Sep 5 2024
The race to replace the powerful greenhouse gas that underpins the power grid

The power grid is underpinned by a single gas that is used to insulate a range of high-voltage equipment. The problem is, it’s also a super powerful greenhouse gas, a nightmare for climate change.

Sulfur hexafluoride (or SF6) is far from the most common gas that warms the planet, contributing around 1% of warming to date—carbon dioxide and methane are much more well-known and abundant. However, like many other fluorinated gases, SF6 is especially potent: It traps about 20,000 times more energy than carbon dioxide does over the course of a century, and it can last in the atmosphere for 1,000 years or more.

Despite their relatively small contributions so far, emissions of the gas are ticking up, and the growth rate has been climbing every year. SF6 emissions in China nearly doubled between 2011 and 2021, accounting for more than half the world’s emissions of the gas.

Now, companies are looking to do away with equipment that relies on the gas and searching for replacements that can match its performance. Last week, Hitachi Energy announced it’s producing new equipment that replaces SF6 with other materials. And there’s momentum building to ban SF6 in the power industry, including a recently passed plan in the European Union that will phase out the gas’s use in high-voltage equipment by 2032. 

As equipment manufacturers work to produce alternatives, some researchers say that we should go even further and are trying to find solutions that avoid fluorine-containing materials entirely.

High voltage, high stakes

You probably have a circuit-breaker box in your home—if a circuit gets overloaded, the breaker flips, stopping the flow of electricity. The power grid has something similar, called switchgear.  

The difference is, it often needs to handle something like a million times more energy than your home’s equipment does, says Markus Heimbach, executive vice president and managing director of the high-voltage products business unit at Hitachi Energy. That’s because parts of the power grid operate at high voltages, allowing them to move energy around while losing as little as possible. Those high voltages require careful insulation at all times and safety measures in case something goes wrong.

Some switchgear uses the same materials as your home circuit-breaker boxes—there’s air around it to insulate it. But when it’s scaled up to handle high voltage, it ends up being gigantic and requiring a large land footprint, making it inconvenient for larger, denser cities.

The solution today is SF6, “a super gas, from a technology point of view,” Heimbach says. It’s able to insulate equipment during normal operation and help interrupt current when needed. And the whole thing has a much smaller footprint than air-insulated equipment.

The problem is, small amounts of SF6 leak out of equipment during normal operation, and more can be released during a failure or when old equipment isn’t handled properly. When the gas escapes, its strong ability to trap heat and the fact that it has such a long lifetime makes it a menace in the atmosphere.

Some governments will soon ban the gas for the power industry, which makes up the vast majority of the emissions. The European Union agreed to ban SF6-containing medium-voltage switchgear by 2030, and high-voltage switchgear that uses the gas by 2032. Several states in the US have proposed or adopted limits and phaseouts.

Making changes 

Hitachi Energy recently announced it’s producing high-voltage switchgear that can handle up to 550 kilovolts (kV). The model follows products rated for 420 kV the company began installing in 2023—there are more than 250 booked by customers today, Heimbach says.  

Hitachi Energy’s new switchgear substitutes SF6 with a gas mixture that contains mostly carbon dioxide and oxygen. It works as well as SF6 and is as safe and reliable but with a much lower global warming potential, trapping 99% less energy in the atmosphere, Heimbach says. 

However, for some of its new equipment, Hitachi Energy still uses some C4-fluoronitriles, which helps with insulation, Heimbach says. This gas is present at a low fraction, less than 5% of the mixture, and it’s less potent than SF6, Heimbach says. But C4-fluoronitriles are still powerful greenhouse gases, up to a few thousand times more potent than carbon dioxide. These and other fluorinated substances could soon be in trouble too—chemical giant 3M announced in late 2022 that the company would stop manufacturing all fluoropolymers, fluorinated fluids, and PFAS-additive products by 2025.

In order to eliminate the need for fluorine-containing gases, some researchers are looking into the grid’s past for alternatives. “We know that there’s no one-for-one replacement gas that has the properties of SF6,” says Lukas Graber, an associate professor in electrical engineering at Georgia Institute of Technology.

SF6 is both extremely stable and extremely electronegative, meaning it tends to grab onto free electrons, and nothing else can quite match it, Graber says. So he’s working on a research project that aims to replace SF6 gas with supercritical carbon dioxide. (Supercritical fluids are those at temperatures and pressures so high that distinct liquid and gas phases don’t quite exist.) The inspiration came from equipment that used to use oil-based materials—instead of trying to grab electrons like SF6, supercritical carbon dioxide can basically slow them down.

Graber and his research team received project funding from the US Department of Energy’s Advanced Research Projects Agency for Energy. The first small-scale prototype is nearly finished, he adds, and the plan is to test out a full-scale prototype in 2025.

Utilities are known for being conservative, since the safety and reliability of the electrical grid have high stakes, Hitachi Energy’s Heimbach says. But with more SF6 bans coming, they’ll need to find and adopt solutions that don’t rely on the gas.

Ecommerce MGMT 0 Comments
Sep 5 2024
How “personhood credentials” could help prove you’re a human online

As AI models become better at mimicking human behavior, it’s becoming increasingly difficult to distinguish between real human internet users and sophisticated systems imitating them. 

That’s a real problem when those systems are deployed for nefarious ends like spreading misinformation or conducting fraud, and it makes it a lot harder to trust what you encounter online.

