Category: Science & Tech

On February 6, Meta said it was going to label AI-generated images on Facebook, Instagram, and Threads. When someone uses Meta’s AI tools to create images, the company will add visible markers to the image, as well as invisible watermarks and metadata in the image file. The company says its standards are in line with best practices laid out by the Partnership on AI, an AI research nonprofit.

Big Tech is also throwing its weight behind a promising technical standard that could add a “nutrition label” to images, video, and audio. Called C2PA, it’s an open-source internet protocol that relies on cryptography to encode details about the origins of a piece of content, or what technologists refer to as “provenance” information. The developers of C2PA often compare the protocol to a nutrition label, but one that says where content came from and who—or what—created it. Read more about it here. 

On February 8, Google announced it is joining other tech giants such as Microsoft and Adobe in the steering committee of C2PA and will include its watermark SynthID in all AI-generated images in its new Gemini tools. Meta says it is also participating in C2PA. Having an industry-wide standard makes it easier for companies to detect AI-generated content, no matter which system it was created with.

OpenAI too announced new content provenance measures last week. It says it will add watermarks to the metadata of images generated with ChatGPT and DALL-E 3, its image-making AI. OpenAI says it will now include a visible label in images to signal they have been created with AI. 

These methods are a promising start, but they’re not foolproof. Watermarks in metadata are easy to circumvent by taking a screenshot of images and just using that, while visual labels can be cropped or edited out. There is perhaps more hope for invisible watermarks like Google’s SynthID, which subtly changes the pixels of an image so that computer programs can detect the watermark but the human eye cannot. These are harder to tamper with. What’s more, there aren’t reliable ways to label and detect AI-generated video, audio, or even text. 

But there is still value in creating these provenance tools. As Henry Ajder, a generative-AI expert, told me a couple of weeks ago when I interviewed him about how to prevent deepfake porn, the point is to create a “perverse customer journey.” In other words, add barriers and friction to the deepfake pipeline in order to slow down the creation and sharing of harmful content as much as possible. A determined person will likely still be able to override these protections, but every little bit helps. 

There are also many nontechnical fixes tech companies could introduce to prevent problems such as deepfake porn. Major cloud service providers and app stores, such as Google, Amazon, Microsoft, and Apple could move to ban services that can be used to create nonconsensual deepfake nudes. And watermarks should be included in all AI-generated content across the board, even by smaller startups developing the technology.

What gives me hope is that alongside these voluntary measures we’re starting to see binding regulations, such as the EU’s AI Act and the Digital Services Act, which require tech companies to disclose AI-generated content and take down harmful content faster. There’s also renewed interest among US lawmakers in passing some binding rules on deepfakes. And following AI-generated robocalls of President Biden telling voters not to vote, the US Federal Communications Commission announced last week that it was banning the use of AI in these calls. 

Frogs can be hard to detect, and they are not, of course, the only species that eludes more traditional, boots-on-the-ground detection. Thomsen began work on another organism that notoriously confounds measurement: fish. Counting fish is sometimes said to vaguely resemble counting trees — except they’re free-roaming, in dark places, and fish counters are doing their tally while blindfolded. Environmental DNA dropped the blindfold. One review of published literature on the technology — though it came with caveats, including imperfect and imprecise detections or details on abundance — found that eDNA studies on freshwater and marine fish and amphibians outnumbered terrestrial counterparts 7:1.

In 2011, Thomsen, then a Ph.D. candidate in Willerslev’s lab, published a paper demonstrating that the method could detect rare and threatened species, such as those in low abundance in Europe, including amphibians, mammals like the otter, crustaceans, and dragonflies. “We showed that only, like, a shot glass of water really was enough to detect these organisms,” he told Undark. It was clear: The method had direct applications in conservation biology for the detection and monitoring of species.

