The question of who has standing to fight the president in court could be clarified in a case over California's ability to regulate car pollution.

A Pentagon effort to develop a hybrid-electric system for Bradley fighting vehicles began during Trump's first administration.

The request from South Carolina Rep. Joe Wilson comes as JBS prepares to join the New York Stock Exchange.

Advocates are urging the state to end a policy that waives penalties for oil and gas companies. Industry warns the move would raise gas prices.

Democrats postponed hearings on a pair of bills to assess climate damages on oil companies.

As natural disasters become more frequent, the U.S. is bearing the brunt of insured losses, said Swiss Re.

The study challenges the notion that the rise of antimicrobial resistance is solely due to the overconsumption of antibiotics.

The fires over the weekend in South Africa may have been started deliberately.

The plot being sold on Norway’s archipelago of Svalbard is the size of Manhattan; has 3 miles of coastline; and is full of mountains, fjords and wildlife.

In the future, quantum computers could rapidly simulate new materials or help scientists develop faster machine-learning models, opening the door to many new possibilities.

But these applications will only be possible if quantum computers can perform operations extremely quickly, so scientists can make measurements and perform corrections before compounding error rates reduce their accuracy and reliability.

The efficiency of this measurement process, known as readout, relies on the strength of the coupling between photons, which are particles of light that carry quantum information, and artificial atoms, units of matter that are often used to store information in a quantum computer.

Now, MIT researchers have demonstrated what they believe is the strongest nonlinear light-matter coupling ever achieved in a quantum system. Their experiment is a step toward realizing quantum operations and readout that could be performed in a few nanoseconds.

The researchers used a novel superconducting circuit architecture to show nonlinear light-matter coupling that is about an order of magnitude stronger than prior demonstrations, which could enable a quantum processor to run about 10 times faster.

There is still much work to be done before the architecture could be used in a real quantum computer, but demonstrating the fundamental physics behind the process is a major step in the right direction, says Yufeng “Bright” Ye SM ’20, PhD ’24, lead author of a paper on this research.

“This would really eliminate one of the bottlenecks in quantum computing. Usually, you have to measure the results of your computations in between rounds of error correction. This could accelerate how quickly we can reach the fault-tolerant quantum computing stage and be able to get real-world applications and value out of our quantum computers,” says Ye.

He is joined on the paper by senior author Kevin O’Brien, an associate professor and principal investigator in the Research Laboratory of Electronics at MIT who leads the Quantum Coherent Electronics Group in the Department of Electrical Engineering and Computer Science (EECS), as well as others at MIT, MIT Lincoln Laboratory, and Harvard University. The research appears today in Nature Communications.

A new coupler

This physical demonstration builds on years of theoretical research in the O’Brien group.

After Ye joined the lab as a PhD student in 2019, he began developing a specialized photon detector to enhance quantum information processing.

Through that work, he invented a new type of quantum coupler, which is a device that facilitates interactions between qubits. Qubits are the building blocks of a quantum computer. This so-called quarton coupler had so many potential applications in quantum operations and readout that it quickly became a focus of the lab.

This quarton coupler is a special type of superconducting circuit that has the potential to generate extremely strong nonlinear coupling, which is essential for running most quantum algorithms. As the researchers feed more current into the coupler, it creates an even stronger nonlinear interaction. In this sense, nonlinearity means a system behaves in a way that is greater than the sum of its parts, exhibiting more complex properties.

“Most of the useful interactions in quantum computing come from nonlinear coupling of light and matter. If you can get a more versatile range of different types of coupling, and increase the coupling strength, then you can essentially increase the processing speed of the quantum computer,” Ye explains.

For quantum readout, researchers shine microwave light onto a qubit and then, depending on whether that qubit is in state 0 or 1, there is a frequency shift on its associated readout resonator. They measure this shift to determine the qubit’s state.

Nonlinear light-matter coupling between the qubit and resonator enables this measurement process.

The MIT researchers designed an architecture with a quarton coupler connected to two superconducting qubits on a chip. They turn one qubit into a resonator and use the other qubit as an artificial atom which stores quantum information. This information is transferred in the form of microwave light particles called photons.

“The interaction between these superconducting artificial atoms and the microwave light that routes the signal is basically how an entire superconducting quantum computer is built,” Ye explains.

Enabling faster readout

The quarton coupler creates nonlinear light-matter coupling between the qubit and resonator that’s about an order of magnitude stronger than researchers had achieved before. This could enable a quantum system with lightning-fast readout.

“This work is not the end of the story. This is the fundamental physics demonstration, but there is work going on in the group now to realize really fast readout,” O’Brien says.

That would involve adding additional electronic components, such as filters, to produce a readout circuit that could be incorporated into a larger quantum system.

The researchers also demonstrated extremely strong matter-matter coupling, another type of qubit interaction that is important for quantum operations. This is another area they plan to explore with future work.

Fast operations and readout are especially important for quantum computers because qubits have finite lifespans, a concept known as coherence time.

Stronger nonlinear coupling enables a quantum processor to run faster and with lower error, so the qubits can perform more operations in the same amount of time. This means the qubits can run more rounds of error correction during their lifespans.