A group of 32 researchers from institutions including OpenAI, Microsoft, MIT, and Harvard has developed a potential solution—a verification concept called “personhood credentials.” These credentials prove that their holder is a real person, without revealing any further information about the person’s identity. The team explored the idea in a non-peer-reviewed paper posted to the arXiv preprint server earlier this month.

Personhood credentials rely on the fact that AI systems still cannot bypass state-of-the-art cryptographic systems or pass as people in the offline, real world. 

To request such credentials, people would have to physically go to one of a number of issuers, like a government or some other kind of trusted organization. They would be asked to provide evidence of being a real human, such as a passport or biometric data. Once approved, they’d receive a single credential to store on their devices the way it’s currently possible to store credit and debit cards in smartphones’ wallet apps.

To use these credentials online, a user could present them to a third-party digital service provider, which could then verify them using a cryptographic protocol called a zero-knowledge proof. That would confirm the holder was in possession of a personhood credential without disclosing any further unnecessary information.

The ability to filter out anyone other than verified humans on a platform could be useful in many ways. People could reject Tinder matches that don’t come with personhood credentials, for example, or choose not to see anything on social media that wasn’t definitely posted by a person. 

The authors want to encourage governments, companies, and standards bodies to consider adopting such a system in the future to prevent AI deception from ballooning out of our control. 

“AI is everywhere. There will be many issues, many problems, and many solutions,” says Tobin South, a PhD student at MIT who worked on the project. “Our goal is not to prescribe this to the world, but to open the conversation about why we need this and how it could be done.”

Possible technical options already exist. For example, a network called Idena claims to be the first blockchain proof-of-person system. It works by getting humans to solve puzzles that would be difficult for bots within a short time frame. The controversial Worldcoin program, which collects users’ biometric data, bills itself as the world’s largest privacy-preserving human identity and financial network. It recently partnered with the Malaysian government to provide proof of humanness online by scanning users’ irises, which creates a code. As in the concept of personhood credentials, each code is protected using cryptography.

However, the project has been criticized for using deceptive marketing practices, collecting more personal data than acknowledged, and failing to obtain meaningful consent from users. Regulators in Hong Kong and Spain banned Worldcoin from operating earlier this year, while its operations have been suspended in countries including Brazil, Kenya, and India. 

So fresh concepts are still needed. The rapid rise of accessible AI tools has ushered in a dangerous period in which internet users are hyper-suspicious about what is and isn’t true online, says Henry Ajder, an expert on AI and deepfakes who is an advisor to Meta and the UK government. And while ideas for verifying personhood have been around for some time, these credentials feel like one of the most substantive ideas for how to push back against encroaching skepticism, he says.

But the biggest challenge the credentials will face is getting enough platforms, digital services, and governments to adopt them, since they may feel uncomfortable conforming to a standard they don’t control. “For this to work effectively, it would have to be something which is universally adopted,” he says. “In principle the technology is quite compelling, but in practice and the messy world of humans and institutions, I think there would be quite a lot of resistance.”

Martin Tschammer, head of security at the startup Synthesia, which creates AI-generated hyperrealistic deepfakes, says he agrees with the principle driving personhood credentials: the need to verify humans online. However, he is unsure whether it’s the right solution or whether it would be practical to implement. He also expresses skepticism over who would run such a scheme.  

“We may end up in a world in which we centralize even more power and concentrate decision-making over our digital lives, giving large internet platforms even more ownership over who can exist online and for what purpose,” he says. “And given the lackluster performance of some governments in adopting digital services, and autocratic tendencies that are on the rise, is it practical or realistic to expect this type of technology to be adopted en masse and in a responsible way by the end of this decade?” 

Rather than waiting for collaboration across industries, Synthesia is currently evaluating how to integrate other personhood-proving mechanisms into its products. He says it already has several measures in place. For example, it requires businesses to prove that they are legitimate registered companies, and will ban and refuse refunds to customers found to have broken its rules. 

One thing is clear: We are in urgent need of ways to differentiate humans from bots, and encouraging discussions between stakeholders in the tech and policy worlds is a step in the right direction, says Emilio Ferrara, a professor of computer science at the University of Southern California, who was not involved in the project. 

“We’re not far from a future where, if things remain unchecked, we’re going to be essentially unable to tell apart interactions that we have online with other humans or some kind of bots. Something has to be done,” he says. “We can’t be naïve as previous generations were with technologies.”

Ecommerce MGMT 0 Comments
Sep 5 2024
AI’s impact on elections is being overblown

This year, close to half the world’s population has the opportunity to participate in an election. And according to a steady stream of pundits, institutions, academics, and news organizations, there’s a major new threat to the integrity of those elections: artificial intelligence. 

The earliest predictions warned that a new AI-powered world was, apparently, propelling us toward a “tech-enabled Armageddon” where “elections get screwed up”, and that “anybody who’s not worried [was] not paying attention.” The internet is full of doom-laden stories proclaiming that AI-generated deepfakes will mislead and influence voters, as well as enabling new forms of personalized and targeted political advertising. Though such claims are concerning, it is critical to look at the evidence. With a substantial number of this year’s elections concluded, it is a good time to ask how accurate these assessments have been so far. The preliminary answer seems to be not very; early alarmist claims about AI and elections appear to have been blown out of proportion.

