Let’s be frank: When you go to see your personal doctor or stop at an urgent care for a quick visit, you probably don’t always tell the physician the whole truth and nothing but the truth. According to one major study, between 60 and 80 percent of US adults have reported they withheld at least one pertinent piece of medical information from a healthcare provider during a visit.

And we do it for the most natural and human reasons. The most common explanations that patients gave for declining to disclose relevant information were not wanting to be lectured, not wanting to be judged for their behavior, and being embarrassed. Women, young people, and the people with the worst health (according to their own self-report) were most likely to hold something back. 

If you have faced discrimination from society because of who you are, you may feel more distrustful of the medical system and its practitioners. Research has found that people who have faced prior discrimination are more likely to withhold information from a healthcare provider. And doctors sometimes earn that distrust, especially when it comes to women and people of color. Our era of do-it-yourself medicine only adds more barriers between patients and their doctors. One survey found that more than 40 percent of US adults who had used alternative or complementary medicine did not disclose that fact to their primary care doctor.

Doctors bear some of the blame for this disconnect. But, at a time when people say they want more agency over their healthcare, the rest of us have to hold up our side of the conversation, too. Here’s how to think about how honest you should be with your doctor.

The key question you should ask yourself about being honest with your doctor

There is no shame in feeling uncomfortable sharing information with your doctor. I spoke with Dr. Ronald Epstein, a professor of family medicine at the University of Rochester Medical Center who has written extensively on patient-doctor communication and authored the book Attending: Medicine, Mindfulness, and Humanity as a guide for doctors who want to practice a more humane kind of medicine. Epstein said he’s been tempted to not tell his doctor when he misses doses of his prescribed medication because he felt embarrassed, and even withheld information once when he didn’t want to spark a lot of burdensome follow-up work; he had fallen down while skiing after bumping into a rock. Because Epstein is 71, his provider routinely screens for any recent falls, and he knew that answering “yes” would lead to a range of tests and consultations with specialists. So he answered “no,” because he felt certain the fall had been a freak accident, not a symptom of an underlying health problem.

I appreciated Epstein making that confession, because it was a reminder that any of us, even somebody who has committed much of their career to open doctor-patient communication, can be tempted to hold something back from a doctor because it seems like it’ll be a hassle.

But even as he shared it, he offered a warning to other people not to follow his example.

“The danger in doing that, of course, is if someone actually fell while turning around on a landing while going downstairs, that could signal some kind of balance or neurologic disorder,” he told me.

To state the obvious, your clinician can’t provide you with the best care if they don’t know everything about your health. That’s especially important in high-stakes situations. One study found that one in four patients who were experiencing an imminent threat to their well-being — depression, suicidal ideation, abuse, or sexual assault — withheld that information from their doctor. Once again, they feared being embarrassed or being judged.

So, to start with, there are certain things your doctor really needs to know if they are going to properly evaluate your health and that you should never lie about or withhold:

• Which prescription drugs you are currently on
• Whether you’re actually taking them as prescribed
• Any other substances you are consuming, including those that are legal but potentially harmful (like tobacco or alcohol) and ones that are illegal, and how much/how often you are using them

Even if it is awkward or you fear judgment, the risks in withholding that kind of information are high. Beyond that, Epstein suggested one question that every patient should ask themselves when interacting with a doctor.

“I think the key question for patients to ask themselves and for families to ask themselves is: Would withholding this information threaten or enhance the health of the person involved?” he told me.

Doctors have a responsibility to make their patients feel comfortable enough to be honest. But from the patient’s perspective, this question cuts to the heart of the matter. If you withhold a certain piece of information from your provider, does it put your and a loved one’s health at risk? You should also ask the inverse of this question: Could sharing the information lead to a better outcome for you or your loved one? 

At the end of the day, these are judgment calls for each individual to make. But think the situation through and be honest with yourself. Are you really equipped to be the judge of whether a certain piece of information is relevant to your or your loved one’s health — or should you let the doctor decide? If you withhold some information, Epstein said, “you’re taking the risk that you know better versus being transparent and letting your doctor make that judgment.”

Communication between doctor and patient is a two-way street

Building a good relationship between doctors and patients is, as Epstein put it to me, “a shared responsibility.” 