“The more runs of error correction you can get in, the lower the error will be in the results,” Ye says.

In the long run, this work could help scientists build a fault-tolerant quantum computer, which is essential for practical, large-scale quantum computation.

This research was supported, in part, by the Army Research Office, the AWS Center for Quantum Computing, and the MIT Center for Quantum Engineering.

Newly published results of a randomized, controlled clinical trial in Japan among more than 170 children aged 1 to 6 who underwent surgery show that by using electroencephalogram (EEG) readings of brain waves to monitor unconsciousness, an anesthesiologist can significantly reduce the amount of the anesthesia administered to safely induce and sustain each patient’s anesthetized state. On average, the little patients experienced significant improvements in several post-operative outcomes, including quicker recovery and reduced incidence of delirium.

“I think the main takeaway is that in kids, using the EEG, we can reduce the amount of anesthesia we give them and maintain the same level of unconsciousness,” says study co-author Emery N. Brown, the Edward Hood Taplin Professor of Medical Engineering and Computational Neuroscience at MIT, an anesthesiologist at Massachusetts General Hospital, and a professor at Harvard Medical School. The study appeared April 21 in JAMA Pediatrics.

Yasuko Nagasaka, chair of anesthesiology at Tokyo Women’s Medical University and a former colleague of Brown’s in the United States, designed the study. She asked Brown to train and advise lead author Kiyoyuki Miyasaka of St. Luke’s International Hospital in Tokyo on how to use EEG to monitor unconsciousness and adjust anesthesia dosing in children. Miyasaka then served as the anesthesiologist for all patients in the trial. Attending anesthesiologists not involved in the study were always on hand to supervise.

Brown’s research in The Picower Institute for Learning and Memory, the Institute for Medical Engineering and Science, and the Department of Brain and Cognitive Sciences at MIT has shown that a person’s level of consciousness under any particular anesthetic drug is discernible from patterns of their brain waves. Each child’s brain waves were measured with EEG, but in the control group Miyasaka adhered to standard anesthesia dosing protocols while in the experimental group he used the EEG measures as a guide for dosing. The results show that when he used EEG, he was able to induce the desired level of unconsciousness with a concentration of 2 percent sevoflurane gas, rather than the standard 5 percent. Maintenance of unconsciousness, meanwhile, only turned out to require 0.9 percent concentration, rather than the standard 2.5 percent.

Meanwhile, a separate researcher, blinded to whether EEG or standard protocols were used, assessed the kids for “pediatric anesthesia emergence delirium” (PAED), in which children sometimes wake up from anesthesia with a set of side effects including lack of eye contact, inconsolability, unawareness of surroundings, restlessness, and non-purposeful movements. Children who received standard anesthesia dosing met the threshold for PAED in 35 percent of cases (30 out of 86), while children who received EEG-guided dosing met the threshold in 21 percent of cases (19 out of 91). The difference of 14 percentage points was statistically significant.

Meanwhile, the authors reported that, on average, EEG-guided patients had breathing tubes removed 3.3 minutes earlier, emerged from anesthesia 21.4 minutes earlier, and were discharged from post-acute care 16.5 minutes earlier than patients who received anesthesia according to the standard protocol. All of these differences were statistically significant. Also, no child in the study ever became aware during surgery.

The authors noted that the quicker recovery among patients who received EEG-guided anesthesia was not only better medically, but also reduced health-care costs. Time in post-acute care in the United States costs about $46 a minute, so the average reduced time of 16.5 minutes would save about $750 per case. Sevoflurane is also a potent greenhouse gas, Brown notes, so reducing its use is better for the environment.

In the study, the authors also present comparisons of the EEG recordings from children in the control and experimental groups. There are notable differences in the “spectrograms” that charted the power of individual brain wave frequencies both as children were undergoing surgery and while they were approaching emergence from anesthesia, Brown says.

For instance, among children who received EEG-guided dosing, there are well-defined bands of high power at about 1-3 Hertz and 10-12 Hz. In children who received standard protocol dosing, the entire range of frequencies up to about 15 Hz are at high power. In another example, children who experienced PAED showed higher power at several frequencies up to 30Hz than children who did not experience PAED.

The findings further validate the idea that monitoring brain waves during surgery can provide anesthesiologists with actionable guidance to improve patient care, Brown says. Training in reading EEGs and guiding dosing can readily be integrated in the continuing medical education practices of hospitals, he adds.

In addition to Miyasuka, Brown, and Nagasaka, Yasuyuki Suzuki is a study co-author.

Funding sources for the study include the MIT-Massachusetts General Brigham Brain Arousal State Control Innovation Center, the Freedom Together Foundation, and the Picower Institute.

For more than 30 years, Course 7 (Biology) students have descended to the expansive, windowless basement of Building 68 to learn practical skills that are the centerpiece of undergraduate biology education at the Institute. The lines of benches and cabinets of supplies that make up the underground MIT Biology Teaching Lab could easily feel dark and isolated. 

In the corner of this room, however, sits Senior Technical Instructor Vanessa Cheung ’02, who manages to make the space seem sunny and communal.