While there will be more elections this year where AI could have an effect, the United States being one likely to attract particular attention, the trend observed thus far is unlikely to change. AI is being used to try to influence electoral processes, but these efforts have not been fruitful. Commenting on the upcoming US election, Meta’s latest Adversarial Threat Report acknowledged that AI was being used to meddle—for example, by Russia-based operations—but that “GenAI-powered tactics provide only incremental productivity and content-generation gains” to such “threat actors.” This echoes comments from the company’s president of global affairs, Nick Clegg, who earlier this year stated that “it is striking how little these tools have been used on a systematic basis to really try to subvert and disrupt the elections.”

Far from being dominated by AI-enabled catastrophes, this election “super year” at that point was pretty much like every other election year.

While Meta has a vested interest in minimizing AI’s alleged impact on elections, it is not alone. Similar findings were also reported by the UK’s respected Alan Turing Institute in May. Researchers there studied more than 100 national elections held since 2023 and found “just 19 were identified to show AI interference.” Furthermore, the evidence did not demonstrate any “clear signs of significant changes in election results compared to the expected performance of political candidates from polling data.”

This all raises a question: Why were these initial speculations about AI-enabled electoral interference so off, and what does it tell us about the future of our democracies? The short answer: Because they ignored decades of research on the limited influence of mass persuasion campaigns, the complex determinants of voting behaviors, and the indirect and human-mediated causal role of technology. 

First, mass persuasion is notoriously challenging. AI tools may facilitate persuasion, but other factors are critical. When presented with new information, people generally update their beliefs accordingly; yet even in the best conditions, such updating is often minimal and rarely translates into behavioral change. Though political parties and other groups invest colossal sums to influence voters, evidence suggests that most forms of political persuasion have very small effects at best. And in most high-stakes events, such as national elections, a multitude of factors are at play, diminishing the effect of any single persuasion attempt.

Second, for a piece of content to be influential, it must first reach its intended audience. But today, a tsunami of information is published daily by individuals, political campaigns, news organizations, and others. Consequently, AI-generated material, like any other content, faces significant challenges in cutting through the noise and reaching its target audience. Some political strategists in the United States have also argued that the overuse of AI-generated content might make people simply tune out, further reducing the reach of manipulative AI content. Even if a piece of such content does reach a significant number of potential voters, it will probably not succeed in influencing enough of them to alter election results.

Third, emerging research challenges the idea that using AI to microtarget people and sway their voting behavior works as well as initially feared. Voters seem to not only recognize excessively tailored messages but actively dislike them. According to some recent studies, the persuasive effects of AI are also, at least for now, vastly overstated. This is likely to remain the case, as ever-larger AI-based systems do not automatically translate to better persuasion. Political campaigns seem to have recognized this too. If you speak to campaign professionals, they will readily admit that they are using AI, but mainly to optimize “mundane” tasks such as fundraising, get-out-the-vote efforts, and overall campaign operations rather than generating new AI-generated, highly tailored content.

Fourth, voting behavior is shaped by a complex nexus of factors. These include gender, age, class, values, identities, and socialization. Information, regardless of its veracity or origin—whether made by an AI or a human—often plays a secondary role in this process. This is because the consumption and acceptance of information are contingent on preexisting factors, like whether it chimes with the person’s political leanings or values, rather than whether that piece of content happens to be generated by AI.

Concerns about AI and democracy, and particularly elections, are warranted. The use of AI can perpetuate and amplify existing social inequalities or reduce the diversity of perspectives individuals are exposed to. The harassment and abuse of female politicians with the help of AI is deplorable. And the perception, partially co-created by media coverage, that AI has significant effects could itself be enough to diminish trust in democratic processes and sources of reliable information, and weaken the acceptance of election results. None of this is good for democracy and elections. 

However, these points should not make us lose sight of threats to democracy and elections that have nothing to do with technology: mass voter disenfranchisement; intimidation of election officials, candidates, and voters; attacks on journalists and politicians; the hollowing out of checks and balances; politicians peddling falsehoods; and various forms of state oppression (including restrictions on freedom of speech, press freedom and the right to protest). 

Of at least 73 countries holding elections this year, only 47 are classified as full (or at least flawed) democracies, according to Our World in Data/Economist Democracy Index, with the rest being hybrid or authoritarian regimes. In countries where elections are not even free or fair, and where political choice that leads to real change is an illusion, people have arguably bigger fish to fry.

And still, technology—including AI—often becomes a convenient scapegoat, singled out by politicians and public intellectuals as one of the major ills befalling democratic life. Earlier this year, Swiss president Viola Amherd warned at the World Economic Forum in Davos, Switzerland, that “advances in artificial intelligence allow … false information to seem ever more credible” and present a threat to trust. Pope Francis, too, warned that fake news could be legitimized through AI. US Deputy Attorney General Lisa Monaco said that AI could supercharge mis- and disinformation and incite violence at elections. This August, the mayor of London, Sadiq Kahn, called for a review of the UK’s Online Safety Act after far-right riots across the country, arguing that “the way the algorithms work, the way that misinformation can spread very quickly and disinformation … that’s a cause to be concerned. We’ve seen a direct consequence of this.”

The motivations to blame technology are plenty and not necessarily irrational. For some politicians, it can be easier to point fingers at AI than to face scrutiny or commit to improving democratic institutions that could hold them accountable. For others, attempting to “fix the technology” can seem more appealing than addressing some of the fundamental issues that threaten democratic life. Wanting to speak to the zeitgeist might play a role, too.

