Dr. Fill

Many years ago, my mother successfully negotiated the NYTimes crossword puzzles on a regular basis. Seems Dr. Fill does the same thing as seen by the clip above. Seems AI has moved in on yet again another human endeavor without hesitation.

Nearly 1,300 people spent this past weekend racing to fill little boxes inside larger boxes, ever mindful of spelling, trivia, wordplay, and a ticking clock. They were competitors—newcomers, ardent hobbyists, and elite speed solvers—in the American Crossword Puzzle Tournament, the pastime’s most prestigious competition. And most of them got creamed by some software.

The annual event, normally set in a packed hotel ballroom with solvers separated by yellow dividers, was virtual this year, pencils swapped for keyboards. After millions of little boxes had been filled, a computer program topped the leaderboard for the first time.

Dr. Fill is the algorithmic creation of Matt Ginsberg, an Oxford-trained astrophysicist—and computer scientist, stunt pilot, bridge player, novelist, and magician—who lives in Oregon. When he began the project a decade ago, his motivation was simple: “I sucked at crosswords, and it just pissed me off.” Ginsberg hoped one day to walk into the tournament hall, wave his laptop above his head, and show the humans who’s boss. Now, if only virtually, he has.

Photo illustration by Slate. Photo by BrianAJackson/iStock/Getty Images Plus.

Full On ...

Photograph/Getty Images

Well, the military is full on in creating Skynet now that autonomous control of armed bots is in play as per a Darpa exercise when AI became the untethered entity running swarms of cheap and potentially lethal tech beyond direct control of humans, an issue BRT has talked about for quite some time.

Helena Bonham Carter as Skynet, under the guise of the late Dr. Serena Kogan on a monitor, with Sam Worthington as Marcus Wright.

LAST AUGUST, SEVERAL dozen military drones and tanklike robots took to the skies and roads 40 miles south of Seattle. Their mission: Find terrorists suspected of hiding among several buildings.

So many robots were involved in the operation that no human operator could keep a close eye on all of them. So they were given instructions to find—and eliminate—enemy combatants when necessary.

The mission was just an exercise, organized by the Defense Advanced Research Projects Agency, a blue-sky research division of the Pentagon; the robots were armed with nothing more lethal than radio transmitters designed to simulate interactions with both friendly and enemy robots.

Just a matter of time as to when they become armed.

The drill was one of several conducted last summer to test how artificial intelligence could help expand the use of automation in military systems, including in scenarios that are too complex and fast-moving for humans to make every critical decision. The demonstrations also reflect a subtle shift in the Pentagon’s thinking about autonomous weapons, as it becomes clearer that machines can outperform humans at parsing complex situations or operating at high speed.

Any questions?

4 takes on AI ...

Dina Litovsky/Redux

4 takes on AI shows just how powerful this tech has become in just a few years, a fact showing just how open ended this software truly is. To whit.

In the NYTimes piece titled Why Trump Still Has Millions of Americans in His Grip, the article starts with why Trump keeps his voters enthralled but quickly segues to AI and what it will mean to work in America.

By the 1970s, many white Americans — who had taken their own centrality for granted — felt that they were being shouldered aside, left to face alone the brunt of the long process of deindustrialization: a cluster of adverse economic trends including the decline in manufacturing employment, the erosion of wages by foreign competition and the implosion of trade unionism.

Galston has grasped a genuine phenomenon. But white men are not the only victims of deindustrialization. We are now entering upon an era in which vast swaths of the population are potentially vulnerable to the threat — or promise — of a Fourth Industrial Revolution.

This revolution is driven by unprecedented levels of technological innovation as artificial intelligence joins forces with automation and takes aim not only at employment in what remains of the nation’s manufacturing heartland, but also increasingly at the white-collar managerial and professional occupational structure.

Daron Acemoglu, an economist at M.I.T., described in an email the most likely trends as companies increasingly adopt A.I. technologies.

A.I. is in its infancy. It can be used for many things, some of them very complementary to humans. But right now it is going more and more in the direction of displacing humans, like a classic automation technology. Put differently, the current business model of leading tech companies is pushing A.I. in a predominantly automation direction.

As a result, Acemoglu continued, “we are at a tipping point, and we are likely to see much more of the same types of disruptions we have seen over the last decades.”

Take 2. The dark side ...

Courtesy of AI Dungeon

In a popular game, when AI is brought into the equation, the dark side emerges ...

IN DECEMBER 2019, Utah startup Latitude launched a pioneering online game called AI Dungeon that demonstrated a new form of human-machine collaboration. The company used text-generation technology from artificial intelligence company OpenAI to create a choose-your-own adventure game inspired by Dungeons & Dragons. When a player typed out the action or dialog they wanted their character to perform, algorithms would craft the next phase of their personalized, unpredictable adventure.