“We joke that we could rig up a system of mirrors to get just enough daylight to bounce down from the stairwell,” Cheung says with a laugh. “It is a basement, but I am very lucky to have this teaching lab space. It is huge and has everything we need.”

This optimism and gratitude fostered by Cheung is critical, as MIT undergrad students enrolled in classes 7.002 (Fundamentals of Experimental Molecular Biology) and 7.003 (Applied Molecular Biology Laboratory) spend four-hour blocks in the lab each week, learning the foundations of laboratory technique and theory for biological research from Cheung and her colleagues.

Running toward science education

Cheung’s love for biology can be traced back to her high school cross country and track coach, who also served as her second-year biology teacher. The sport and the fundamental biological processes she was learning about in the classroom were, in fact, closely intertwined. 

“He told us about how things like ATP [adenosine triphosphate] and the energy cycle would affect our running,” she says. “Being able to see that connection really helped my interest in the subject.”

That inspiration carried her through a move from her hometown of Pittsburgh, Pennsylvania, to Cambridge, Massachusetts, to pursue an undergraduate degree at MIT, and through her thesis work to earn a PhD in genetics at Harvard Medical School. She didn’t leave running behind either: To this day, she can often be found on the Charles River Esplanade, training for her next marathon. 

She discovered her love of teaching during her PhD program. She enjoyed guiding students so much that she spent an extra semester as a teaching assistant, outside of the one required for her program. 

“I love research, but I also really love telling people about research,” Cheung says.

Cheung herself describes lab instruction as the “best of both worlds,” enabling her to pursue her love of teaching while spending every day at the bench, doing experiments. She emphasizes for students the importance of being able not just to do the hands-on technical lab work, but also to understand the theory behind it.

“The students can tend to get hung up on the physical doing of things — they are really concerned when their experiments don’t work,” she says. “We focus on teaching students how to think about being in a lab — how to design an experiment and how to analyze the data.”

Although her talent for teaching and passion for science led her to the role, Cheung doesn’t hesitate to identify the students as her favorite part of the job. 

“It sounds cheesy, but they really do keep the job very exciting,” she says.

Using mind and hand in the lab

Cheung is the type of person who lights up when describing how much she “loves working with yeast.” 

“I always tell the students that maybe no one cares about yeast except me and like three other people in the world, but it is a model organism that we can use to apply what we learn to humans,” Cheung explains.

Though mastering basic lab skills can make hands-on laboratory courses feel “a bit cookbook,” Cheung is able to get the students excited with her enthusiasm and clever curriculum design. 

“The students like things where they can get their own unique results, and things where they have a little bit of freedom to design their own experiments,” she says. So, the lab curriculum incorporates opportunities for students to do things like identify their own unique yeast mutants and design their own questions to test in a chemical engineering module.

Part of what makes theory as critical as technique is that new tools and discoveries are made frequently in biology, especially at MIT. For example, there has been a shift from a focus on RNAi to CRISPR as a popular lab technique in recent years, and Cheung muses that CRISPR itself may be overshadowed within only a few more years — keeping students learning at the cutting edge of biology is always on Cheung’s mind. 

“Vanessa is the heart, soul, and mind of the biology lab courses here at MIT, embodying ‘mens et manus’ [‘mind and hand’],” says technical lab instructor and Biology Teaching Lab Manager Anthony Fuccione. 

Support for all students

Cheung’s ability to mentor and guide students earned her a School of Science Dean’s Education and Advising Award in 2012, but her focus isn’t solely on MIT undergraduate students. 

In fact, according to Cheung, the earlier students can be exposed to science, the better. In addition to her regular duties, Cheung also designs curriculum and teaches in the LEAH Knox Scholars Program. The two-year program provides lab experience and mentorship for low-income Boston- and Cambridge-area high school students. 

Paloma Sanchez-Jauregui, outreach programs coordinator who works with Cheung on the program, says Cheung has a standout “growth mindset” that students really appreciate.

“Vanessa teaches students that challenges — like unexpected PCR results — are part of the learning process,” Sanchez-Jauregui says. “Students feel comfortable approaching her for help troubleshooting experiments or exploring new topics.”

Cheung’s colleagues report that they admire not only her talents, but also her focus on supporting those around her. Technical Instructor and colleague Eric Chu says Cheung “offers a lot of help to me and others, including those outside of the department, but does not expect reciprocity.”

Professor of biology and co-director of the Department of Biology undergraduate program Adam Martin says he “rarely has to worry about what is going on in the teaching lab.” According to Martin, Cheung is ”flexible, hard-working, dedicated, and resilient, all while being kind and supportive to our students. She is a joy to work with.” 

The ocean’s deep-sea bed is scattered with ancient rocks, each about the size of a closed fist, called “polymetallic nodules.” Elsewhere, along active and inactive hydrothermal vents and the deep ocean’s ridges, volcanic arcs, and tectonic plate boundaries, and on the flanks of seamounts, lie other types of mineral-rich deposits containing high-demand minerals.

The minerals found in the deep ocean are used to manufacture products like the lithium-ion batteries used to power electric vehicles, cell phones, or solar cells. In some cases, the estimated resources of critical mineral deposits in parts of the abyssal ocean exceed global land-based reserves severalfold.