Yet we should remember that there’s a cost to overreaction based on ill-founded assumptions, especially when other critical issues go unaddressed. Overly alarmist narratives about AI’s presumed effects on democracy risk fueling distrust and sowing confusion among the public—potentially further eroding already low levels of trust in reliable news and institutions in many countries. One point often raised in the context of these discussions is the need for facts. People argue that we cannot have democracy without facts and a shared reality. That is true. But we cannot bang on about needing a discussion rooted in facts when evidence against the narrative of AI turbocharging democratic and electoral doom is all too easily dismissed. Democracy is under threat, but our obsession with AI’s supposed impact is unlikely to make things better—and could even make them worse when it leads us to focus solely on the shiny new thing while distracting us from the more lasting problems that imperil democracies around the world. 

Felix M. Simon is a research fellow in AI and News at the Reuters Institute for the Study of Journalism; Keegan McBride is an assistant professor in AI, government, and policy at the Oxford Internet Institute; Sacha Altay is a research fellow in the department of political science at the University of Zurich.

Ecommerce MGMT 0 Comments
Sep 5 2024
Here’s how ed-tech companies are pitching AI to teachers

This story is from The Algorithm, our weekly newsletter on AI. To get it in your inbox first, sign up here.

This back-to-school season marks the third year in which AI models like ChatGPT will be used by thousands of students around the globe (among them my nephews, who tell me with glee each time they ace an assignment using AI). A top concern among educators remains that when students use such models to write essays or come up with ideas for projects, they miss out on the hard and focused thinking that builds creative reasoning skills. 

But this year, more and more educational technology companies are pitching schools on a different use of AI. Rather than scrambling to tamp down the use of it in the classroom, these companies are coaching teachers how to use AI tools to cut down on time they spend on tasks like grading, providing feedback to students, or planning lessons. They’re positioning AI as a teacher’s ultimate time saver. 

One company, called Magic School, says its AI tools like quiz generators and text summarizers are used by 2.5 million educators. Khan Academy offers a digital tutor called Khanmigo, which it bills to teachers as “your free, AI-powered teaching assistant.” Teachers can use it to assist students in subjects ranging from coding to humanities. Writing coaches like Pressto help teachers provide feedback on student essays.  

The pitches from ed-tech companies often cite a 2020 report from McKinsey and Microsoft, which found teachers work an average of 50 hours per week. Many of those hours, according to the report, consist of “late nights marking papers, preparing lesson plans, or filling out endless paperwork.” The authors suggested that embracing AI tools could save teachers 13 hours per week. 

Companies aren’t the only ones making this pitch. Educators and policymakers have also spent the last year pushing for AI in the classroom. Education departments in South Korea, Japan, Singapore, and US states like North Carolina and Colorado have issued guidance for how teachers can positively and safely incorporate AI. 

But when it comes to how willing teachers are to turn over some of their responsibilities to an AI model, the answer really depends on the task, according to Leon Furze, an educator and PhD candidate at Deakin University who studies the impact of generative AI on writing instruction and education.

“We know from plenty of research that teacher workload actually comes from data collection and analysis, reporting, and communications,” he says. “Those are all areas where AI can help.”

Then there are a host of not-so-menial tasks that teachers are more skeptical AI can excel at. They often come down to two core teaching responsibilities: lesson planning and grading. A host of companies offer large language models that they say can generate lesson plans to conform to different curriculum standards. Some teachers, including in some California districts, have also used AI models to grade and provide feedback for essays. For these applications of AI, Furze says, many of the teachers he works with are less confident in its reliability. 

When companies promise time savings for planning and grading, it is “a huge red flag,” he says, because “those are core parts of the profession.” He adds, “Lesson planning is—or should be—thoughtful, creative, even fun.” Automated feedback on creative skills like writing is controversial too: “Students want feedback from humans, and assessment is a way for teachers to get to know students. Some feedback can be automated, but not all.” 

So how eager are teachers to adopt AI to save time? Earlier this year, in May, a Pew research poll found that only 6% of teachers think AI can provide more benefits than harm in education. But with AI changing faster than ever, this school year might be when ed-tech companies start to win them over.

Now read the rest of The Algorithm

Deeper learning

How machine learning is helping us probe the secret names of animals

Until now, only humans, dolphins, elephants, and probably parrots had been known to use specific sounds to call out to other individuals. But now, researchers armed with audio recorders and pattern-recognition software are making unexpected discoveries about the secrets of animal names—at least with small monkeys called marmosets. They’ve found that the animals will adjust the sounds they make in a way that’s specific to whoever they’re “conversing” with at the time.  

Why this matters: In years past, it’s been argued that human language is unique and that animals lack both the brains and vocal apparatus to converse. But there’s growing evidence that isn’t the case, especially now that the use of names has been found in at least four distantly related species. Read more from Antonio Regalado.

Bits and bytes

How will AI change the future of sex? 

Porn and real-life sex affect each other in a loop. If people become accustomed to getting exactly what they want from erotic media, this could further affect their expectations of relationships. (MIT Technology Review)

There’s a new way to build neural networks that could make AI more understandable

The new method, studied in detail by a group led by researchers at MIT, could make it easier to understand why neural networks produce certain outputs, help verify their decisions, and even probe for bias. (MIT Technology Review)

Researchers built an “AI scientist.” What can it do?