Last summer, OpenAI gave Latitude early access to a more powerful, commercial version of its technology. In marketing materials, OpenAI touted AI Dungeon as an example of the commercial and creative potential of writing algorithms.

Then, last month, OpenAI says, it discovered AI Dungeon also showed a dark side to human-AI collaboration. A new monitoring system revealed that some players were typing words that caused the game to generate stories depicting sexual encounters involving children. OpenAI asked Latitude to take immediate action. "Content moderation decisions are difficult in some cases, but not this one,” OpenAI CEO Sam Altman said in a statement. “This is not the future for AI that any of us want."

Take 3, Aggression 2Step ...

Google’s AI got aggressive over apple picking.

When resources get scarce, AI bots become all too human.

In a study published online (but not yet in a peer-reviewed journal), Deep Mind researchers had AI “agents” compete against each other in 40 million rounds of a fruit-gathering computer game. In the game, each agent had to collect as many apples as possible. They also could temporarily knock an opponent out of the game by hitting them with a laser beam.

When apples were abundant, the two agents were happy to collect their fruit without targeting each other. But in more scarce scenarios with fewer apples around, the agents became more aggressive. The researchers also found that the greater the “cognitive capacity” of the agent, the more frequently they attacked their opponent. This makes sense, as in this scenario attacking an opponent is more complex behavior and so requires greater intelligence.

Take 4, Dreaming of Shakespeare ...

Illustration by Charlotte Fu

Writing is hard. One has to practice at it for a long time before it becomes instinctive, something this writer has learned about the hard way as practice doesn't make perfect, it makes better.

Sudowrite uses, as its base, GPT-3, the latest version of a deep-learning neural network that can auto-generate text. The organization that created GPT-3, OpenAI, was founded as a nonprofit with a mission “to advance digital intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return.” In July of 2019, Microsoft invested a billion dollars, which allowed OpenAI to create a supercomputer with two hundred and eighty-five thousand C.P.U. cores, ten thousand G.P.U.s, and four hundred gigabits per second of network connectivity per server. Microsoft claims that it ranks in the top five supercomputers in the world, processing more than twenty-three thousand teraflops per second. The power of the supercomputer has been transformative. GPT-2, which John Seabrook took for a test drive in 2019, asking it to write an article for The New Yorker, had 1.5 billion parameters. GPT-3 has a hundred and seventy-five billion. (A parameter is a configuration variable required to make a prediction in machine learning.) Brute computational power is why Sudowrite can write like Kafka.

GPT-3 hints at a world in which machines can generate language. The consequences are vertiginous. To spend ten minutes with Sudowrite is to recognize that the undergraduate essay, the basic pedagogical mode of all humanities, will soon be under severe pressure. Take an A paper, change a few words in the first paragraph, push buttons three times, and you have an essay that fits the assignment. Whatever field you are in, if it uses language, it is about to be transformed. The changes that are coming are fundamental to every method of speaking and writing that presently exists.

Any questions?

Spaghettification

No words needed.

The awesome spectacle of a black hole ripping a star to shreds can be seen in this striking new visualization from the Deutsches Elektronen-Synchrotron (DESY), a particle accelerator lab in Hamburg, Germany. 

Such events are known as stellar tidal disruptors, and they are fairly rare, occurring just once every 10,000 years in a typical galaxy, according to NASA. Stars are typically flung toward a ravenous black hole after interacting gravitationally with another star or massive object, only to become stretched and devoured should they come too close to the black hole's maw in a process called spaghettification. 

Intensity plus.

Echoes of childhood

Playing Children, by Song dynasty Chinese artist Su Hanchen, c. 1150 AD

This wonderful  picture of childhood indirectly connects to a story by Garrison Keillor reminding us of what children have lost with the onslaught of tech and the negative impact it's had on the art of growing up. From this writer's perspective, I wouldn't trade my childhood with the one today's kids are experiencing in any way, shape or fashion.

I’m a happy man. I had a happy frugal childhood, riding my bike around the countryside back before cellphones and apps that parents could track you with on a laptop, but I avoid talking about happiness because I have young leftist friends who, if I admit to being happy, say, “Well, that’s very nice for you but not everyone is as privileged as you were.”

Privilege was not what made me happy. Dad worked for the post office, we were six kids, so though we weren’t impoverished, we could see it from there. No, it was the bike and freedom and the truck farmers who’d pay a kid to hoe corn and pick strawberries and I’d take my dough to the corner store and buy a couple Pearson’s Salted Nut Rolls and take them down to the Mississippi and eat them and skip stones. It wasn’t about privilege. Why can’t a man talk about happiness without getting a poke in the eye from someone who’s just read a book about systemic inequity and wants you to know it?

The great privilege of my childhood was hoeing and weeding, which is denied to kids now whose moms go to Whole Foods to purchase raspberries from New Zealand and a bag of baby arugula hand-raised in the coastal foothills of Northern California by liberal arts graduates, instead of growing food in a garden and affording their children a useful education.