In 2012, the journal Molecular Ecology published a special issue on eDNA, and Taberlet and several colleagues outlined a working definition of eDNA as any DNA isolated from environmental samples. The method described two similar but slightly different approaches: One can answer a yes or no question: Is the bullfrog (or whatever) present or not? It does so by scanning the metaphoric barcode, short sequences of DNA that are particular to a species or family, called primers; the checkout scanner is a common technique called quantitative real-time polymerase chain reaction, or qPCR.

Scientists use eDNA to track creatures of all shapes and sizes, be it tiny bits of invasive algae, eels in Loch Ness, or a sightless sand-dwelling mole that hasn’t been seen in nearly 90 years.

Another approach, commonly known as DNA metabarcoding, essentially spits out a list of organisms present in a given sample. “You sort of ask the question, what is here?” Thomsen said. “And then you get all of the known things, but you also get some surprises, right? Because there were some species that you didn’t know were actually present.”

One aims to find the needle in a haystack; the other attempts to reveal the whole haystack. eDNA differs from more traditional sampling techniques where organisms, like fish, are caught, manipulated, stressed, and sometimes killed. The data obtained are objective; it’s standardized and unbiased.

“eDNA, one way or the other, is going to stay as one of the important methodologies in biological sciences,” said Mehrdad Hajibabaei, a molecular biologist at University of Guelph, who pioneered the metabarcoding approach, and who traced fish some 9,800 feet under the Labrador Sea. “Every day I see something bubbling up that didn’t occur to me.”

In recent years, the field of eDNA has expanded. The method’s sensitivity allows researchers to sample previously out-of-reach environments, for example, capturing eDNA from the air — an approach that highlights eDNA’s promises and its potential pitfalls. Airborne eDNA appears to circulate on a global dust belt, suggesting its abundance and omnipresence, and it can be filtered and analyzed to monitor plants and terrestrial animals. But eDNA blowing in the wind can lead to inadvertent contamination.

In 2019, Thomsen, for instance, left two bottles of ultra-pure water out in the open — one in a grassland, and the other near a marine harbor. After a few hours, the water contained detectable eDNA associated with birds and herring, suggesting that traces of non-terrestrial species settled into the samples; the organisms obviously did not inhabit the bottles. “So it must come from the air,” Thomsen told Undark. The results suggest a two-fold problem: For one, trace evidence can move around, where two organisms that come into contact can then tote around the other’s DNA, and just because certain DNA is present doesn’t mean that the species is actually there.

Heat pumps have been around for decades, but the technology has been experiencing a clear moment in the sun in recent years, with global sales increasing by double digits in both 2021 and 2022, according to the International Energy Agency (IEA). Heat pumps were featured on MIT Technology Review’s 2024 list of 10 Breakthrough Technologies. 

Sales fell by nearly 17% in 2023 in the US, one of the technology’s largest markets, according to new data from the Air-Conditioning, Heating, and Refrigeration Institute. The slowdown comes after nearly a decade of constant growth. The AHRI data isn’t comprehensive, but the organization includes manufacturers accounting for about 90% of the units sold in the US annually.

However, the decline likely says less about heat pumps than it does about the whole HVAC sector, since gas furnaces and air conditioners saw even steeper drops. Gas furnace sales declined even more than heat pumps did in 2023, so heat pumps actually made up a slightly larger percentage of sales this year than in 2022.

The broad slowdown reflects broader consumer pessimism amid higher interest rates and inflation, says Yannick Monschauer, an analyst at the IEA, via email. 

“We have also been observing slowing heat pump sales in other parts of the world for 2023,” Monschauer adds. In Europe, a rush to electrify, driven by the energy crisis and rising natural gas prices, has slowed. 

Whether any of these strategies will fix the bias in pulse oximeters remains to be seen. But it’s likely that by the time improved devices are up for regulatory approval, the bar for performance will be higher. At the meeting last week, committee members reviewed a proposal that would require companies to test the device in at least 24 people whose skin tones span the entirety of a 10-shade scale. The current requirement is that the trial must include 10 people, two of whom have “darkly pigmented” skin.