The large language model does everything from reading the literature to writing and reviewing its own papers, but it has a limited range of applications so far. (Nature)

OpenAI is weighing changes to its corporate structure as it seeks more funding 

These discussions come as Apple, Nvidia, and Microsoft are considering a funding round that would value OpenAI at more than $100 billion. (Financial Times)

Ecommerce MGMT 0 Comments
Sep 5 2024
The UK is building an alarm system for climate tipping points

The UK’s new moonshot research agency just launched an £81 million ($106 million) program to develop early warning systems to sound the alarm if Earth gets perilously close to crossing climate tipping points.

A climate tipping point is a threshold beyond which certain ecosystems or planetary processes begin to shift from one stable state to another, triggering dramatic and often self-reinforcing changes in the climate system. 

The Advanced Research and Invention Agency (ARIA) will announce today that it’s seeking proposals to work on systems for two related climate tipping points. One is the accelerating melting of the Greenland Ice Sheet, which could raise sea levels dramatically. The other is the weakening of the North Atlantic Subpolar Gyre, a huge current rotating counterclockwise south of Greenland that may have played a role in triggering the Little Ice Age around the 14th century. 

The goal of the five-year program will be to reduce scientific uncertainty about when these events could occur, how they would affect the planet and the species on it, and over what period those effects might develop and persist. In the end, ARIA hopes to deliver a proof of concept demonstrating that early warning systems can be “affordable, sustainable, and justified.” No such dedicated system exists today, though there’s considerable research being done to better understand the likelihood and consequences of surpassing these and other climate tipping points.

Sarah Bohndiek, a program director for the tipping points research program, says we underappreciate the possibility that crossing these points could significantly accelerate the effects of climate change and increase the dangers, possibly within the next few decades.

By developing an early warning system, “we might be able to change the way that we think about climate change and think about our preparedness for it,” says Bohndiek, a professor of biomedical physics at the University of Cambridge. 

ARIA intends to support teams that will work toward three goals: developing low-cost sensors that can withstand harsh environments and provide more precise and needed data about the conditions of these systems; deploying those and other sensing technologies to create “an observational network to monitor these tipping systems”; and building computer models that harness the laws of physics and artificial intelligence to pick up “subtle early warning signs of tipping” in the data.

But observers stress that designing precise early warning systems for either system would be no simple feat and might not be possible anytime soon. Not only do scientists have limited understanding of these systems, but the data  on how they’ve behaved in the past is patchy and noisy, and setting up extensive monitoring tools in these environments is expensive and cumbersome. 

Still, there’s wide agreement that we need to better understand these systems and the risks that the world may face.

Unlocking breakthroughs

It is clear that the tipping of either of these systems could have huge effects on Earth and its inhabitants.

As the world warmed in recent decades, trillions of tons of ice melted off the Greenland Ice Sheet, pouring fresh water into the North Atlantic, pushing up ocean levels, and reducing the amount of heat that the snow and ice reflected back into space. 

Melting rates are increasing as Arctic warming speeds ahead of the global average and hotter ocean waters chip away at ice shelves that buttress land-based glaciers. Scientists fear that as those shelves collapse, the ice sheet will become increasingly unstable. 

The complete loss of the ice sheet would raise global sea levels by more than 20 feet (six meters), submerging coastlines and kick-starting mass climate migration around the globe.

But at any point along the way, the influx of water into the North Atlantic could also substantially slow down the convection systems that help to drive the Subpolar Gyre, because fresher water isn’t as dense and prone to sinking. (Saltier, cooler water readily sinks.)

The weakening of the Subpolar Gyre could cool parts of northwest Europe and eastern Canada, shift the jet stream northward, create more erratic weather patterns across Europe, and undermine the productivity of agriculture and fisheries, according to one study last year. 

The Subpolar Gyre may also influence the strength of the Atlantic Meridional Overturning Circulation (AMOC), a network of ocean currents that moves massive amounts of heat, salt, and carbon dioxide around the globe. The specifics of how a weakened Subpolar Gyre would affect the AMOC are still the subject of ongoing research, but a dramatic slowdown or shutdown of that system is considered one of the most dangerous climate tipping points. It could substantially cool Northern Europe, among other wide-ranging effects.  

The tipping of the AMOC itself, however, is not the focus of the ARIA research program. 

The agency, established last year to “unlock scientific and technological breakthroughs,” is a UK answer to the US’s DARPA and ARPA-E research programs. Other projects it’s funding include efforts to develop precision neurotechnologies, improve robot dexterity, and build safer and more energy-efficient AI systems. ARIA is also setting up programs for developing synthetic plants and exploring climate interventions that could cool the planet, including solar geoengineering. 

Bohndiek and the other program director of the tipping points program—Gemma Bale, an assistant professor at the University of Cambridge—are both medical physicists who previously focused on developing medical devices. At ARIA, they initially expected to work on efforts to decentralize health care.

But Bohndiek says they soon realized that “a lot of these things that need to change at the individual health level will be irrelevant if climate change truly is going to cross these big thresholds.” She adds, “If we’re going to end up in a society where the world is so much warmer … does the problem of decentralizing health care matter anymore?” 

Bohndiek and Bale stress that they hope the program will draw applications from researchers who haven’t traditionally worked on climate change. They add that any research teams proposing to work in or around Greenland must take appropriate steps to engage with local communities, governments, and other research groups.

Tipping dangers

Efforts are already underway to develop greater understanding of the Subpolar Gyre and the Greenland Ice Sheet, including the likelihood, timing, and consequences of their tipping into different states.

There are, for instance, regular field expeditions to measure and refine modeling of ice loss in Greenland. A variety of research groups have set up sensor networks that cross various points of the Atlantic to more closely monitor the shifting conditions of current systems. And several studies have already highlighted the appearance of some “early warning signals” of a potential collapse of the AMOC in the coming decades.

