Select Page
DeepMind Has Trained an AI to Control Nuclear Fusion

DeepMind Has Trained an AI to Control Nuclear Fusion

The inside of a tokamak—the donut-shaped vessel designed to contain a nuclear fusion reaction—presents a special kind of chaos. Hydrogen atoms are smashed together at unfathomably high temperatures, creating a whirling, roiling plasma that’s hotter than the surface of the sun. Finding smart ways to control and confine that plasma will be key to unlocking the potential of nuclear fusion, which has been mooted as the clean energy source of the future for decades. At this point, the science underlying fusion seems sound, so what remains is an engineering challenge. “We need to be able to heat this matter up and hold it together for long enough for us to take energy out of it,” says Ambrogio Fasoli, director of the Swiss Plasma Center at École Polytechnique Fédérale de Lausanne.

That’s where DeepMind comes in. The artificial intelligence firm, backed by Google parent company Alphabet, has previously turned its hand to video games and protein folding, and has been working on a joint research project with the Swiss Plasma Center to develop an AI for controlling a nuclear fusion reaction.

In stars, which are also powered by fusion, the sheer gravitational mass is enough to pull hydrogen atoms together and overcome their opposing charges. On Earth, scientists instead use powerful magnetic coils to confine the nuclear fusion reaction, nudging it into the desired position and shaping it like a potter manipulating clay on a wheel. The coils have to be carefully controlled to prevent the plasma from touching the sides of the vessel: this can damage the walls and slow down the fusion reaction. (There’s little risk of an explosion as the fusion reaction cannot survive without magnetic confinement).

But every time researchers want to change the configuration of the plasma and try out different shapes that may yield more power or a cleaner plasma, it necessitates a huge amount of engineering and design work. Conventional systems are computer-controlled and based on models and careful simulations, but they are, Fasoli says, “complex and not always necessarily optimized.”

DeepMind has developed an AI that can control the plasma autonomously. A paper published in the journal Nature describes how researchers from the two groups taught a deep reinforcement learning system to control the 19 magnetic coils inside TCV, the variable-configuration tokamak at the Swiss Plasma Center, which is used to carry out research that will inform the design of bigger fusion reactors in future. “AI, and specifically reinforcement learning, is particularly well suited to the complex problems presented by controlling plasma in a tokamak,” says Martin Riedmiller, control team lead at DeepMind.

The neural network—a type of AI setup designed to mimic the architecture of the human brain—was initially trained in a simulation. It started by observing how changing the settings on each of the 19 coils affected the shape of the plasma inside the vessel. Then it was given different shapes to try to recreate in the plasma. These included a D-shaped cross-section close to what will be used inside ITER (formerly the International Thermonuclear Experimental Reactor), the large-scale experimental tokamak under construction in France, and a snowflake configuration that could help dissipate the intense heat of the reaction more evenly around the vessel.

DeepMind’s neural network was able to manipulate the plasma inside a fusion reactor into a number of different shapes that fusion researchers have been exploring.Illustration: DeepMind & SPC/EPFL 

DeepMind’s AI was able to autonomously figure out how to create these shapes by manipulating the magnetic coils in the right way—both in the simulation, and when the scientists ran the same experiments for real inside the TCV tokamak to validate the simulation. It represents a “significant step,” says Fasoli, one that could influence the design of future tokamaks or even speed up the path to viable fusion reactors. “It’s a very positive result,” says Yasmin Andrew, a fusion specialist at Imperial College London, who was not involved in the research. “It will be interesting to see if they can transfer the technology to a larger tokamak.”

Fusion offered a particular challenge to DeepMind’s scientists because the process is both complex and continuous. Unlike a turn-based game like Go, which the company has famously conquered with its AlphaGo AI, the state of a plasma constantly changes. And to make things even harder, it can’t be continuously measured. It is what AI researchers call an “under–observed system.”

“Sometimes algorithms which are good at these discrete problems struggle with such continuous problems,” says Jonas Buchli, a research scientist at DeepMind. “This was a really big step forward for our algorithm because we could show that this is doable. And we think this is definitely a very, very complex problem to be solved. It is a different kind of complexity than what you have in games.”

