cwalshuk - Tumblr blog

cwalshuk · 2 months

Text

Humans are not perfectly vigilant

I'm on tour with my new, nationally bestselling novel The Bezzle! Catch me in BOSTON with Randall "XKCD" Munroe (Apr 11), then PROVIDENCE (Apr 12), and beyond!

Here's a fun AI story: a security researcher noticed that large companies' AI-authored source-code repeatedly referenced a nonexistent library (an AI "hallucination"), so he created a (defanged) malicious library with that name and uploaded it, and thousands of developers automatically downloaded and incorporated it as they compiled the code:

https://www.theregister.com/2024/03/28/ai_bots_hallucinate_software_packages/

These "hallucinations" are a stubbornly persistent feature of large language models, because these models only give the illusion of understanding; in reality, they are just sophisticated forms of autocomplete, drawing on huge databases to make shrewd (but reliably fallible) guesses about which word comes next:

https://dl.acm.org/doi/10.1145/3442188.3445922

Guessing the next word without understanding the meaning of the resulting sentence makes unsupervised LLMs unsuitable for high-stakes tasks. The whole AI bubble is based on convincing investors that one or more of the following is true:

There are low-stakes, high-value tasks that will recoup the massive costs of AI training and operation;

There are high-stakes, high-value tasks that can be made cheaper by adding an AI to a human operator;

Adding more training data to an AI will make it stop hallucinating, so that it can take over high-stakes, high-value tasks without a "human in the loop."

These are dubious propositions. There's a universe of low-stakes, low-value tasks – political disinformation, spam, fraud, academic cheating, nonconsensual porn, dialog for video-game NPCs – but none of them seem likely to generate enough revenue for AI companies to justify the billions spent on models, nor the trillions in valuation attributed to AI companies:

https://locusmag.com/2023/12/commentary-cory-doctorow-what-kind-of-bubble-is-ai/

The proposition that increasing training data will decrease hallucinations is hotly contested among AI practitioners. I confess that I don't know enough about AI to evaluate opposing sides' claims, but even if you stipulate that adding lots of human-generated training data will make the software a better guesser, there's a serious problem. All those low-value, low-stakes applications are flooding the internet with botshit. After all, the one thing AI is unarguably very good at is producing bullshit at scale. As the web becomes an anaerobic lagoon for botshit, the quantum of human-generated "content" in any internet core sample is dwindling to homeopathic levels:

https://pluralistic.net/2024/03/14/inhuman-centipede/#enshittibottification

This means that adding another order of magnitude more training data to AI won't just add massive computational expense – the data will be many orders of magnitude more expensive to acquire, even without factoring in the additional liability arising from new legal theories about scraping:

https://pluralistic.net/2023/09/17/how-to-think-about-scraping/

That leaves us with "humans in the loop" – the idea that an AI's business model is selling software to businesses that will pair it with human operators who will closely scrutinize the code's guesses. There's a version of this that sounds plausible �� the one in which the human operator is in charge, and the AI acts as an eternally vigilant "sanity check" on the human's activities.

For example, my car has a system that notices when I activate my blinker while there's another car in my blind-spot. I'm pretty consistent about checking my blind spot, but I'm also a fallible human and there've been a couple times where the alert saved me from making a potentially dangerous maneuver. As disciplined as I am, I'm also sometimes forgetful about turning off lights, or waking up in time for work, or remembering someone's phone number (or birthday). I like having an automated system that does the robotically perfect trick of never forgetting something important.

There's a name for this in automation circles: a "centaur." I'm the human head, and I've fused with a powerful robot body that supports me, doing things that humans are innately bad at.

That's the good kind of automation, and we all benefit from it. But it only takes a small twist to turn this good automation into a nightmare. I'm speaking here of the reverse-centaur: automation in which the computer is in charge, bossing a human around so it can get its job done. Think of Amazon warehouse workers, who wear haptic bracelets and are continuously observed by AI cameras as autonomous shelves shuttle in front of them and demand that they pick and pack items at a pace that destroys their bodies and drives them mad:

https://pluralistic.net/2022/04/17/revenge-of-the-chickenized-reverse-centaurs/

Automation centaurs are great: they relieve humans of drudgework and let them focus on the creative and satisfying parts of their jobs. That's how AI-assisted coding is pitched: rather than looking up tricky syntax and other tedious programming tasks, an AI "co-pilot" is billed as freeing up its human "pilot" to focus on the creative puzzle-solving that makes coding so satisfying.

But an hallucinating AI is a terrible co-pilot. It's just good enough to get the job done much of the time, but it also sneakily inserts booby-traps that are statistically guaranteed to look as plausible as the good code (that's what a next-word-guessing program does: guesses the statistically most likely word).