“There are two aspects of any healing tradition. One is the technical and instrumental pieces. That is the drugs you prescribe, the surgeries, the manipulations you do,” he said. “The other is relational. That is the achievement of shared understanding, trust, confidence, and sometimes optimism so that people can really make decisions on their own behalf and feel empowered.”

Epstein trained at Harvard Medical School in the 1980s, as the AIDS crisis was beginning, and he still recalls the sometimes awkward conversations doctors and the people they were treating had to have about sex. Even though the stakes were clearly life and death, many doctors were too embarrassed to ask detailed questions about people’s safer sex practices. Sex, drugs, and money remain sensitive and uncomfortable subjects, but they can have a huge bearing on a person’s health. 

Doctors can easily build walls between themselves and their patients without even realizing it — and, sometimes, even a subtle change can make patients feel more welcome. Asking “are you taking your medications daily?” in an abrupt, scolding fashion could lead a person to lie and say yes to avoid being judged. A softer approach — “it’s hard to remember to take pills every day, how are you doing with that?” — could yield more honest answers for a question that has serious clinical implications.

“If the pills are something that if you miss a dose, you might suffer severe consequences, then it can be life and death,” Epstein said.

People often feel nervous during medical appointments — I know that’s true for myself — and Epstein said doctors should take care to make sure their patients are absorbing the information. Asking them to repeat back instructions, for example, can act as a backstop to avoid any miscommunication. 

Sometimes, the solution can be as simple as phrasing questions in an open-ended way. “Do you have any more questions?” could lead to a quick “no” from an anxious patient, while posing “what questions do you have?” might invite a more open dialogue.

Epstein has a favorite Frank Kafka quote that illustrates the challenge: “To write prescriptions is easy. To come to human understanding is difficult.” Good medicine requires both.

Ismail Harerimana grew up in Uganda not knowing why he was always sick. 

His childhood in the 1990s was a string of recurrent infections: malaria, diarrhea, headaches, and skin rashes. By 14, he was scarily thin, at which point doctors put him on a new medication that seemed to help. It was for kidney disease, his father falsely told him. But a classmate with the same prescription knew better. “Are you also suffering from kidney disease?” Harerimana remembers asking him. “And the boy said, ‘No — I’m suffering from AIDS.’”

In the 1990s, at the height of the AIDS crisis in Uganda, hundreds of thousands of babies like Harerimana were born with HIV each year, contracting the virus from their HIV-positive parents in utero, during childbirth, or while breastfeeding. About half did not live to see their second birthday.  

But those outcomes have changed in radical, often remarkable ways over the past three decades. In some parts of Uganda, as many as one in four infants were once infected with HIV at birth, leading to 32,000 new childhood HIV infections annually in the mid-1990s. Today, that infection rate has plummeted to fewer than 5,000. 

This changed because Uganda — along with much of the world — has diligently perfected the simple interventions needed to keep babies safe from the virus: repeated HIV testing for all expectant parents, and widely available anti-retroviral therapies for those who test positive, which makes the virus virtually untransmittable. In some countries, Botswana among them, new childhood infections are now so exceedingly rare that every new baby born with HIV prompts a comprehensive federal audit.

“I’m filled with hope because now, as Africans, we’re not asking whether elimination is possible,” said Doris Macharia, president of the Elizabeth Glaser Pediatric AIDS Foundation. “We are actually confronting what it will take to finish this job. That is profound. That is progress. And that’s where we should be.”

But finishing the job would mean building a world where no babies are born with HIV at all, and many African countries with the highest HIV burdens remain far from that goal. About 120,000 children are still newly infected with HIV each year, most of them before or shortly after birth, accounting for nearly 10 percent of all new infections. That’s one child every four and a half minutes. 

Thanks to advancements in treatments, even babies born with HIV today can go on to live long, healthy, happy lives. But it is more difficult, because the same barriers that prevent their parents from getting on treatment while pregnant mean that many of their children struggle to access care. As a result, roughly 75,000 kids die from AIDS-related causes each year, typically before their fourth birthday. That is almost definitely an undercount, as it likely excludes many of the roughly 34 percent of children living with HIV who are never accurately diagnosed. 