“Society wants electric-powered vehicles, solar cells for clean energy, but all of this requires resources,” says Thomas Peacock, professor of mechanical engineering at MIT, in a video discussing his research. “Land-based resources are getting depleted, or are more challenging to access. In parts of the ocean, there are much more of these resources than in land-based reserve. The question is: Can it be less impactful to mine some of these resources from the ocean, rather than from land?”

Deep-sea mining is a new frontier in mineral extraction, with potentially significant implications for industry and the global economy, and important environmental and societal considerations. Through research, scientists like Peacock study the impacts of deep-sea mining activity objectively and rigorously, and can bring evidence to bear on decision-making. 

Mining activities, whether on land or at sea, can have significant impacts on the environment at local, regional, and global scales. As interest in deep-seabed mining is increasing, driven by the surging demand for critical minerals, scientific inquiries help illuminate the trade-offs.

Peacock has long studied the potential impacts of deep-sea mining in a region of the Pacific Ocean known as the Clarion Clipperton Zone (CCZ), where polymetallic nodules abound. A decade ago, his research group began studying deep-sea mining, seeing a critical need to develop monitoring and modeling capabilities for assessing the scale of impact.

Today, his MIT Environmental Dynamics Laboratory (ENDLab) is at the forefront of advancing understanding for emerging ocean utilization technologies. With research anchored in fundamental fluid dynamics, the team is developing cutting-edge monitoring programs, novel sensors, and modeling tools.

“We are studying the form of suspended sediment from deep sea mining operations, testing a new sensor for sediment and another new sensor for turbulence, studying the initial phases of the sediment plume development, and analyzing data from the 2021 and 2022 technology trials in the Pacific Ocean,” he explains.

In deep-sea nodule mining, vehicles collect nodules from the ocean floor and convey them back to a vessel above. After the critical materials are collected on the vessel, some leftover sediment may be returned to the deep-water column. The resulting sediment plumes, and their potential impacts, are a key focus of the team’s work.

A 2022 study conducted in the CCZ investigated the dynamics of sediment plumes near a deep-seabed polymetallic nodule mining vehicle. The experiments reveal most of the released sediment-laden water, between 92 and 98 percent, stayed close to the sea-bed floor, spreading laterally. The results suggest that turbidity current dynamics set the fraction of sediment that remains suspended in the water, along with the scale of the subsequent ambient sediment plume. The implications of the process, which had been previously overlooked, are substantial for plume modeling and informative for environmental impact statements.

“New model breakthroughs can help us make increasingly trustworthy predictions,” he says. The team also contributed to a recent study, published in the journal Nature, which showed that sediment deposited away from a test mining site gets cleared away, most likely by ocean currents, and reported on any observed biological recovery.

Researchers observed a site four decades after a nodule test mining experiment. Although biological impacts in many groups of organisms were present, populations of several organisms, including sediment macrofauna, mobile deposit feeders, and even large-sized sessile fauna, had begun to reestablish despite persistent physical changes at the seafloor. The study was led by the National Oceanography Centre in the U.K.

“A great deal has been learned about the fluid mechanics of deep-sea mining, in particular when it comes to deep-sea mining sediment plumes,” says Peacock, adding that the scientific progress continues with more results on the way. The work is setting new standards for in-situ monitoring of suspended sediment properties, and for how to interpret field data from recent technical trials.

Whether you are a person about town or a worm in a dish, life can throw all kinds of circumstances your way. What you need is a nervous system flexible enough to cope. In a new study, MIT neuroscientists show how even a simple animal can repurpose brain circuits and the chemical signals, or “neuromodulators,” in its brain to muster an adaptive response to an infection. The study therefore may provide a model for understanding how brains in more complex organisms, including ourselves, manage to use what they have to cope with shifting internal states. 

“Neuromodulators play pivotal roles in coupling changes in animals’ internal states to their behavior,” the scientists write in their paper, recently published in Nature Communications. “How combinations of neuromodulators released from different neuronal sources control the diverse internal states that animals exhibit remains an open question.”

When C. elegans worms fed on infectious Pseudomonas bacteria, they ate less and became more lethargic. When the researchers looked across the nervous system to see how that behavior happened, they discovered that the worm had completely revamped the roles of several of its 302 neurons and some of the peptides they secrete across the brain to modulate behavior. Systems that responded to stress in one case or satiety in another became reconfigured to cope with the infection.

“This is a question of, how do you adapt to your environment with the highest level of flexibility given the set of neurons and neuromodulators you have,” says postdoc Sreeparna Pradhan, co-lead author of the new study in Nature Communications. “How do you make the maximum set of options available to you?”

The research to find out took place in the lab of senior author Steve Flavell, an associate professor in The Picower Institute for Learning and Memory and the Department of Brain and Cognitive Sciences and an investigator of the Howard Hughes Medical Institute. Pradhan, who was supported by a fellowship from MIT’s K. Lisa Yang Brain-Body Center during the work, teamed up with former Flavell Lab graduate student Gurrein Madan to lead the research.