Weeding is editing and editing is a basic skill desperately needed now that the computer has led to floods, downpours, typhoons of verbiage. Everything is ten times too long. (I had a couple thousand words here about my old editor William Shawn, which I’ve taken out, as you can see.) I read a memoir now and then and I think, “This person never mowed a lawn or weeded a flower bed.” Their book has, in a manner of speaking, a lot of old rusted cars and busted appliances sitting in tall weeds that need to be thrown down in a coulee and the grass mowed.

Any questions?

Not Ironman but close :)

Not quite Ironman but close. :)

A new video uploaded by UK-based jetpack maker Gravity Industries shows a Royal Marine using a jet engine-powered suit to land on a nearby ship after approaching it on a powerboat. In one iteration of the exercise, the soldier lands on the ship and then pulls out a handgun — a preview of how military and police want to use jetpacks for rapid infiltration missions.

The marine can be seen effortlessly taking to the skies, reaching the upper deck of a nearby vessel in record time.

All that's really needed are truly powerful batteries to power the engines. 

From Ice to Flowers

The transition from ice to flowers starts slowly but picks up speed after passing the Vernal Equinox. Spring has finally sprung in CT. :)

Time keeps on ticking ...

 Clocks release disorder, or entropy, as they tick. A new experiment shows that, for one tiny clock, the more accurate the clock, the more disorder it generates. Jatocreate/Pixabay

Entropy is everywhere thanks to the two laws of thermodynamics and the Arrow of Time as disorder always grows as time passes. Thank Sir Arthur Eddington for the  Arrow of Time bit.

Sir Arthur Stanley Eddington

Today’s most advanced clocks keep time with an incredibly precise rhythm. But a new experiment suggests that clocks’ precision comes at a price: entropy.

Entropy, or disorder, is created each time a clock ticks. Now, scientists have measured the entropy generated by a clock that can be run at varying levels of accuracy. The more accurate the clock’s ticks, the more entropy it emitted, physicists report in a paper accepted to Physical Review X.

“If you want a better clock, you have to pay for it,” says physicist Natalia Ares of the University of Oxford.

That's why There Ain't No Thing as a Free Lunch

TANSTAAFL: a plan for a new economic world order. (Pierre Dos Utt, 1949)

To whit yet again ...

Time and entropy are closely intertwined concepts. Entropy is known as the “arrow of time,” because entropy tends to grow as time passes — the universe seems to consistently move from lower entropy to higher entropy (SN: 7/10/15). This march toward increasing entropy explains why some processes can proceed forward in time but not in reverse: It’s easy to mix cream into coffee but exceedingly difficult to separate it again. Machines also increase disorder as they operate, for example by giving off heat that boosts the entropy of their surroundings. That means even a standard, battery-powered clock produces entropy as it ticks.

Reality bites.

Relentless ...

 

Relentless says it all.  

Commencement Day

In this concept illustration provided by NASA, NASA's Ingenuity Mars Helicopter stands on the Red Planet's surface. NASA/Getty Images

Ginny's graduated. No longer a test but rather a partner to Percy as the little chopper has passed every test as a viable scout to a rover. Tech at it's best defines just how well NASA designed this system for an economical 85 million. Awesome says it all.

"Ingenuity's transition from conducting a technology demonstration to an operations demonstration brings with it a new flight envelope," NASA said.

"There will be more precision maneuvering, greater use of its aerial-observation capabilities, and more risk overall," it added.

The additional testing will be conducted over the next few months, ending no later than August.

While the first helicopter flight went just 10 feet above the ground, each voyage has drastically increased in distance. On the fourth flight, the helicopter rose 16 feet in the air and made a round trip of 872 feet across the planet.

Details ...

NASA’s Ingenuity Mars Helicopter (above center to the right) is viewed by one of the hazard cameras aboard the Perseverance rover during the helicopter’s fourth flight on April 30, 2021. Credit: NASA/JPL-Caltech.

Ingenuity successfully completed its fourth flight today, and we couldn’t be happier. The helicopter took off at 10:49 a.m. EDT (7:49 a.m. PDT, or 12:33 local Mars time), climbing to an altitude of 16 feet (5 meters) before flying south approximately 436 feet (133 meters) and then back, for an 872-foot (266-meter) round trip. In total, we were in the air for 117 seconds. That’s another set of records for the helicopter, even compared to the spectacular third flight.

We also managed to capture lots of images during the flight with the color camera and with Ingenuity’s black-and-white navigation camera, which tracks surface features as it flies. Images from that navigation camera are typically used by Ingenuity’s flight controller and then thrown away unless we specifically tell the helicopter to store them for later use. During this flight, we saved even more images than we did on our previous flights: about 60 total during the last 164 feet (50 meters) before the helicopter returned to its landing site.

Commencement Day, the day marking the start of a new chapter certainly applies to Ginny, right? :)