Reaching these kids is what Macharia calls the last mile in preventing childhood HIV. It is also the hardest to cross — and particularly so now. Cuts to foreign assistance from the US and other countries have hampered progress, and in some harrowing cases, even reversed it. A projection by UNAIDS found that sustained aid cuts could lead to 1.1 million additional HIV infections in children between 2024 and 2040, and 820,000 more deaths.

Harerimana, who has found his calling as a community health worker, is already seeing some of those dire scenarios play out. For the first time in years, he’s seen an uptick in babies being born with HIV in his town.

“It takes me back to those days,” he said, “when there was no access to medication, where there was no access to research,” there was only “a disease everyone fears, a disease that has no concrete cure.”

Regression is not inevitable. Even the Trump administration — which deeply destabilized global HIV services last year — has supported the rollout of Lenacapavir, a potentially game-changing HIV prevention drug, for expectant parents at risk of HIV. Stopping babies from being born with HIV is, after all, about as sympathetic a case as you can get with foreign aid. But the very aid systems that have helped us reach the cusp of an HIV-free generation are now confronting a massive transition, one that makes all elements of care far more difficult. 

The secret to making sure kids don’t get HIV

After Harerimana learned he had HIV, he began zoning out in class. He couldn’t understand how a kid like him could get a virus he thought spread only through unprotected sex. 

“I would just sit and get lost. My mind would only think about how I’m going to lose my friends, how I’m going to die very soon,” he said. “And I started to ask God, like, ‘God, where did I get this disease?’”

Even many adults at the time didn’t realize there were other ways to contract HIV. Pervasive stigmas around HIV have made correcting such misconceptions an uphill battle around the world. As recently as 2016, only 56 percent of young women in Uganda knew much about vertical transmission, which is how the vast majority of children acquire HIV. Nearly half of babies born to an HIV-positive parent who is not on treatment will contract the virus. In comparison, there is at most a 1 in 72 chance of contracting the virus if you have unprotected sex with an untreated HIV-positive partner, and a 1 in 158 chance if you share needles with them.

But as awful as it sounds, at the height of the HIV epidemic, there “was not a market” for investing in pediatric treatment and prevention, said Florence Riako Anam, co-executive director of the Global Network of People Living with HIV. That was because “most of the children who acquired HIV did not live long. Many of them did not go beyond months, frankly.”

But some, like Harerimana, did live long enough to see a renaissance of new treatments and discoveries. The medication he began as a teen was an anti-retroviral therapy, or ARV, that these days is so effective, it can virtually eliminate HIV from your bloodstream. 

In 1994, a group of American researchers found that people who are pregnant and on treatment have a minuscule chance of passing the virus on to their baby, results so impressive that they halted their medical trial so they could offer treatment to the placebo group. Nearly 80 percent of HIV-positive pregnant people in the US were on ARVs by 1999. By 2003, just 1.2 percent of those parents passed the virus to their children.

But it would take many years for these miracle drugs to reach most African countries. Philippa Musoke, a pediatric infectious disease specialist in Uganda, led a landmark study in 1999 that found just two doses of the HIV drug Nevirapine — which cost $2 at the time per dose — slashed the chance a newborn would contract the virus by 50 percent. Other treatments relied on a “cocktail” of drugs that were much more effective, but often prohibitively expensive, costing $815 for a month-long course in the US.  

“It opened people’s eyes that a simple regimen could actually prevent mother-to-child transmission globally,” Musoke told me. Within a few years, many countries began rolling out free Nevirapine programs  — and later, more effective combined drug treatments — for pregnant people living with HIV. 

Most of the world saw its childhood infection rate collapse, but the undisputed breakout star was Botswana, which, in 1999, became the first African country to offer free HIV drugs to all pregnant women. At the time, a woman in the country had a one in four chance of having HIV, among the highest rates in the world. If she had three children in the years that followed, at least one would likely become infected before or during childbirth or breastfeeding. 

But thanks to the free treatment program, and a robust maternal health system that integrates universal HIV testing, a young Botswanan woman living with HIV today has an under 1.2 percent chance of passing the virus to her kids. Last year, the World Health Organization certified Botswana as the first country in the world with a high HIV rate to eliminate mother-to-child transmissions as a public health threat.