Pradhan says the team discovered several surprises in the course of the study, including that a neuropeptide called FLP-13 completely flipped its function in infected animals versus animals experiencing other forms of stress. Previous research had shown that when worms are stressed by heat, a neuron called ALA releases FLP-13 to cause the worms to go into quiescence, a sleep-like state. But when the worms in the new study ate Pseudomonas bacteria, a band of other neurons released FLP-13 to fight off quiescence, enabling the worms to survive longer. Meanwhile, ALA took on a completely different role during sickness: leading the charge to suppress feeding by emitting a different group of peptides.

A comprehensive approach

To understand how the worms responded to infection, the team tracked many features of the worms’ behavior for days and made genetic manipulations to probe the underlying mechanisms at play. They also recorded activity across the worms' whole brains. This kind of a comprehensive observation and experimentation is difficult to achieve in more complex animals, but C. elegans’ relative simplicity makes it a tractable testbed, Pradhan says. The team’s approach also is what allowed it to make so many unexpected findings.

For instance, Pradhan didn’t suspect that the ALA neuron would turn out to be the neuron that suppressed feeding, but when she observed their behavior for long enough, she started to realize the reduced feeding arose from the worms taking little breaks that they wouldn’t normally take. As she and Madan were manipulating more than a dozen genes they thought might be affecting behavior and feeding in the worm, she included another called ceh-17 that she had read about years ago that seemed to promote bouts of “microsleep” in the worms. When they knocked out ceh-17, they found that those worms didn’t reduce feeding when they got infected, unlike normal animals. It just so happens that ceh-17 is specifically needed for ALA to function properly, so that’s when the team realized ALA might be involved in the feeding-reduction behavior.

To know for sure, they then knocked out the various peptides that ALA releases and saw that when they knocked out three in particular, flp-24, nlp-8 and flp-7, infected worms didn’t exhibit reduced feeding upon infection. That clinched that ALA drives the reduced feeding behavior by emitting those three peptides.

Meanwhile, Pradhan and Madan’s screens also revealed that when infected worms were missing flp-13, they would go into a quiescence state much sooner than infected worms with the peptide available. Notably, the worms that fought off the quiescence state lived longer. They found that fighting off quiescence depended on the FLP-13 coming from four neurons (I5, I1, ASH and OLL), but not from ALA. Further experiments showed that FLP-13 acted on a widespread neuropeptide receptor called DMSR-1 to prevent quiescence.

Having a little nap

The last major surprise of the study was that the quiescence that Pseudomonas infection induces in worms is not the same as other forms of sleepiness that show up in other contexts, such as after satiety or heat stress. In those cases, worms don’t wake easily (with a little poke), but amid infection their quiescence was readily reversible. It seemed more like lethargy than sleep. Using the lab’s ability image all neural activity during behavior, Pradhan and Madan discerned that a neuron called ASI was particularly active during the bouts of lethargy. That observation solidified further when they showed that ASI’s secretion of the peptide DAF-7 was required for the quiescence to emerge in infected animals.

In all, the study showed that the worms repurpose and reconfigure — sometimes to the point of completely reversing — the functions of neurons and peptides to mount an adaptive response to infection, versus a different problem like stress. The results therefore shed light on what has been a tricky question to resolve. How do brains use their repertoire of cells, circuits, and neuromodulators to deal with what life hands them? At least part of the answer seems to be by reshuffling existing components, rather than creating unique ones for each situation.

“The states of stress, satiety, and infection are not induced by unique sets of neuromodulators," the authors wrote in their paper. "Instead, one larger set of neuromodulators may be deployed from different sources and in different combinations to specify these different internal states.”

In addition to Pradhan, Madan, and Flavell, the paper’s other authors are Di Kang, Eric Bueno, Adam Atanas, Talya Kramer, Ugur Dag, Jessica Lage, Matthew Gomes, Alicia Kun-Yang Lu, and Jungyeon Park.

Support for the research came from the the Picower Institute, the Freedom Together Foundation, the K. Lisa Yang Brain-Body Center, and the Yang Tan Collective at MIT; the National Institutes of Health; the McKnight Foundation; the Alfred P. Sloan Foundation; and the Howard Hughes Medical Institute.

When farmers apply pesticides to their crops, 30 to 50 percent of the chemicals end up in the air or soil instead of on the plants. Now, a team of researchers from MIT and Singapore has developed a much more precise way to deliver substances to plants: tiny needles made of silk.

In a study published today in Nature Nanotechnology, the researchers developed a way to produce large amounts of these hollow silk microneedles. They used them to inject agrochemicals and nutrients into plants, and to monitor their health.

“There’s a big need to make agriculture more efficient,” says Benedetto Marelli, the study’s senior author and an associate professor of civil and environmental engineering at MIT. “Agrochemicals are important for supporting our food system, but they’re also expensive and bring environmental side effects, so there’s a big need to deliver them precisely.”

Yunteng Cao PhD ’22, currently a postdoc Yale University, and Doyoon Kim, a former postdoc in the Marelli lab, led the study, which included a collaboration with the Disruptive and Sustainable Technologies for Agricultural Precision (DiSTAP) interdisciplinary research group at the Singapore-MIT Alliance for Research and Technology (SMART).