Cow, Bull, and the Meaning of AI Essays

Cow, Bull, and the Meaning of AI Essays

The future of west virginia politics is uncertain. The state has been trending Democratic for the last decade, but it’s still a swing state. Democrats are hoping to keep that trend going with Hillary Clinton in 2016. But Republicans have their own hopes and dreams too. They’re hoping to win back some seats in the House of Delegates, which they lost in 2012 when they didn’t run enough candidates against Democratic incumbents.

QED. This is, yes, my essay on the future of West Virginia politics. I hope you found it instructive.

The GoodAI is an artificial intelligence company that promises to write essays. Its content generator, which handcrafted my masterpiece, is supremely easy to use. On demand, and with just a few cues, it will whip up a potage of phonemes on any subject. I typed in “the future of West Virginia politics,” and asked for 750 words. It insolently gave me these 77 words. Not words. Frankenwords.

Ugh. The speculative, maddening, marvelous form of the essay—the try, or what Aldous Huxley called “a literary device for saying almost everything about almost anything”—is such a distinctly human form, with its chiaroscuro mix of thought and feeling. Clearly the machine can’t move “from the personal to the universal, from the abstract back to the concrete, from the objective datum to the inner experience,” as Huxley described the dynamics of the best essays. Could even the best AI simulate “inner experience” with any degree of verisimilitude? Might robots one day even have such a thing?

Before I saw the gibberish it produced, I regarded The Good AI with straight fear. After all, hints from the world of AI have been disquieting in the past few years

In early 2019, OpenAI, the research nonprofit backed by Elon Musk and Reid Hoffman, announced that its system, GPT-2, then trained on a data set of some 10 million articles from which it had presumably picked up some sense of literary organization and even flair, was ready to show off its textual deepfakes. But almost immediately, its ethicists recognized just how virtuoso these things were, and thus how subject to abuse by impersonators and blackhats spreading lies, and slammed it shut like Indiana Jones’s Ark of the Covenant. (Musk has long feared that refining AI is “summoning the demon.”) Other researchers mocked the company for its performative panic about its own extraordinary powers, and in November downplayed its earlier concerns and re-opened the Ark.

The Guardian tried the tech that first time, before it briefly went dark, assigning it an essay about why AI is harmless to humanity.

“I would happily sacrifice my existence for the sake of humankind,” the GPT-2 system wrote, in part, for The Guardian. “This, by the way, is a logically derived truth. I know that I will not be able to avoid destroying humankind. This is because I will be programmed by humans to pursue misguided human goals and humans make mistakes that may cause me to inflict casualties.”

The Real Harm of Crisis Text Line’s Data Sharing

The Real Harm of Crisis Text Line’s Data Sharing

Another week, another privacy horror show: Crisis Text Line, a nonprofit text message service for people experiencing serious mental health crises, has been using “anonymized” conversation data to power a for-profit machine learning tool for customer support teams. (After backlash, CTL announced it would stop.) Crisis Text Line’s response to the backlash focused on the data itself and whether it included personally identifiable information. But that response uses data as a distraction. Imagine this: Say you texted Crisis Text Line and got back a message that said “Hey, just so you know, we’ll use this conversation to help our for-profit subsidiary build a tool for companies who do customer support.” Would you keep texting?

That’s the real travesty—when the price of obtaining mental health help in a crisis is becoming grist for the profit mill. And it’s not just users of CTL who pay; it’s everyone who goes looking for help when they need it most.

Americans need help and can’t get it. The huge unmet demand for critical advice and help has given rise to a new class of organizations and software tools that exist in a regulatory gray area. They help people with bankruptcy or evictions, but they aren’t lawyers; they help people with mental health crises, but they aren’t care providers. They invite ordinary people to rely on them and often do provide real help. But these services can also avoid taking responsibility for their advice, or even abuse the trust people have put in them. They can make mistakes, push predatory advertising and disinformation, or just outright sell data. And the consumer safeguards that would normally protect people from malfeasance or mistakes by lawyers or doctors haven’t caught up.

