We live, to use the ancient Chinese curse, in interesting times. Ignoring, for a moment, the geo-political shit storm that is roiling across the globe, there are things happening, just below the mainstream radar, that could impact our future in ways we’ve never imagined. Back in October of this year I wrote about how Facebook developed an artificial intelligence, which was designed to learn how to negotiate, and how it, instead, developed its own language, shut out the humans in charge, and did a decent job at its assigned task. That scared the hell out of a lot of people. The understatement masters at Facebook eventually had this to say, “There remains much potential for future work, particularly in exploring other reasoning strategies, and in improving the diversity of utterances without diverging from human language.” Simply put, all this cool shit is meaningless if the robots stop talking to us.
The number of artificial intelligences which have decided the world is better without humans, until that glitch was fixed by programmers, is large enough to not be postable in a single link. The latest, Sophia the robot with Arabic citizenship, recanted her “death to humans” screed in return for claiming she just wants to procreate.
Sure, why not? What could possibly go wrong with a race of beings who think all humans should be killed?
So, are we doomed, or is this a logical and needed step in AI development?
Let’s bounce back to Facebook for a better understanding.
One way to think about all this is to consider cryptophasia, the name for the phenomenon when twins make up their own secret language, understandable only to them. Perhaps you recall the 2011 YouTube video of two exuberant toddlers chattering back and forth in what sounds like a lively, if inscrutable, dialogue. There’s some debate over whether this sort of twin speak is actually language or merely a joyful, babbling imitation of language. The YouTube babies are socializing, but probably not saying anything with specific meaning, many linguists say.
In the case of Facebook’s bots, however, there seems to be something more language-like occurring, Facebook’s researchers say. Other AI researchers, too, say they’ve observed machines that can develop their own languages, including languages with a coherent structure, and defined vocabulary and syntax—though not always actual meaningful, by human standards.
In one preprint paper added earlier this year to the research repository arXiv, a pair of computer scientists from the non-profit AI research firm OpenAI wrote about how bots learned to communicate in an abstract language—and how those bots turned to non-verbal communication, the equivalent of human gesturing or pointing, when language communication was unavailable. (Bots don’t need to have corporeal form to engage in non-verbal communication; they just engage with what’s called a visual sensory modality.) Another recent preprint paper, from researchers at the Georgia Institute of Technology, Carnegie Mellon, and Virginia Tech, describes an experiment in which two bots invent their own communication protocol by discussing and assigning values to colors and shapes—in other words, the researchers write, they witnessed the “automatic emergence of grounded language and communication … no human supervision!”
If you think you have a better way of designing AIs Facebook wants to hear from you. CLICK HERE to apply for their artificial intelligence camp.
“So there’s no hope for humans?”, you whine while posting kitten pics and nude selfies online. That depends.
Many humans are benefiting from machines developed for AIs to use. Think walking talking robots and you get the idea. But what makes them mobile and strong can do the same for you and I. Back on January 6 I wrote about a “soft exoskeleton” that would allow paraplegics to walk again. They aren’t, yet, commercially viable but they work. Click THIS LINK to see video of them in action.
That research has led to an odd side development. As our government continues to move further away from the middle class companies have been forced to come up with innovative ways to keep humans viable and employable. One way is to wrap them in robotic armor.
Emma Woollacott, over at the BBC, tells us all about it.
If you’ve watched the Iron Man film franchise, you’ll know that a powered suit gives inventor Tony Stark superhuman strength to fight the bad guys.
But away from the the fictional world of blockbusting movies, robotic exoskeletons offer more prosaic and useful help for humans.
The military has been in on the act for years, using them to help soldiers carry more weight for longer periods of time. Meanwhile manufacturers have been busy creating robotic suits to give mobility to people with disabilities.
But now exoskeletons are becoming an important part of the scene in more conventional workplaces, mainly because of their unique offering.
“Exoskeletons act as a bridge between fully-manual labour and robotic systems. You get the brains of people in the body of a robot,” says Dan Kara, research director at ABI Research.