Other countries have also managed to pull off remarkable, albeit more modest, progress. In Kenya, where Anam lives, more than three-quarters of pregnant people with HIV received treatment in 2008, up from virtually none in 2003. In those five years, the number of children newly infected with HIV fell by 75 percent. 

After contracting HIV, “I don’t think many of us thought we could have kids,” not safely at least, said Anam, who tested positive for the virus shortly after giving birth to her first child 26 years ago. “And then over time, with advancement in treatment, it became an option for women.” 

Many of her friends who thought they could never have more children, some of whom lost their first babies to HIV in the 1990s, suddenly found they could have kids safely. Their second children, she says, are now in their tweens. 

Botswana cracked the code. Why can’t everyone else?

Even with all that progress, hundreds of babies are still being born with HIV each day. Other than Botswana, no country with a high HIV rate has managed to all but eliminate childhood HIV. Despite decades of progress and far better treatments, the rest of the world is still stubbornly far from that goal. 

“We’ve really made significant progress, but we’re not there yet,” Musoke said. “That is really unacceptable because we have all the knowledge, we have all the resources” to ensure no child is born with HIV in theory.

Yet about one in six pregnant people living with HIV is still not on treatment. And about half of those who are on treatment don’t take it as consistently as they should. Together, their children account for the vast majority of the 328 infected with HIV every single day.

Reaching these parents is critical. The problem is that many of them do not know they have the virus and live in rural areas where there are few providers who can test them for it. 

“Eliminating pediatric HIV and mother-to-child transmission is no longer a scientific question,” Macharia said. “It’s really a delivery and a systems question,” which will require more outreach workers, especially peer mentors, people living with HIV who’ve been trained to help others like themselves navigate their treatment and prevention options.

Liako Serobanyane tested positive for HIV in 2007, when she was pregnant with her second child. She trained as a mentor mother through the group Mothers2Mothers in Lesotho because she wanted to help “other women going through what I went through, even though I didn’t get the support I needed at the time,” she said. “There is no other model better than this, because we have been there. We know how it feels to be HIV-positive. We know how it feels to be rejected.”

The progress that’s been made so far against mother-to-child transmission has largely stemmed from parents who were easier to reach. They were already receiving prenatal care or giving birth at a clinic or hospital, as 99.8 percent of expectant parents in Botswana do. But there are still many parents with limited access to care. In Nigeria, which accounts for one in seven of the world’s babies born with HIV, about half of parents give birth at home with no skilled health worker present. The country has offered free HIV treatment to its citizens for nearly two decades now. But not enough pregnant people are taking them up on it. It is mentors like Serobanyane who have the best shot at making sure they do.

“We can’t just wait for people to come to the clinic” anymore, said Macharia of the Elizabeth Glaser Pediatric AIDS Foundation. “We have to go to them.” 

The US built the system to keep babies HIV-free. It’s now dismantling it.

But bringing together all of those factors – strengthening delivery systems, hiring more peer mentors, normalizing HIV testing, and convincing more parents to give birth at the hospital – is neither easy nor cheap.

Maybe the biggest difference between Botswana and other countries with high HIV rates is that Botswana has diamonds. Lots of diamonds. Enough diamonds to turn Botswana into one of Africa’s richest countries per capita. 

That’s allowed Botswana to largely bankroll its own HIV response. As Alankar Malviya, Botswana country director for UNAIDS, told me, the country pays for about 70 percent of all testing, treatment, and outreach costs. Other less well-off countries like Nigeria have built about 90 percent of their HIV response primarily with the help of PEPFAR, the US-funded HIV program that began in 2003. It’s no coincidence that much of the world’s success in fighting off childhood HIV infections so far began that year. PEPFAR has helped make sure that at least 7.8 million babies were not born with HIV over the past 26 years. 

PEPFAR continues to fund lifesaving HIV treatment around the world, according to newly released data, but the Trump administration has severely disrupted its support for prevention and outreach work. That includes cuts to many outreach programs aimed at preventing mother-to-child HIV transmission, though the administration has maintained funding for some services, such as prenatal testing. 