In demonstrations, the team used the technique to give plants iron to treat a disease known as chlorosis, and to add vitamin B12 to tomato plants to make them more nutritious. The researchers also showed the microneedles could be used to monitor the quality of fluids flowing into plants and to detect when the surrounding soil contained heavy metals.

Overall, the researchers believe the microneedles could serve as a new kind of plant interface for real-time health monitoring and biofortification.

“These microneedles could be a tool for plant scientists so they can understand more about plant health and how they grow,” Marelli says. “But they can also be used to add value to crops, making them more resilient and possibly even increasing yields.”

The inner workings of plants

Accessing the inner tissues of living plants requires scientists to get through the plants’ waxy skin without causing too much stress. In previous work, the researchers used silk-based microneedles to deliver agrochemicals to plants in lab environments and to detect pH changes in living plants. But these initial efforts involved small payloads, limiting their applications in commercial agriculture.

“Microneedles were originally developed for the delivery of vaccines or other drugs in humans,” Marelli explains. “Now we’ve adapted it so that the technology can work with plants, but initially we could not deliver sufficient doses of agrochemicals and nutrients to mitigate stressors or enhance crop nutritional values.”

Hollow structures could increase the amount of chemicals microneedles can deliver, but Marelli says creating those structures at scale has historically required clean rooms and expensive facilities like the ones found inside the MIT.nano building.

For this study, Cao and Kim created a new way to manufacture hollow silk microneedles by combining silk fibroin protein with a salty solution inside tiny, cone-shaped molds. As water evaporated from the solution, the silk solidified into the mold while the salt forms crystalline structures inside the molds. When the salt was removed, it left behind in each needle a hollow structure or tiny pores, depending on the salt concentration and the separation of the organic and inorganic phases.

“It’s a pretty simple fabrication process. It can be done outside of a clean room — you could do it in your kitchen if you wanted,” Kim says. “It doesn’t require any expensive machinery.”

The researchers then tested their microneedles’ ability to deliver iron to iron-deficient tomato plants, which can cause a disease known as chlorosis. Chlorosis can decrease yields, but treating it by spraying crops is inefficient and can have environmental side effects. The researchers showed that their hollow microneedles could be used for the sustained delivery of iron without harming the plants.

The researchers also showed their microneedles could be used to fortify crops while they grow. Historically, crop fortification efforts have focused on minerals like zinc or iron, with vitamins only added after the food is harvested.

In each case, the researchers applied the microneedles to the stalks of plants by hand, but Marelli envisions equipping autonomous vehicles and other equipment already used in farms to automate and scale the process.

As part of the study, the researchers used microneedles to deliver vitamin B12, which is primarily found naturally in animal products, into the stalks of growing tomatoes, showing that vitamin B12 moved into the tomato fruits before harvest. The researchers propose their method could be used to fortify more plants with the vitamin.

Co-author Daisuke Urano, a plant scientist with DiSTAP, explains that “through a comprehensive assessment, we showed minimal adverse effects from microneedle injections in plants, with no observed short- or long-term negative impacts.”

“This new delivery mechanism opens up a lot of potential applications, so we wanted to do something nobody had done before,” Marelli explains.

Finally, the researchers explored the use of their microneedles to monitor the health of plants by studying tomatoes growing in hydroponic solutions contaminated with cadmium, a toxic metal commonly found in farms close to industrial and mining sites. They showed their microneedles absorbed the toxin within 15 minutes of being injected into the tomato stalks, offering a path to rapid detection.

Current advanced techniques for monitoring plant health, such as colorimetric and hyperspectral lead analyses, can only detect problems after plants growth is already being stunted. Other methods, such as sap sampling, can be too time-consuming.

Microneedles, in contrast, could be used to more easily collect sap for ongoing chemical analysis. For instance, the researchers showed they could monitor cadmium levels in tomatoes over the course of 18 hours.

A new platform for farming

The researchers believe the microneedles could be used to complement existing agricultural practices like spraying. The researchers also note the technology has applications beyond agriculture, such as in biomedical engineering.

“This new polymeric microneedle fabrication technique may also benefit research in microneedle-mediated transdermal and intradermal drug delivery and health monitoring,” Cao says.

For now, though, Marelli believes the microneedles offer a path to more precise, sustainable agriculture practices.

“We want to maximize the growth of plants without negatively affecting the health of the farm or the biodiversity of surrounding ecosystems,” Marelli says. “There shouldn’t be a trade-off between the agriculture industry and the environment. They should work together.”

This work was supported, in part, by the U.S. Office of Naval Research, the U.S. National Science Foundation, SMART, the National Research Foundation of Singapore, and the Singapore Prime Minister’s Office.

This is the second part of a three-part series about age verification in the European Union. In this blog post, we take a deep dive into the age verification app solicited by the European Commission, based on digital identities. Part one gives an overview of the political debate around age verification in the EU and part three explores measures to keep all users safe that do not require age checks. 