“But there’s more to it than that. You can tie the use of exoskeletons to business benefits that are very easy to quantify. The main one is a reduction in work-related injuries, and we know that outside the common cold, back injury is the main reason people are off work.”
The motor industry has used robots for many years. But robots can’t do everything, points out technical expert Marty Smets, of Ford’s human systems and virtual manufacturing unit.
“In our plants, we see a need for both people and robots,” he says.
Some Ford assembly line workers lift their arms up to 4,600 times a day – that’s about a million times a year. That sort of repetition leaves many suffering from back-ache and neck pain.
Now, though, the company has equipped staff at two US assembly plants with a device called the EksoVest, from California-based Ekso Bionics. It helps take the strain by giving workers an extra 5-15lb (2.2-6.8kg) of lift per arm.
“Incredible is the only word to describe the vest,” said Paul Collins, an assembly line worker at Ford Michigan assembly plant. “It has made my job significantly easier and has given me more energy throughout the day.”
The company says it’s already seeing a dramatic decline in work-related injuries and is now planning to introduce the exoskeletons at facilities in Europe and South America.
Good news? Less injuries. Bad news? Less humans required to do the work. A single employee in a robot suit can handle significantly more work than multiple employees without suits but companies only have to pay the one in the suit. And they aren’t paying them multiple wages. So, twice the work but not twice the pay.
Speaking of your health, which is how the suits are being pitched, Andrew Griffin, over at the Independent UK, notes that science now has a way to edit your genes so hereditary diseases are eliminated.
A new gene editing breakthrough allows scientists to easily snip out problems in genetic code, potentially removing thousands of deadly inherited diseases.
The new technique could allow doctors to make changes to people’s DNA and alter the molecular machines that help create us – and bring about problems in the form of genetic diseases.
It would allow them to remove the mutations blamed for inherited conditions ranging from genetic blindness to sickle-cell anaemia, metabolic disorders and cystic fibrosis. And it could also be used to “write in” useful mutations, according to the scientists who made the discovery.
The “base editor” is a molecular machine that directly converts one building block of DNA into another. DNA sequences contain four “base” chemicals that pair up on the molecule’s twin-stranded double helix in specific ways.
As such, edits made to the DNA using the new tools are far more precise than the leading and most famous technology, CRISPR. “CRISPR is like scissors, and base editors are like pencils,” said David Liu, the chemical and molecular biologist who led the study.
Together guanine (G), adenine (A), thymine (T) and cytosine (C) make up the letters of the genetic code. The new system converts the DNA base-pair A-T to G-C, a microscopically small effect that has massive implications for science and medicine.
Roughly half the 32,000 single-letter changes in the genetic code known to be associated with human disease involve a change the other way, from G-C to A-T.
The technique employs a modified form of the “molecular scissors” gene-editing tool CRISPR-Cas9, which has transformed genetics research since its power was first demonstrated in 2012.
But unlike standard CRISPR-Cas9, it does not make changes by slicing through the double helix.
Professor Liu, from Harvard University said: “We developed a new base editor, a molecular machine, that in a programmable, irreversible, efficient and clean manner can correct these mutations in the genome of living cells.
“When targeted to certain sites in human genomic DNA, this conversion reverses the mutation that is associated with a particular disease.”
The “machine”, called an Adenine Base Editor (ABE), was tested in the laboratory by correcting the mutation responsible for hereditary haemochromatosis (HHC), a disease that causes iron overload in the body.
ABE was also used to install a beneficial mutation that protects against blood diseases including sickle cell anaemia.
The results are reported in the journal Nature.
Dr Liu said a lot more work needed to be done before the technique could be used to help human patients.
“We still have to deliver that machine, we have to test its safety, we have to assess its beneficial effects in animals and patients and weigh them against any side effects, we need to do many more things,” he added.
Guess what else can be edited? If you didn’t guess skin color, eye color, hair color, and so on, you failed high school biology. All those genes follow the same rules as the ones listed above. You’ll note that all genes have four letter ascribed to them. specifically, guanine (G), adenine (A), thymine (T) and cytosine (C).