In the meantime, health-care workers are grappling with how to use the existing tools and whether to trust them. In the advisory committee meeting on Friday, one committee member asked a representative from Medtronic, one of the largest providers of pulse oximeters, if the company had considered a voluntary recall of its devices. “We believe with 100% certainty that our devices conform to current FDA standards,” said Sam Ajizian, Medtronic’s chief medical officer of patient monitoring. A recall “would undermine public safety because this is a foundational device in operating rooms and ICUs, ERs, and ambulances and everywhere.”

But not everyone agrees that the benefits outweigh the harms. Last fall, a community health center in Oakland California, filed a lawsuit against some of the largest manufacturers and sellers of pulse oximeters, asking the court to prohibit sale of the devices in California until the readings are proved accurate for people with dark skin, or until the devices carry a warning label.

“The pulse oximeter is an example of the tragic harm that occurs when the nation’s health-care industry and the regulatory agencies that oversee it prioritize white health over the realities of non-white patients,” said Noha Aboelata, CEO of Roots Community Health Center, in a statement. “The story of the making, marketing and use of racially biased pulse oximeters is an indictment of our health-care system.”

Read more from MIT Technology Review’s archive

Melissa Heikkilä’s reporting showed her just how “pale, male, and stale” the humans of AI are. Could we just ask it to do better? 

No surprise that technology perpetuates racism, wrote Charlton McIlwain in 2020. That’s the way it was designed. “The question we have to confront is whether we will continue to design and deploy tools that serve the interests of racism and white supremacy.”

We’ve seen that deep-learning models can perform as well as medical professionals when it comes to imaging tasks, but they can also perpetuate biases. Some researchers say the way to fix the problem is to stop training algorithms to match the experts, reported Karen Hao in 2021. 

From around the web

The high lead levels found in applesauce pouches came from a single cinnamon processing plant in Ecuador. (NBC)

Finally, Google is launching a new subscription service called Gemini Advanced that extends its existing Google One Premium Plan (which gives users extra storage and a few other perks). For $19.99 a month, the new Google One Premium AI Plan will give you access to Google’s most powerful model, Gemini Ultra, for the first time as well as all the plan’s usual perks. This is very similar to OpenAI’s offering, where for $19.99 a month ChatGPT Premium buys you access to GPT-4 rather than GPT-3.5.

When can you get it? The free Gemini app (powered by Gemini Pro) is available from today in English in the US. Starting next week, you’ll be able to access it across the Asia Pacific region in English and in Japanese and Korean. But there is no word on when the app will come to the UK, EU countries or Switzerland.

Gemini Advanced (the paid-for service that gives access to Gemini Ultra) is available in English in more than 150 countries, including the UK and EU (but not France). Google says it is analyzing local requirements and fine-tuning Gemini for cultural nuance in different countries. But it claims that more languages and regions are coming.

What can you do with it? Google says it has developed its Gemini products with the help of more than 100 testers and power-users. At the press conference yesterday, Google execs outlined a handful of use cases, such as getting Gemini to help write a cover letter for a job application. “This can help you come across as more professional and increase your relevance to recruiters,” Kat Behr said

Or you could take a picture of your flat tire and ask Gemini how to fix it. A more elaborate example involved Gemini managing a snack-rota for the parents of kids on a soccer team. Gemini would come up with the rota of who brought snacks and when; help you to email other parents and field their replies. In future versions, Gemini will be able to draw on data in your Google Drive that could help manage carpooling around game schedules, Behr said.   

But we should expect users to find a lot more uses for these tools. “I’m really excited to see how people around the world are going to push the envelope on this AI,” Hsaio said.