With less funding for HIV screenings and prevention, fewer pregnant people will know they need antiretrovirals in the first place. They won’t have the condoms they need to prevent the spread. And if their babies contract the virus in utero or while breastfeeding, their parents might not know why they are so sick until it is too late.

“We are in a period of transition,” a senior official from the US State Department, which now oversees PEPFAR, told me under the condition of anonymity. “And during that transition, yes, there may be a few people who used to go to a particular community site that isn’t there anymore, and are having to figure out where to get those services from.”

The official insisted that the US still cares about preventing mother-to-child transmission. The Trump administration has shifted the way aid works by channeling it through bilateral agreements that require countries to partially pay their own way. It throws the old, and in many ways, highly successful system of HIV aid — which relied on international organizations as partners — out the window.

“Yes, it saved lives. Yes, it made progress,” the official said of the old aid order. “But it isn’t a model we can keep going with.”

Josephine Nabukenya, a pediatric HIV advocate who, like Harerimana, was born with the virus in the 1990s, agrees that having countries take more ownership of their health care system is a good thing in the long run. “But you do it in a phased approach,” she said, to avoid letting parents and children fall through the cracks. 

So far, that’s not how it’s played out. Mothers2Mothers, an organization that, since 2001, has trained HIV-positive moms like Serobanyane to be peer health mentors — a uniquely effective intervention — lost most of its funding last year. They closed offices in four countries and laid off hundreds of workers and peer mothers, shutting off outreach services for 450,000 people.

Serobanyane is based in Lesotho, one of the few countries where the group still operates. Because of funding cuts, she is one of just two mentor mothers in her district, down from six. “We love our job. We are doing it passionately,” she said, “but not knowing if the funding is going to be there or is going to be cut off is depressing and tiring.” 

She also worries for the mothers whose treatment or testing she can no longer follow as closely. Reminding them to attend their prenatal screenings or refill their treatment prescriptions requires resources and support that are no longer as available to her. 

Lesotho is one of the over 30 countries that have signed bilateral health aid deals with the State Department so far. The country is set to receive $232 million over 5 years from the US, which its government could theoretically use to hire its own mentor mothers and otherwise make up for lapses in HIV care and outreach. “It’s our dream that the mentor mother model be absorbed by the government one day,” Serobanyane said.

But the reality is, said Mpolokeng Mohloai, director of Mothers2Mothers in Lesotho, “the government is not yet ready to absorb it all.” 

“Every child that is infected with HIV is unacceptable.”

In an absolute worst-case scenario, if US-funded HIV programs aren’t adequately replaced, then a total of up to 1.7 million more children could die of AIDS-related causes by 2040, according to UNAIDS, a devastating leap in the wrong direction on an issue where the world had been making so much progress.

Even if governments do manage to plug some gaps, a large number of parents and children will lose access to support in the short term as a result of funding cuts. This means more mothers who don’t know they’re HIV-positive until it’s too late, more parents who fall behind on their medications, and more children who grow up to be very sick.

“Every child that is infected with HIV is unacceptable. Any mom who acquires HIV during pregnancy, breastfeeding, or even before then — that is also unacceptable,” said Macharia of the Elizabeth Glaser Pediatric AIDS Foundation. “Those have to be unacceptable facts for us.”

Harerimana lost his job as a community health worker last year when the Trump administration put a pause on all foreign assistance funding. He has continued to work without pay, supporting children and their parents, some of whom he says have already missed out on critical treatment.

“I can now comfortably say that over the past year, when the aid cuts and confusion started, we are now seeing children getting infected by HIV through mother-to-child transmission again,” he said. “By the time the system stabilizes, the world will know how much the aid cuts have caused.”

In a darkened convention hall in Chicago on May 31, a Harvard oncologist named Brian Wolpin stood at a podium and in a voice that sounded as if he was reading from the phone book, recited a set of numbers that brought a roomful of cancer doctors to their feet for 42 seconds. Adam Feuerstein, a biotech correspondent for the health news site Stat who has covered cancer conferences like this for two decades, said he had never witnessed anything like it. The applause lasted so long that Wolpin, caught off-guard, ad-libbed: “That time was not built into my talk.” 