In  part one of this series on age verification in the European Union, we gave an overview of the state of the debate in the EU and introduced an age verification app, or mini-wallet, that the European Commission has commissioned. In this post, we will take a more detailed look at the app, how it will work and what some of its shortcomings are.

According to the original tender and the app’s recently published specifications, the Commission is soliciting the creation of a mobile application that will act as a digital wallet by storing a proof of age to enable users to verify their ages and access age-restricted content.

After downloading the app, a user would request proof of their age. For this crucial step, the Commission foresees users relying on a variety of age verification methods, including national eID schemes, physical ID cards (acknowledging that biometric analysis would be necessary for identifying a user corresponding to an ID), linking the app to another app that contains information about a user’s age, like a banking app, or age assessment through third parties like post offices. 

In the next step, the age verification app would generate a proof of age. Once the user would access a website restricting content for certain age cohorts, the platform would request proof of the user’s age through the app. The app would then present proof of the user’s age via the app, allowing online services to verify the age attestation and the user would then access age-restricted websites or content in question. The goal is to build an app that will be aligned and allows for integration with the architecture of the upcoming EU Digital Identity Wallet. 

The user journey of the European Commission's age verification app

Review of the Commission’s Specifications for an Age Verification Mini-ID Wallet 

According to the specifications for the app, interoperability, privacy and security are key concerns for the Commission in designing the main requirements of the app. It acknowledges that the development of the app is far from finished, but an interactive process, and that key areas require feedback from stakeholders across industry and civil society. 

The specifications consider important principles to ensure the security and privacy of users verifying their age through the app, including data minimization, unlinkability (to ensure that only the identifiers required for specific linkable transactions are disclosed), storage limitations, transparency and measures to secure user data and prevent the unauthorized interception of personal data. 

However, taking a closer look at the specifications, many of the mechanisms envisioned to protect users’ privacy are not necessary requirements, but optional. For example, the app  shall implement salted hashes and Zero Knowledge Proofs (ZKPs), but is not required to do so. Indeed, the app’s specifications seem to heavily rely on ZKPs, while simultaneously acknowledging that no compatible ZKP solution is currently available. This warrants a closer inspection of what ZKPs are and why they may not be the final answer to protecting users’ privacy in the context of age verification. 

A Closer Look at Zero Knowledge Proofs

Zero Knowledge Proofs provide a cryptographic way to not give something away, like your exact date of birth and age, while proving something about it. They can offer a “yes-or-no” claim (like above or below 18) to a verifier requiring a legal age threshold. Two properties of ZKPs are “soundness” and “zero knowledge.” Soundness is appealing to verifiers and to governments to make it hard for a prover to present forged information. Zero-Knowledge can be beneficial to the holder, because they don’t have to share explicit information, just the proof that said information exists. This is objectively more secure than uploading a picture of your ID  to multiple sites or applications, but it still requires an initial ID upload process as mentioned above for activation.

This scheme makes several questionable assumptions. First, that frequently used ZKPs will avoid privacy concerns, and second, that verifiers won’t combine this data with existing information, such as account data, profiles, or interests, for other purposes, such as advertising. The European Commission plans to test this assumption with extremely sensitive data: government-issued IDs. Though ZKPs are a better approach, this is a brand new system affecting millions of people, who will be asked to provide an age proof with potentially higher frequency than ever before. This rolls the dice with the resiliency of these privacy measures over time. Furthermore, not all ZKP systems are the same, and while there is  research about its use on mobile devices, this rush to implementation before the research matures puts all of the users at risk.

Who Can Ask for Proof of Your Age?

Regulation on verifiers (the service providers asking for age attestations) and what they can ask for is also just as important to limit a potential flood of verifiers that didn’t previously need age verification. This is especially true for non Know-Your-Customer (KYC) cases, in which service providers are not required to perform due diligence on their users. Equally important are rules that determine the consequences for when verifiers violate those regulations. Up until recently, the eIDAS framework, of which the technical implementation is still being negotiated, required registration certificates across all EU member states for verifiers. By forcing verifiers to register the data categories they intend to ask for, issues like illegal data requests were supposed to be mitigated. But now, this requirement has been rolled back again and the Commission’s planned mini-AV wallet will not require it in the beginning. Users will be asked to prove how old they are without the restraint on verifiers that protects from request abuse. Without verifier accountability, or at least industry-level data categories being given a determined scope, users are being asked to enter into an imbalanced relationship. An earlier mock-up gave some hope for  empowered selective disclosure, where a user could toggle giving discrete information on and off during the time of the verifier request. It would be more proactive to provide that setting to the holder in their wallet settings, before a request is made from a relying party.

Privacy tech is offered in this system as a concession to users forced to share information even more frequently, rather than as an additional way to bring equity in existing interactions with those who hold power, through mediating access to information, loans, jobs, and public benefits. Words mean things, and ZKPs are not the solution, but a part of one. Most ZKP systems are more focused on making proof and verification time more efficient than they are concerned with privacy itself. The result of the latest research with digital credentials are more privacy oriented ways to share information. But at this scale, we will need regulation and added measures on aggressive verification to complete the promise of better privacy for eID use.

Who Will Have Access to the Mini-ID Wallet, and Who Will Be Left Out?