Is it safe? Google has been working hard to make sure its slick AI-powered products are safe to use. But no amount of testing can anticipate all the ways that tech will get used and misused once it is released. In the last few months, Meta saw its image-making app used to produce pictures of Mickey Mouse with guns and Spongebob Squarepants flying a jet into two towers. Microsoft’s image-making software was used to create fake pornographic images of Taylor Swift.

The AI Act aims to mitigate some—but not all—these problems. For example, it requires the makers of powerful AI like Gemini to build in safeguards, such as watermarking for generated images and steps to avoid reproducing copyrighted material. Google says that all images generated by its products will include its SynthID watermarks. 

Like most companies, Google was knocked onto the back foot when ChatGPT arrived. Microsoft’s partnership with OpenAI has given it a boost over its old rival. But with Gemini, Google has come back strong: this is the slickest packaging of this generation’s tech yet. 

“The biggest misconception that people have is that sustainability is about carbon emissions,” says Pablo Orvananos, global sustainability consulting lead at Hitachi Digital Services. “That’s what we call carbon tunnel vision. Sustainability is much more than carbon. It’s a plethora of environmental issues and social issues, and companies need to focus on all of it.”

Companies looking to act will find a great deal of complexity surrounding corporate sustainability efforts. Companies are responsible not only for their own emissions and fossil fuels usage (Scope 1), but also the sustainability efforts of their energy suppliers (Scope 2) and their supply chain partners (Scope 3). New regulations require organizations to look beyond just emissions. Companies must ask questions about a broad range of environmental and societal issues: Are supply chain partners sourcing raw materials in an environmentally conscious manner? Are they treating workers fairly?

Sustainability can’t be siloed into one specific task, such as decarbonizing the data center. The only way to achieve sustainability is with a comprehensive, holistic approach, says Daniel Versace, an ESG research analyst at IDC. “A siloed approach to ESG is an approach that’s bound to fail,” he adds.

This content was produced by Insights, the custom content arm of MIT Technology Review. It was not written by MIT Technology Review’s editorial staff.

One significant difference is VPPs’ ability to shape consumers’ energy use in real time. Unlike conventional power plants, VPPs can communicate with distributed energy resources and allow grid operators to control the demand from end users.

For example, smart thermostats linked to air conditioning units can adjust home temperatures and manage how much electricity the units consume. On hot summer days these thermostats can pre-cool homes before peak hours, when air conditioning usage surges. Staggering cooling times can help prevent abrupt demand hikes that might overwhelm the grid and cause outages. Similarly, electric vehicle chargers can adapt to the grid’s requirements by either supplying or utilizing electricity. 

These distributed energy sources connect to the grid through communication technologies like Wi-Fi, Bluetooth, and cellular services. In aggregate, adding VPPs can increase overall system resilience. By coordinating hundreds of thousands of devices, VPPs have a meaningful impact on the grid—they shape demand, supply power, and keep the electricity flowing reliably.

How popular are VPPs now?

Until recently, VPPs were mostly used to control consumer energy use. But because solar and battery technology has evolved, utilities can now use them to supply electricity back to the grid when needed.

In the United States, the Department of Energy estimates VPP capacity at around 30 to 60 gigawatts. This represents about 4% to 8% of peak electricity demand nationwide, a minor fraction within the overall system. However, some states and utility companies are moving quickly to add more VPPs to their grids.

Green Mountain Power, Vermont’s largest utility company, made headlines last year when it expanded its subsidized home battery program. Customers have the option to lease a Tesla home battery at a discounted rate or purchase their own, receiving assistance of up to $10,500, if they agree to share stored energy with the utility as required. The Vermont Public Utility Commission, which approved the program, said it can also provide emergency power during outages.

In Massachusetts, three utility companies (National Grid, Eversource, and Cape Light Compact) have implemented a VPP program that pays customers in exchange for utility control of their home batteries.

Meanwhile, in Colorado efforts are underway to launch the state’s first VPP system. The Colorado Public Utilities Commission is urging Xcel Energy, its largest utility company, to develop a fully operational VPP pilot by this summer.