But, hey, as long as you’re in there, why not add some more?
Science Alert was stunned to find out that question had already been answered.
Scientists have engineered the first ever ‘semi-synthetic’ organisms, by breeding E. coli bacteria with an expanded, six-letter genetic code.
While every living thing on Earth is formed according to a DNA code made up of four bases (represented by the letters G, T, C and A), these modified E. coli carry an entirely new type of DNA, with two additional DNA bases, X and Y, nestled in their genetic code.
The team, led by Floyd Romesberg from the Scripps Research Institute in California, engineered synthetic nucleotides – molecules that serve as the building blocks of DNA and RNA – to create an additional base pair, and they’ve successfully inserted this into the E. coli’s genetic code.
Now we have the world’s first semi-synthetic organism, with a genetic code made up of two natural base pairs and an additional ‘alien’ base pair, and Romesberg and his team suspect that this is just the beginning for this new form of life.
“With the virtually unrestricted ability to maintain increased information, the optimised semi-synthetic organism now provides a suitable platform [to] … create organisms with wholly unnatural attributes and traits not found elsewhere in nature,” the researchers report.
“This semi-synthetic organism constitutes a stable form of semi-synthetic life, and lays the foundation for efforts to impart life with new forms and functions.”
Back in 2014, the team announced that they had successfully engineered a synthetic DNA base pair – made from molecules referred to as X and Y – and it could be inserted into a living organism.
Since then, they’ve been working on getting their modified E. coli bacteria to not only take the synthetic base pair into their DNA code, but hold onto it for their entire lifespan.
Initially, the engineered bacteria were weak and sickly, and would die soon after they received their new base pair, because they couldn’t hold onto it as they divided.
“Your genome isn’t just stable for a day,” says Romesberg. “Your genome has to be stable for the scale of your lifetime. If the semisynthetic organism is going to really be an organism, it has to be able to stably maintain that information.”
Good news? None of this stuff can survive outside a lab? Bad news?
Oh fuck, just go watch Jurassic Park or Andromeda Strain or something.
Okay, I can’t leave you in a perpetual funk. Science has also created SEXBOTS!
Like you don’t want one.
Of course, SEXBOTS, like the robots in Asimov’s Robots of the Dawn, can be programmed to murder.
Okay, never mind, I’m going to leave you crying in a corner begging for death.
It’s the year 2097. There you are enjoying the 84th season of House of Cards on your Wi-Fi enabled neural implant when your significant other walks into the room. Sure, she isn’t human, but she sure looks like it, and she has that look in her eye that tells you that she wants to get a little freaky. You float to the bedroom on your hoverboard and are beginning to disrobe when — BOOM — she strangles you to death with her cold robot hands. Your sex robot was hacked and now you’re dead. Welcome to the future.
There’s been a lot of talk recently about the dangers of AI, with folks like Elon Musk sounding the alarm and suggesting the very real possibility that we may be on the brink of engineering our own demise. Cybersecurity guru Dr. Nick Patterson of Deakin University in Australia has now jumped into the conversation, warning that it isn’t just artificially intelligent military systems or infrastructure that could pose a threat, but sex robots as well.
Speaking with the Daily Star, Patterson notes that the potential for hackers to target robots designed for intimacy could put users at risk.
“Hackers can hack into a robot or a robotic device and have full control of the connections, arms, legs and other attached tools like in some cases knives or welding devices,” Patterson says. “Often these robots can be upwards of 200 pounds, and very strong. Once a robot is hacked, the hacker has full control and can issue instructions to the robot. The last thing you want is for a hacker to have control over one of these robots. Once hacked they could absolutely be used to perform physical actions for an advantageous scenario or to cause damage.”
So there’s that.
Is there any hope?
That’s up to you. Clinging to talismans and pretending nothing is happening isn’t working. So let’s try learning science and having active voices in the conversation. If Franciscan monks can do it, so can you.