What Wolpin had just shown attendees at the American Society of Clinical Oncology’s (ASCO) annual meeting was a simple line graph: a drug called daraxonrasib had nearly doubled median overall survival in a 500-patient trial of a form of previously treated advanced pancreatic cancer. ASCO’s chief medical officer Julie Gralow termed the result not a home run but a “grand slam.” Toronto oncologist Jennifer Knox called it a “game changer.”

Wolpin received such a rapturous response at ASCO because pancreatic cancer is among the most pernicious and treatment-resistant cancers in existence, killing more than 50,000 Americans a year, among them Supreme Court Ruth Bader Ginsburg. The cancer has a five-year survival rate in the low teens. 

Wolpin, who began his career in the mid-2000s at the world-class Dana-Farber Cancer Institute, told The Bulwark: “I think I saw several patients that first year of fellowship who had pancreatic cancer, and they all died in like three months. It’s not supposed to happen here, right? You’re supposed to have figured this out.” For decades after President Richard Nixon declared a “war on cancer,” deaths continued to mount and medical progress on many cancers remained all too limited. 

But a change is well underway. The US death rate from cancer has fallen 34 percent from its 1991 peak through 2023, and the five-year relative survival for all cancers combined reached 70 percent for people diagnosed between 2015 tto 2021, up from 50 percent in the 1970s. And while daraxonrasib got the standing ovation, it was only the loudest moment in a week — and a decade — of steady, compounding victories over cancer.

The immune system, turned up

One major driver of the shift is immunotherapy. Rather than attacking a tumor directly as conventional chemotherapy does, these treatments use a patient’s own immune system to hunt and kill cancer cells. You can see immunotherapy’s powerful effects through the story of former President Jimmy Carter, who was diagnosed in 2015 at age 90 with metastatic melanoma that had spread to his liver and brain. That should have been a sign for newspaper editors to update their planned obituaries immediately; yet after being treated with the immunotherapy drug pembrolizumab, as well as surgery and radiation, Carter watched his tumors vanish and managed to live another decade. 

And scientists keep pushing the frontier further. Moderna and Merck reported that the combination of a personalized mRNA vaccine — the technology behind the Covid shots, retrained on each patient’s own tumor — and an immuontherapy drug (pembrolizumab) reduced the risk of recurrence or death for high-risk melanoma by 49 percent after five years. In a small, early Memorial Sloan Kettering trial of a similar vaccine appeared to help some pancreatic cancer patients stay cancer-free longer after surgery. Seven of the eight patients who responded to the vaccine were still alive four to six years later, with a larger trial now underway.

A Memorial Sloan Kettering trial of a similar vaccine in 2024 kept pancreatic cancer at bay in patients whose immune systems responded to it. And for blood cancers, a single infusion of reengineered immune cells — called CAR T-cell therapy — has begun producing something that looks close to a cure: Emily Whitehead, the first child with cancer ever treated with CAR T, is now more than a decade cancer-free and attending college. (I wrote in more detail about immunotherapy and CAR T last year.) 

From treatment to prevention

And scientists’ ambitions are growing, from treating cancer to stopping it before it starts. Last week, a team led by the Francis Crick Institute’s Charles Swanton reported that a blood test measuring 14 proteins, combined with basic risk factors like age, smoking, and lung disease, could help identify people likely to develop lung cancer years before diagnosis. They also found an intriguing clue from an older drug trial: An anti-inflammatory drug seemed to cut lung cancer risk nearly in half among people with the highest inflammation levels. 

This is still early evidence — not yet a blood test and prevention treatment doctors can offer patients — but Swanton compared it to how statins work for heart disease. Just as cholesterol tests can predict a person’s risk of heart disease, and then statins can be given to lower cholesterol, the protein test identifies lung cancer risk and the anti-inflammatory drug reduces it. 

And no story on modern medical miracles would be complete without an appearance from GLP-1 drugs, which truly do seem to do everything. A University of Pennsylvania study of more than 110,000 women, also reported at the ASCO meeting this week, found that taking GLP-1 drugs like Ozempic was associated with about 30 percent lower breast cancer incidence.

Both findings are early, so we shouldn’t expect major changes overnight. It took decades between the development of a test for LDL cholesterol levels, the introduction of statins, and the undeniable proof of heart disease prevention. But oncology is clearly moving toward catching cancer before it takes hold, just as we have with heart attacks. 