But the goal of the ARIA program is to accelerate such research efforts and sharpen the field’s focus on improving our ability to predict tipping events. 

William Johns, an oceanographer focused on observation of the AMOC at the University of Miami, says the field is a long way from being able to state confidently that systems like the Subpolar Gyre or AMOC will weaken beyond the bounds of normal natural fluctuations, much less say with any precision when they would do so. 

He stresses that there’s still wide disagreement between models on these sorts of questions and limited evidence of what took place before they tipped in the ancient past, all of which makes it difficult to even know what signals we should be monitoring for most closely.

Jaime Palter, an associate professor of oceanography at the University of Rhode Island, adds that she found it a “puzzling” choice to fund a research program focused on the tipping of the Subpolar Gyre. She notes that researchers believe the wind drives the system more than convection, that its connection to the AMOC isn’t well understood, and that the slowdown of the latter system is the one that more of the field is focused on—and more of the world is worried about.

But she and Johns both said that providing funds to monitor these systems more closely is critical to improve scientific understanding of how they work and the odds that they will tip.

Radical interventions

So what could the world do if ARIA or anyone else does manage to develop systems that can predict, with high confidence, that one of these systems will shift into a new state in, say, the next decade?

Bohndiek stresses that the effects of reaching a tipping point wouldn’t be immediate, and that the world would still have years or even decades to take actions that might prevent the breakdown of such systems, or begin adapting to the changes they’ll bring. In the case of runaway melting of the ice sheet, that could mean building higher seawalls or relocating cities. In the case of the Subpolar Gyre weakening, big parts of Europe might have to look to other areas of the world for their food supplies.

More reliable predictions might also alter people’s thinking about more dramatic interventions, such as massive and hugely expensive engineering projects to prop up ice shelves or to freeze glaciers more stably onto the bedrock they’re sliding upon. 

Similarly, they might shift how some people weigh the trade-offs between the dangers of climate change and the risks of interventions like solar geoengineering, which would involve releasing particles in the atmosphere that could reflect more heat back into space.

But some observers note that if enough fresh water is pouring into the Atlantic to weaken the gyre and substantially slow the broader Atlantic current system, there’s very little the world can do to stop it.

“I’m afraid I don’t really see an action you could take,” Johns says. “You can’t go vacuum up all the fresh water—it’s not going to be feasible—and you can’t stop it from melting on the scale we’d have to.”

Bale readily acknowledges that they’ve selected a very hard problem to solve, but she stresses that the point of ARIA research programs is to work at the “edge of the possible.” 

“We genuinely don’t know if an early warning system for these systems is possible,” she says. “But I think if it is possible, we know that it would be valuable and important for society, and that’s part of our mission.”

Ecommerce MGMT 0 Comments
Sep 5 2024
The Download: climate tipping point alarms, and AI’s vision of the 3028 Olympics

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

The UK is building an alarm system for climate tipping points

The news: The UK’s new moonshot research agency just launched an £81 million ($106 million) program to develop early warning systems to sound the alarm if Earth gets perilously close to crossing climate tipping points.

How they’re doing it: The teams the agency supports will work toward three goals: developing low-cost sensors to provide more precise data about the conditions of these systems; deploying those and other sensing technologies to create an observational network to monitor these tipping systems; and building computer models that harness physics and artificial intelligence to pick up subtle early warning signs of tipping in the data.

What it matters: The goal of the five-year program will be to reduce scientific uncertainty about when these events could occur, how they would affect the planet and the species on it, and over what period those effects might develop and persist. Read the full story.

—James Temple

What this futuristic Olympics video says about the state of generative AI

The Olympic Games in Paris just finished last month and the Paralympics are still underway, so the 2028 Summer Olympics in Los Angeles feel like a lifetime from now. But the prospect of watching the games in his home city has Josh Kahn, a filmmaker in the sports entertainment world, thinking even further into the future: What might an LA Olympics in the year 3028 look like?

It’s the perfect type of creative exercise for AI video generation, which allows users to generate fairly high-definition video in minutes. More than anything, the video shows what a boon the generative technology may be for creators. However, it also indicates what’s holding it back.

Read the full story.

—James O’Donnell

Coming soon: Our 2024 list of Innovators Under 35

To tackle complex global problems such as preventing disease and mitigating climate change, we’re going to need new ideas from our brightest minds. Every year, MIT Technology Review identifies a new class of Innovators Under 35 taking on these and other challenges.

On September 10, we will honor the 2024 class of Innovators Under 35. These 35 researchers and entrepreneurs are rising stars in their fields pursuing ambitious projects. Each is doing groundbreaking work to advance one of five areas: materials science, biotechnology, robotics, artificial intelligence, or climate and energy.

Keep an eye on this newsletter next Tuesday for our announcement of the 2024 list, which will be made available exclusively to subscribers. Read the full story for more details, and if you’re not a subscriber already, sign up here to save 25%.

—Amy Nordum

If you can’t wait until then, we’ll reveal our Innovator of the Year during a live broadcast on LinkedIn on Monday, September 9. Register here to be among the first to find out who it is, and learn about their work and the impact they’re having.

MIT Technology Review Narrated: How a tiny Pacific Island became the global capital of cybercrime

Tokelau is a group of three isolated atolls strung out across the Pacific Ocean between New Zealand and Hawaii.