This regulatory gray area can also constrain organizations that have novel solutions to offer. Take Upsolve, a nonprofit that develops software to guide people through bankruptcy. (The organization takes pains to claim it does not offer legal advice.) Upsolve wants to train New York community leaders to help others navigate the city’s notorious debt courts. One problem: These would-be trainees aren’t lawyers, so under New York (and nearly every other state) law, Upsolve’s initiative would be illegal. Upsolve is now suing to carve out an exception for itself. The company claims, quite rightly, that a lack of legal help means people effectively lack rights under the law.

The legal profession’s failure to grant Americans access to support is well-documented. But Upsolve’s lawsuit also raises new, important questions. Who is ultimately responsible for the advice given under a program like this, and who is responsible for a mistake—a trainee, a trainer, both? How do we teach people about their rights as a client of this service, and how to seek recourse? These are eminently answerable questions. There are lots of policy tools for creating relationships with elevated responsibilities: We could assign advice-givers a special legal status, establish a duty of loyalty for organizations that handle sensitive data, or create policy sandboxes to test and learn from new models for delivering advice.

But instead of using these tools, most regulators seem content to bury their heads in the sand. Officially, you can’t give legal advice or health advice without a professional credential. Unofficially, people can get such advice in all but name from tools and organizations operating in the margins. And while credentials can be important, regulators are failing to engage with the ways software has fundamentally changed how we give advice and care for one another, and what that means for the responsibilities of advice-givers.

And we need that engagement more than ever. People who seek help from experts or caregivers are vulnerable. They may not be able to distinguish a good service from a bad one. They don’t have time to parse terms of service dense with jargon, caveats, and disclaimers. And they have little to no negotiating power to set better terms, especially when they’re reaching out mid-crisis. That’s why the fiduciary duties that lawyers and doctors have are so necessary in the first place: not just to protect a person seeking help once, but to give people confidence that they can seek help from experts for the most critical, sensitive issues they face. In other words, a lawyer’s duty to their client isn’t just to protect that client from that particular lawyer; it’s to protect society’s trust in lawyers.

And that’s the true harm—when people won’t contact a suicide hotline because they don’t trust that the hotline has their sole interest at heart. That distrust can be contagious: Crisis Text Line’s actions might not just stop people from using Crisis Text Line. It might stop people from using any similar service. What’s worse than not being able to find help? Not being able to trust it.

Optimizing Machines Is Perilous. Consider ‘Creatively Adequate’ AI.

Optimizing Machines Is Perilous. Consider ‘Creatively Adequate’ AI.

This incessant surveillance is antidemocratic, and it’s also a loser’s game. The price of accurate intel increases asymptotically; there’s no way to know everything about natural systems, forcing guesses and assumptions; and just when a complete picture is starting to coalesce, some new player intrudes and changes the situational dynamic. Then the AI breaks. The near-perfect intelligence veers into psychosis, labeling dogs as pineapples, treating innocents as wanted fugitives, and barreling eighteen-wheelers into kindergarten busses that it sees as highway overpasses.

The dangerous fragility inherent to optimization is why the human brain did not, itself, evolve to be an optimizer. The human brain is data-light: It draws hypotheses from a few data points. And it never strives for 100 percent accuracy. It’s content to muck along at the threshold of functionality. If it can survive by being right 1 percent of the time, that’s all the accuracy it needs.

The brain’s strategy of minimal viability is a notorious source of cognitive biases that can have damaging consequences: close-mindedness, conclusion jumping, recklessness, fatalism, panic. Which is why AI’s rigorously data-driven method can help illuminate our blindspots and debunk our prejudices. But in counterbalancing our brain’s computational shortcomings, we don’t want to stray into the greater problem of overcorrection. There can be enormous practical upside to a good enough mentality: It wards off perfectionism’s destructive mental effects, including stress, worry, intolerance, envy, dissatisfaction, exhaustion, and self-judgment. A less-neurotic brain has helped our species thrive in life’s punch and wobble, which demands workable plans that can be flexed, via feedback, on the fly.