This turns AI-"assisted" coders into reverse centaurs. The AI can churn out code at superhuman speed, and you, the human in the loop, must maintain perfect vigilance and attention as you review that code, spotting the cleverly disguised hooks for malicious code that the AI can't be prevented from inserting into its code. As "Lena" writes, "code review [is] difficult relative to writing new code":

https://twitter.com/qntm/status/1773779967521780169

Why is that? "Passively reading someone else's code just doesn't engage my brain in the same way. It's harder to do properly":

https://twitter.com/qntm/status/1773780355708764665

There's a name for this phenomenon: "automation blindness." Humans are just not equipped for eternal vigilance. We get good at spotting patterns that occur frequently – so good that we miss the anomalies. That's why TSA agents are so good at spotting harmless shampoo bottles on X-rays, even as they miss nearly every gun and bomb that a red team smuggles through their checkpoints:

https://pluralistic.net/2023/08/23/automation-blindness/#humans-in-the-loop

"Lena"'s thread points out that this is as true for AI-assisted driving as it is for AI-assisted coding: "self-driving cars replace the experience of driving with the experience of being a driving instructor":

https://twitter.com/qntm/status/1773841546753831283

In other words, they turn you into a reverse-centaur. Whereas my blind-spot double-checking robot allows me to make maneuvers at human speed and points out the things I've missed, a "supervised" self-driving car makes maneuvers at a computer's frantic pace, and demands that its human supervisor tirelessly and perfectly assesses each of those maneuvers. No wonder Cruise's murderous "self-driving" taxis replaced each low-waged driver with 1.5 high-waged technical robot supervisors:

https://pluralistic.net/2024/01/11/robots-stole-my-jerb/#computer-says-no

AI radiology programs are said to be able to spot cancerous masses that human radiologists miss. A centaur-based AI-assisted radiology program would keep the same number of radiologists in the field, but they would get less done: every time they assessed an X-ray, the AI would give them a second opinion. If the human and the AI disagreed, the human would go back and re-assess the X-ray. We'd get better radiology, at a higher price (the price of the AI software, plus the additional hours the radiologist would work).

But back to making the AI bubble pay off: for AI to pay off, the human in the loop has to reduce the costs of the business buying an AI. No one who invests in an AI company believes that their returns will come from business customers to agree to increase their costs. The AI can't do your job, but the AI salesman can convince your boss to fire you and replace you with an AI anyway – that pitch is the most successful form of AI disinformation in the world.

An AI that "hallucinates" bad advice to fliers can't replace human customer service reps, but airlines are firing reps and replacing them with chatbots:

https://www.bbc.com/travel/article/20240222-air-canada-chatbot-misinformation-what-travellers-should-know

An AI that "hallucinates" bad legal advice to New Yorkers can't replace city services, but Mayor Adams still tells New Yorkers to get their legal advice from his chatbots:

https://arstechnica.com/ai/2024/03/nycs-government-chatbot-is-lying-about-city-laws-and-regulations/

The only reason bosses want to buy robots is to fire humans and lower their costs. That's why "AI art" is such a pisser. There are plenty of harmless ways to automate art production with software – everything from a "healing brush" in Photoshop to deepfake tools that let a video-editor alter the eye-lines of all the extras in a scene to shift the focus. A graphic novelist who models a room in The Sims and then moves the camera around to get traceable geometry for different angles is a centaur – they are genuinely offloading some finicky drudgework onto a robot that is perfectly attentive and vigilant.

But the pitch from "AI art" companies is "fire your graphic artists and replace them with botshit." They're pitching a world where the robots get to do all the creative stuff (badly) and humans have to work at robotic pace, with robotic vigilance, in order to catch the mistakes that the robots make at superhuman speed.

Reverse centaurism is brutal. That's not news: Charlie Chaplin documented the problems of reverse centaurs nearly 100 years ago:

https://en.wikipedia.org/wiki/Modern_Times_(film)

As ever, the problem with a gadget isn't what it does: it's who it does it for and who it does it to. There are plenty of benefits from being a centaur – lots of ways that automation can help workers. But the only path to AI profitability lies in reverse centaurs, automation that turns the human in the loop into the crumple-zone for a robot:

https://estsjournal.org/index.php/ests/article/view/260

If you'd like an essay-formatted version of this post to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:

https://pluralistic.net/2024/04/01/human-in-the-loop/#monkey-in-the-middle

Image: Cryteria (modified) https://commons.wikimedia.org/wiki/File:HAL9000.svg

CC BY 3.0 https://creativecommons.org/licenses/by/3.0/deed.en

Jorge Royan (modified) https://commons.wikimedia.org/wiki/File:Munich_-_Two_boys_playing_in_a_park_-_7328.jpg

CC BY-SA 3.0 https://creativecommons.org/licenses/by-sa/3.0/deed.en

Noah Wulf (modified) https://commons.m.wikimedia.org/wiki/File:Thunderbirds_at_Attention_Next_to_Thunderbird_1_-_Aviation_Nation_2019.jpg

CC BY-SA 4.0 https://creativecommons.org/licenses/by-sa/4.0/deed.en

375 notes · View notes

cwalshuk · 1 year

Text

The Artemis I Mission: To the Moon and Back

The Artemis I mission was the first integrated test of the Orion spacecraft, the Space Launch System (SLS) rocket, and Exploration Ground Systems at NASA’s Kennedy Space Center in Florida. We’ll use these deep space exploration systems on future Artemis missions to send astronauts to the Moon and prepare for our next giant leap: sending the first humans to Mars.