Despite its size, Tokelau has become an internet giant. Until recently, its .tk domain had more users than any other country’s: a staggering 25 million. Yet only one website with a .tk domain is actually from Tokelau. Nearly all the others are used by spammers, phishers, and cybercriminals.

This is the story of how Tokelau unwittingly became the global capital of cybercrime—and its fight to fix its reputation.

This is our latest story to be turned into a MIT Technology Review Narrated podcast. In partnership with News Over Audio, we’ll be making a selection of our stories available, each one read by a professional voice actor. You’ll be able to listen to them on the go or download them to listen to offline.

We’re publishing a new story each week on Spotify and Apple Podcasts, including some taken from our most recent print magazine.

Just navigate to MIT Technology Review Narrated on either platform, and follow us to get all our new content as it’s released.

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 Clearview AI has been slapped with its largest fine to date 
A Dutch regulator has ordered it to pay more than 30 million euros for creating an illegal database of “unique biometric codes.” (The Verge)
+ Clearview has missed its window to appeal the decision. (Ars Technica)
+ It’s not the first time the company has been in legal hot water. (MIT Technology Review)

2 Bluesky is gaining millions of new users
It’s struggling to cope with the influx triggered by X’s shutdown in Brazil. (TechCrunch)
+ Some worry that judge Alexandre de Moraes has overreached his powers. (Economist $)

3 Training AI to appear human-like is unrewarding work
Workers cite long hours, low pay, and unstable employment. (Fast Company $)
+ The people paid to train AI are outsourcing their work… to AI. (MIT Technology Review) 

4 The Chinese Communist Party is silencing protesters in San Francisco
It’s part of a bigger pattern of China overreaching beyond its borders. (WP $)
+ How Twitter’s “Teacher Li” became the central hub of China protest information. (MIT Technology Review)

5 Mega AI data centers are coming
But the companies behind the multi-billion dollar projects are a mystery. (The Information $)
+ Energy-hungry data centers are quietly moving into cities. (MIT Technology Review)

6 New York vigilantes are making big bucks reporting idling polluters
The only problem is, the police keep getting in the way. (NY Mag $)

7 Beware of Bitcoin ATM scams
Criminals are finding new ways to part victims from their hard-earned cash. (The Verge) 

8 What’s going to happen to all our old gas-powered cars?
Giving them a second life undermines efforts to curb climate change. (Vox)
+ This company wants to inject petroleum carbon emissions back underground. (The Atlantic $)
+ The problem with plug-in hybrids? Their drivers. (MIT Technology Review)

9 You can’t play video games on Microsoft’s new AI computers
Their new chip architecture isn’t playing ball with more than 1,000 games. (WSJ $)

10 Yarn made from potato is a sustainable cotton alternative 🥔
There’s no guarantee it’ll taste good, though. (New Scientist $)

Quote of the day

“We’re a Brazilian app now.”

—Bluesky jokes about the millions of new users it’s gained in the wake of Brazil’s Supreme Court banning access to X in the country, the Financial Times reports.

The big story

Hydrogen trains could revolutionize how Americans get around

April 2024

Like a mirage speeding across the dusty desert outside Pueblo, Colorado, the first hydrogen-fuel-cell passenger train in the United States is getting warmed up on its test track. It will soon be shipped to Southern California, where it is slated to carry riders on San Bernardino County’s Arrow commuter rail service before the end of the year.

The best way to decarbonize railroads is the subject of growing debate among regulators, industry, and activists. The debate is partly technological, revolving around whether hydrogen fuel cells, batteries, or overhead electric wires offer the best performance for different railroad situations. But it’s also political: a question of the extent to which decarbonization can, or should, usher in a broader transformation of rail transportation.

In the insular world of railroading, this hydrogen-powered train is a Rorschach test. To some, it represents the future of rail transportation. To others, it looks like a big, shiny distraction. Read the full story.

—Benjamin Schneider

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or tweet ’em at me.)

+ The time and effort it must have taken to paint all 151 original Pokemon is mind-boggling.
+ You’ve heard of Stonehenge, but do you know Kilmartin Glen?
+ The Netherlands recently hosted a celebration for thousands of redheads.
+ Peter Blackert builds some seriously impressive Lego cars.

Ecommerce MGMT 0 Comments
Sep 5 2024
Charts: U.S. Consumer Outlook Q3 2024

McKinsey & Company conducted its latest U.S. “ConsumerWise” survey of roughly 4,000 participants from late July 2024 through the first week of August. According to the survey, U.S. consumer confidence in the economy improved, rising to 41% from 33% in the prior quarter.

–

The data shows that increased optimism has resulted in a higher but cautious inclination to spend in most categories.

–

“Trading down” is changing the type or quantity of purchases for better value and pricing. McKinsey’s survey found that trading down among U.S. consumers continued, as did delays in purchases and using buy-now-pay-later plans.

–

The McKinsey survey asked participants whether they intended to splurge (increase in spending lavishly) in Q3 2024.

Ecommerce MGMT 0 Comments
Sep 4 2024
How New Chrome AI Feature Challenges SEO To Evolve via @sejournal, @martinibuster

A Google Chrome Engineer published a LinkedIn post outlining the new Chrome AI History feature and the signals it uses to surface previously visited sites. The post illustrates that natural language browser history search could become a traffic source, and SEO must evolve in response.

