as our cue, we could simply wire-up a team of meshworms with radio repeaters and send a small, crawling team of spelunking robots far ahead of us into caves where no human body can fit; they would crawl until they lose a signal, move back a few feet to re-establish a secure feed, and then the next one squirms dutifully forward.

You’ve thus built a mobile, semi-autonomous, deep-earth radio network made from repurposed medical devices—equal parts cave-mapping expedition and subterranean pirate radio station—opening up whole new realms of underground exploration (and tactical media).