Beyond the numbers

Medical advances come with a literal cost. The new medicines are brutally expensive, with the average monthly price of a new cancer drug more than doubling between 2009 and 2019, while about half of surveyed American cancer patients and survivors have to take on debt to pay for treatment. 

Many of those high prices will eventually fall, once patents run out and generic versions emerge. But a greater worry is that the scientific engine driving these advances is being throttled. Almost every advance I’ve mentioned can be traced back to federally funded basic research, which the Trump administration has been attacking relentlessly.

In 2025, the administration froze or canceled thousands of National Institutes of Health (NIH) and National Science Foundation (NSF) grants, while new NIH awards fell by billions of dollars. Congress later rejected the deepest proposed NIH cuts, but the damage was already real: Hundreds of NIH-funded clinical trials were disrupted, and early-career scientists became much less likely to win major grants. In saving dollars with those cuts, we risk losing discoveries that would save lives, at the very moment when cancer research is paying off.

The cost of those lives was made visceral at the ASCO meeting. In the opening address, ASCO’s outgoing president Eric Small spoke about his partner, Amy Lin, a University of San Francisco San Francisco oncologist. Lin had died in December of metastatic clear cell ovarian cancer, a deadly disease that still has few treatment options. He brought on the grief expert and author David Kessler to give a talk on compassionate end-of-life care.   

Perhaps more than any other medical specialty, grief and loss have always been an inseparable, if rarely discussed, part of oncology. Brian Wolpin started his career watching pancreatic patients die within months and feeling certain it wasn’t supposed to happen at a place like Dana-Farber. The ovation he got was the sound of a room realizing he might be right — that the disease that once seemed untreatable is starting to lose its terrible power.

A version of this story originally appeared in the Good News newsletter. Sign up here!

AI company CEOs Sam Altman (OpenAI), Demis Hassabis (Google DeepMind), and Dario Amodei (Anthropic) disagree on a lot, like how fast the technology should develop, the best way to regulate it, and how to prepare society for smarter-than-human AI, among other things. 

That makes it all the more remarkable that they — along with 85 other experts in tech, biology, and national security policy — recently signed on to an open letter calling for more robust regulations around gene synthesis. They’re all concerned that AI systems might be used to help develop and even deploy dangerous biological weapons designed through gene synthesis, which is used to chemically build custom DNA sequences in a lab, rather than relying solely on existing natural DNA templates.

The simple fact of multiple CEOs of fiercely competitive AI companies aligning on anything is remarkable. But to understand how they came to this agreement, we have to take a step back to understand what gene synthesis is, how it works, and why the possibility of AI-assisted misuse of the technology generates so much fear.  

Modern microbiology owes a lot to gene synthesis. Researchers can order synthetic genes from commercial DNA providers to develop new vaccines, drugs, and gene therapies for inherited diseases like hemophilia; produce human insulin, boost agricultural output, and more. Gene synthesis is a foundational technology for successful CAR-T cell therapies for cancer and many diagnostic tools. The demand for synthetic DNA is growing globally, and it’s never been cheaper or simpler to write genetic code.  

But for all its power, gene synthesis also carries substantial risk. The same technology that can enable life-saving new gene therapies can also assist in the creation of deadly pathogens by assembling some of the same nucleotides — the genetic building blocks that create the code for all of life — in a different order. 

Most US companies that provide gene synthesis services screen orders for genetic sequences of concern, such as those that can make a pathogen more dangerous or transmissible, and to verify that customers are legitimate. They do so voluntarily, well aware of the potential dangers. 

But not every provider does so. “As long as screening remains voluntary, some companies will not do it,” Becky Mackelprang, the director for security programs at the Engineering Biology Research Consortium, told me over email. There’s a real risk that bad actors could find a gene synthesis company with more lax standards, and that might mean disaster.

We’ve been fortunate so far. “This technology has been commercially deployed for more than 20 years and has never been misused to cause harm,” James Diggans, the vice president of policy and biosecurity at gene synthesis company Twist Bioscience, told me over email.

But AI threatens to complicate matters, opening up new frontiers of risk.