History Search Powered By AI

Google recently announced a new opt-in feature in Chrome that gives users the benefit of AI to search through their browser history and find a page that they have previously visited. This makes it easier for a site that has previously been visited to obtain another visit from the same person.

Chrome AI History Searches Page Content

Chrome Engineering Leader Addy Osmani wrote a description of the new Chrome AI History feature that contained some undocumented information about how it works which shows how text and images are used as data sources for the AI to identify a site that a user had previously visited.

The Chrome Browser history normally just searches the URL and Page Title to find something in the search history. “History Search, powered by AI” looks at the webpage content, including the images.

Osmani shared an example where he identified a page he had previously visited in which the AI used image content to find what he was looking for.

He gave an example of finding a page he visited that’s related to shopping:

“Recently, I was browsing for a new sweater and took a look at a few options across a few sites. I saw some neat Burberry designs. But there was one specific Burberry sweater I liked from a while back that said “England” on it. I can’t remember where I saw it or how to find that page again.

With AI history search, I simply type “Burberry sweater England” and voilà – the exact page appears, even though “England” was only mentioned within an image on the site.”

What does he mean that the word “England” was only mentioned in an image? He doesn’t specifically say that the word was in the image meta data like in the alt tag. I assumed that’s what he meant, that the word England was in the image metadata. So I found the exact page he was looking at (it’s in a video he embedded in his LinkedIn post) and checked the source code and the word “England” was not in the meta data.

If you watch the video the AI Browser history shows multiple pages so it’s possible that the AI simply ignored the word “England” just surfaced everything that had a partial match. But, Osmani said it was surfaced because of the image.

Here’s a screenshot from his video:

Screenshot of a page surfaced by Chrome AI Browser History result

Here’s the AI search results showing multiple pages in the results:

Screenshot of Chrome AI Browser history

The above image shows that the AI history surfaced more than just one page and the other pages weren’t about a shirt that said England, only the one. So it could very well be that the AI history was surfacing the England page not because it had the word England in the image but because it was relevant for the words Burberry and Sweater. But again, it could be because the word was in the image, this is something that needs clarification.

Osmani then offers two more examples that show how using keywords that appear in the page content will help surface web pages that a user had previously visited.

AI Browser Search Documentation

Google maintains a help page dedicated to this new feature where it lists the following tips that also give more information about how the AI browser search works.

  • “When you search short and simple text, you’ll be matched directly to the page title or URL. You won’t find an AI-powered result.
  • You can rate the best match result. At the bottom of the best match result, select Thumbs up or Thumbs down .
  • If you select Thumbs down , you can provide additional feedback on why the result didn’t meet your needs.
  • You can also search for browsing history in the address bar.”

Takeaways

Chrome AI search enables repeat visitors through natural language searches. But when users search with simple text Chrome will default to simple keyword matching to the page title and URL.

  • Exact keywords are not necessary
  • URLs are not necessary
  • Short simple text is matched via Title tag and URL
  • Keywords in title tag and URL that match to how users will remember the site (the topic) can still be important
  • The ability to rate results shows that this feature will continue to evolve

Chrome AI History is a useful feature and will likely become more prominent as people become more aware of it and people become more accustomed to using AI that’s built into their browsers and devices. This doesn’t mean it will become useful to add keywords all over the meta data but it does show how the future of SEO is growing to accommodate more than just search as AI takes a greater role in surfacing web pages.

Featured Image by Shutterstock/Cast Of Thousands

Ecommerce MGMT 0 Comments
Sep 4 2024
WordPress Just Locked Down Security For All Plugins & Themes via @sejournal, @martinibuster

WordPress announced a major clampdown to protect its theme and plugin ecosystem from password insecurity. These improvements follow a flurry of attacks in June that compromised multiple plugins at the source.

Improves Plugin Developer Security

This WordPress security update fixes a flaw that allowed hackers to use compromised passwords from other breaches to unlock developer accounts that used the same credentials and had “commit access” enabling them to make changes to the plugin code right at the source. This closes a WordPress security gap that allowed hackers to compromise multiple plugins beginning in late June of this year.

Double Layer Of Developer Security

WordPress is introducing two layers of security, one on the individual developer account and a second one on the code commit access. This separates the author security credentials from the code committing environment.

1. Two-Factor Authorization

The first improvement to security is the imposition of a mandatory two-factor authorization for all plugin and theme authors that will be enforced beginning on October 1, 2024. WordPress is already prompting users to use 2FA. Users can also visit this page to configure their two-factor authorization.

2. SVN Passwords

WordPress also announced it will begin using SVN (Subversion) passwords, an additional layer of security for authenticating developers as a part of a version control system. SVN ensures that only authorized individuals can make changes to the code, adding a second layer of security to plugins and themes.

The WordPress announcement explains:

“We’ve introduced an SVN password feature to separate your commit access from your main WordPress.org account credentials. This password functions like an application or additional user account password. It protects your main password from exposure and allows you to easily revoke SVN access without having to change your WordPress.org credentials. Generate your SVN password in your WordPress.org profile.”

WordPress noted that technical limitations prevented them from using 2FA to existing code repositories, thereby requiring them to use SVN instead.

Takeaway: Vastly Improved WordPress Security

These changes will results in greater security for the entire WordPress ecosystem and immensely contribute to ensuring that all plugins and themes are trustworthy and not compromised at the source.

Read the announcement

Upcoming Security Changes for Plugin and Theme Authors on WordPress.org

Featured Image by Shutterstock/Cast Of Thousands

Ecommerce MGMT 0 Comments