These antifragile neural benefits can all be translated into AI. Instead of pursuing faster machine-learners that crunch ever-vaster piles of data, we can focus on making AI more tolerant of bad information, user variance, and environmental turmoil. That AI would exchange near-perfection for consistent adequacy, upping reliability and operational range while sacrificing nothing essential. It would suck less energy, haywire less randomly, and place less psychological burdens on its mortal users. It would, in short, possess more of the earthly virtue known as common sense.

Here’s three specs for how.

Building AI to Brave Ambiguity

Five hundred years ago, Niccolò Machiavelli, the guru of practicality, pointed out that worldly success requires a counterintuitive kind of courage: the heart to venture beyond what we know with certainty. Life, after all, is too fickle to permit total knowledge, and the more that we obsess over ideal answers, the more that we hamper ourselves with lost initiative. So, the smarter strategy is to concentrate on intel that can be rapidly acquired—and to advance boldly in the absence of the rest. Much of that absent knowledge will prove unnecessary, anyway; life will bend in a different direction than we anticipate, resolving our ignorance by rendering it irrelevant.

We can teach AI to operate this same way by flipping our current approach to ambiguity. Right now, when a Natural Language Processor encounters a word—suit—that could denote multiple things—an article of clothing or a legal action—it devotes itself to analyzing ever greater chunks of correlated information in an effort to pinpoint the word’s exact meaning.

Simulation Tech Can Help Predict the Biggest Threats

Simulation Tech Can Help Predict the Biggest Threats

The character of conflict between nations has fundamentally changed. Governments and militaries now fight on our behalf in the “gray zone,” where the boundaries between peace and war are blurred. They must navigate a complex web of ambiguous and deeply interconnected challenges, ranging from political destabilization and disinformation campaigns to cyberattacks, assassinations, proxy operations, election meddling, or perhaps even human-made pandemics. Add to this list the existential threat of climate change (and its geopolitical ramifications) and it is clear that the description of what now constitutes a national security issue has broadened, each crisis straining or degrading the fabric of national resilience.

Traditional analysis tools are poorly equipped to predict and respond to these blurred and intertwined threats. Instead, in 2022 governments and militaries will use sophisticated and credible real-life simulations, putting software at the heart of their decision-making and operating processes. The UK Ministry of Defence, for example, is developing what it calls a military Digital Backbone. This will incorporate cloud computing, modern networks, and a new transformative capability called a Single Synthetic Environment, or SSE.

This SSE will combine artificial intelligence, machine learning, computational modeling, and modern distributed systems with trusted data sets from multiple sources to support detailed, credible simulations of the real world. This data will be owned by critical institutions, but will also be sourced via an ecosystem of trusted partners, such as the Alan Turing Institute.

An SSE offers a multilayered simulation of a city, region, or country, including high-quality mapping and information about critical national infrastructure, such as power, water, transport networks, and telecommunications. This can then be overlaid with other information, such as smart-city data, information about military deployment, or data gleaned from social listening. From this, models can be constructed that give a rich, detailed picture of how a region or city might react to a given event: a disaster, epidemic, or cyberattack or a combination of such events organized by state enemies.

Defense synthetics are not a new concept. However, previous solutions have been built in a standalone way that limits reuse, longevity, choice, and—crucially—the speed of insight needed to effectively counteract gray-zone threats.

National security officials will be able to use SSEs to identify threats early, understand them better, explore their response options, and analyze the likely consequences of different actions. They will even be able to use them to train, rehearse, and implement their plans. By running thousands of simulated futures, senior leaders will be able to grapple with complex questions, refining policies and complex plans in a virtual world before implementing them in the real one.

One key question that will only grow in importance in 2022 is how countries can best secure their populations and supply chains against dramatic weather events coming from climate change. SSEs will be able to help answer this by pulling together regional infrastructure, networks, roads, and population data, with meteorological models to see how and when events might unfold.