Beyond its technical specifications, the proposed app raises a number of accessibility and participation issues. At its heart, the mini-ID wallet will rely on the verification of a user’s age through a proof of age. According to the tender, the wallet should support four methods for the issuance and proving of age of a user.

Different age verification methods foreseen by the app

The first options are national eID schemes, which is an obvious choice: Many Member States are currently working on (or have already notified) national eID schemes in the context of the eIDAS, Europe’s eID framework. The goal is to allow the mini-ID wallet to integrate with the eIDAS node operated by the European Commission to verify a user’s age. Although many Member States are working on national eID schemes, previous uptake of eIDs has been reluctant, and it's questionable whether an EU-wide rollout of eIDs will be successful. 

But even if an EU-wide roll out was achievable, many will not be able to participate. Those who are not in possession of ID cards, passports, residence permits, or documents like birth certificates will not be able to attain an eID and will be at risk of losing access to knowledge, information, and services. This is especially relevant for already marginalized groups like refugees or unhoused people who may lose access to critical resources. But also many children and teenagers will not be able to participate in eID schemes. There are no EU-wide rules on when children need to have government-issued IDs, and while some countries, like Germany, mandate that every citizen above the age of 16 possess an ID, others, like Sweden, don’t require their citizens to have an ID or passport. In most EU Member States, the minimum age at which children can apply for an ID without parental consent is 18. So even in cases where children and teenagers may have a legal option to get an ID, their parents might withhold consent, thereby making it impossible for a child to verify their age in order to access information or services online.

The second option are so-called smartcards, or physical eID cards, such as national ID cards, e-passports or other trustworthy physical eID cards. The same limitations as for eIDs apply. Additionally, the Commission’s tender suggests the mini-ID wallet will rely on biometric recognition software to compare a user to the physical ID card they are using to verify their age. This leads to a host of questions regarding the processing and storing of sensitive biometric data. A recent study by the National Institute of Standards and Technology compared different age estimation algorithms based on biometric data and found that certain ethnicities are still underrepresented in training data sets, thus exacerbating the risk age estimation systems of discriminating against people of color. The study also reports higher error rates for female faces compared to male faces and that overall accuracy is strongly influenced by factors people have no control over, including “sex, image quality, region-of-birth, age itself, and interactions between those factors.” Other studies on the accuracy of biometric recognition software have reported higher error rates for people with disabilities as well as trans and non-binary people. 

The third option foresees a procedure to allow for the verification of a user’s identity through institutions like a bank, a notary, or a citizen service center. It is encouraging that the Commission’s tender foresees an option for different, non-state institutions to verify a user’s age. But neither banks nor notary offices are especially accessible for people who are undocumented, unhoused, don’t speak a Member State’s official language, or are otherwise marginalized or discriminated against. Banks and notaries also often require a physical ID in order to verify a client’s identity, so the fundamental access issues outlined above persist.

Finally, the specification suggests that third party apps that already have verified a user's identity, like banking apps or mobile network operators, could provide age verification signals. In many European countries, however, showing an ID is a necessary prerequisite for opening a bank account, setting up a phone contract, or even buying a SIM card. 

In summary, none of the options the Commission considers to allow for proving someone’s age accounts for the obstacles faced by different marginalized groups, leaving potentially millions of people across the EU unable to access crucial services and information, thereby undermining their fundamental rights. 

The question of which institutions will be able to verify ages is only one dimension when considering the ramification of approaches like the mini-ID wallet for accessibility and participation. Although often forgotten in policy discussions, not everyone has access to a personal device. Age verification methods like the mini-ID wallet, which are device dependent, can be a real obstacle to people who share devices, or users who access the internet through libraries, schools, or internet cafés, which do not accommodate the use of personal age verification apps. The average number of devices per household has been  found to correlate strongly with income and education levels, further underscoring the point that it is often those who are already on the margins of society who are at risk of being left behind by age verification mandates based on digital identities. 

This is why we need to push back against age verification mandates. Not because child safety is not a concern – it is. But because age verification mandates risk undermining crucial access to digital services, eroding privacy and data protection, and limiting the freedom of expression. Instead, we must ensure that the internet remains a space where all voices can be heard, free from discrimination, and where we do not have to share sensitive personal data to access information and connect with each other.

This seems like an important advance in LLM security against prompt injection:

Google DeepMind has unveiled CaMeL (CApabilities for MachinE Learning), a new approach to stopping prompt-injection attacks that abandons the failed strategy of having AI models police themselves. Instead, CaMeL treats language models as fundamentally untrusted components within a secure software framework, creating clear boundaries between user commands and potentially malicious content.

[…]

To understand CaMeL, you need to understand that prompt injections happen when AI systems can’t distinguish between legitimate user commands and malicious instructions hidden in content they’re processing...

The move plunges the congressionally mandated report into disarray and raises questions about whether the president will fill the void with pseudoscience.

Climate nonprofits that say they are about to go out of business will have to wait as the court considers whether EPA can terminate $20 billion in awards.

A recent investigation found "no evidence" that FEMA under Joe Biden deliberately avoided helping homes in Florida that backed Donald Trump.