Take a visual journey through the mission, starting from launch, to lunar orbit, to splashdown.

Liftoff

The SLS rocket carrying the Orion spacecraft launched on Nov. 16, 2022, from Launch Complex 39B at NASA’s Kennedy Space Center in Florida. The world’s most powerful rocket performed with precision, meeting or exceeding all expectations during its debut launch on Artemis I.

"This is Your Moment"

Following the successful launch of Artemis I, Launch Director Charlie Blackwell-Thompson congratulates the launch team.

“The harder the climb, the better the view,” she said. “We showed the space coast tonight what a beautiful view it is.”

That's Us

On Orion’s first day of flight, a camera on the tip of one of Orion’s solar arrays captured this image of Earth.

Inside Orion

On the third day of the mission, Artemis I engineers activated the Callisto payload, a technology demonstration developed by Lockheed Martin, Amazon, and Cisco that tested a digital voice assistant and video conferencing capabilities in a deep space environment. In the image, Commander Moonikin Campos occupies the commander’s seat inside the spacecraft. The Moonikin is wearing an Orion Crew Survival System suit, the same spacesuit that Artemis astronauts will use during launch, entry, and other dynamic phases of their missions. Campos is also equipped with sensors that recorded acceleration and vibration data throughout the mission that will help NASA protect astronauts during Artemis II. The Moonikin was one of three “passengers” that flew aboard Orion. Two female-bodied model human torsos, called phantoms, were aboard. Zohar and Helga, named by the Israel Space Agency (ISA) and the German Aerospace Center (DLR) respectively, supported the Matroshka AstroRad Radiation Experiment (MARE), an experiment to provide data on radiation levels during lunar missions. Snoopy, wearing a mock orange spacesuit, also can be seen floating in the background. The character served as the zero-gravity indicator during the mission, providing a visual signifier that Orion is in space.

Far Side of the Moon

A portion of the far side of the Moon looms large in this image taken by a camera on the tip of one of Orion’s solar arrays on the sixth day of the mission.

First Close Approach

The Orion spacecraft captured some of the closest photos of the Moon from a spacecraft built for humans since the Apollo era — about 80 miles (128 km) above the lunar surface. This photo was taken using Orion’s optical navigational system, which captures black-and-white images of the Earth and Moon in different phases and distances.

Distant Retrograde Orbit

Orion entered a distant retrograde orbit around the Moon almost two weeks into the mission. The orbit is “distant” in the sense that it’s at a high altitude approximately 50,000 miles (80,467 km) from the surface of the Moon. Orion broke the record for farthest distance of a spacecraft designed to carry humans to deep space and safely return them to Earth, reaching a maximum distance of 268,563 miles (432,210 km).

Second Close Approach

On the 20th day of the mission, the spacecraft made its second and final close approach to the Moon flying 79.2 miles (127.5 km) above the lunar surface to harness the Moon’s gravity and accelerate for the journey back to Earth.

Cameras mounted on the crew module of the Orion spacecraft captured these views of the Moon’s surface before its return powered flyby burn.

Heading Home

After passing behind the far side of the Moon on Flight Day 20, Orion powered a flyby burn that lasted approximately 3 minutes and 27 seconds to head home. Shortly after the burn was complete, the Orion spacecraft captured these views of the Moon and Earth, which appears as a distant crescent.

Parachutes Deployed

Prior to entering the Earth’s atmosphere, Orion’s crew module separated from its service module, which is the propulsive powerhouse provided by ESA (European Space Agency). During re-entry, Orion endured temperatures about half as hot as the surface of the Sun at about 5,000 degrees Fahrenheit (2,760 degrees Celsius). Within about 20 minutes, Orion slowed from nearly 25,000 mph (40,236 kph) to about 20 mph (32 kph) for its parachute-assisted splashdown.

Splashdown

On Dec. 11, the Orion spacecraft splashed down in the Pacific Ocean off the coast of California after traveling 1.4 million miles (2.3 million km) over a total of 25.5 days in space. Teams are in the process of returning Orion to Kennedy Space Center in Florida. Once at Kennedy, teams will open the hatch and unload several payloads, including Commander Moonikin Campos, the space biology experiments, Snoopy, and the official flight kit. Next, the capsule and its heat shield will undergo testing and analysis over the course of several months.

Make sure to follow us on Tumblr for your regular dose of space!

2K notes · View notes