For good or for ill

Both large language models (LLMs) and AI biodesign tools enable scientists to design entirely novel genetic sequences. This is a boon for industrial and medical applications — and a challenge for current screening systems, which use similarity to known pathogenic or toxic sequences in order to detect risk. A screening system should catch someone trying to order sequences of a known dangerous virus like Ebola, for example, but it might miss a new sequence that could still be risky. Last year, a study published in Science demonstrated that our screening systems have kept pace with AI capabilities so far. “But the industry clearly understands this will not be the case forever,” Diggans said.   

Mackelprang is worried that AI could reduce the knowledge barriers that have historically prevented bad actors from developing bioweapons. Frontier AI systems, for example, seem to already outperform expert virologists on questions about performing complex laboratory procedures. 

But there is knowing and there is doing, and biological lab work is still hard. “Researchers spend years trying to make a protocol work even after consulting directly with others who have perfected that exact same protocol. I think AI can help someone ‘level up’ their laboratory skills, but I do not think AI can enable someone without any biological training to create a serious hazard,” Mackelprang told me.

That means that gene synthesis companies are still a primary chokepoint for anyone trying to produce a novel genetic sequence. Mackelprang’s main concern is that aspiring bioterrorists might use AI to generate harmful genetic sequences that can evade current or future screening systems. “In the near term, I think the likelihood of these types of misuse are quite low. But when the potential consequences are severe and technologies continue to develop rapidly, we have a responsibility…to develop reasonable prevention and mitigation options,” she said.

Maximizing the benefits of gene synthesis while minimizing the risks is difficult, but not impossible. That’s why Diggans and Mackelprang — along with Altman, Hassabis, and Amodei, as well as other gene synthesis providers, tech entrepreneurs, life science executives, and national security experts — signed the open letter calling for mandatory gene synthesis screening and recordkeeping of orders. 

Co-organized by the think tanks Institute for Progress and the Foundation for American Innovation, the open letter also calls for providers to record synthesis orders and sequence data to support biosecurity investigations “so that any threat that might evade initial screening can be traced back to its source…Awareness of traceability itself deters misuse.” This would, ideally, address Mackelprang’s concern that AI might eventually help bad actors evade existing screening protocols.

“Screening every DNA synthesis order before it’s manufactured is the kind of unglamorous, common-sense step that prevents a much bigger problem later,” DJ Kleinbaum, the co-founder of the biotech startup Emerald Cloud Lab, an automated lab scientists can access remotely, and one of the signatories, said. 

But Altman, Hassabis, and Amodei’s shared signatures may be the most meaningful evidence that the letter matters. For all their disagreements, they are well aware that their tools can be used for tremendous — even catastrophic — harm. 

AIxBio risk: A thing on which we can all agree

While it’s far from the first time frontier AI companies have spoken to AI-enabled biological risk, the open letter is the first place they’ve come together to do so in a single voice. “Support for screening does not depend on any particular view of AI,” the letter reads. “This is a rare moment of agreement across stakeholders that are often at odds.”

The letter calls for Congress to act now. “We applaud the legislative efforts currently underway,” the letter says, alluding to the bipartisan Biosecurity Modernization and Innovation Act, a bill that gives the Department of Commerce a year to develop new gene synthesis screening rules. The letter also suggests that US states should implement screening requirements based on federal and industry guidelines to create a unified national standard rather than an inconsistent set of laws.

The letter isn’t about applying biosecurity regulations to the AI companies themselves, which likely would have limited the number of tech signatories. (Though major companies do actively try to prevent their models from giving away dangerous biological knowledge, albeit not always successfully.) Focusing on screening is tractable, has the buy-in of several gene synthesis providers, and provides a concrete example of how AI can lower the barrier to doing both great and terrible things. And of course, it’s ultimately something a human being has to do at this point. 

The AI companies are actively thinking about the catastrophic risks that their technologies might enable. Anthropic is hiring a technical chemical, biological, radiological, and nuclear threat investigator for its threat intelligence team. In May, after launching GPT-Rosalind, a frontier model to accelerate life sciences research and drug discovery, OpenAI introduced Rosalind Biodefense, a program that allows trusted developers to use GPT-Rosalind to build biodefense tools. On June 4, the day after the open letter went live, security specialists at OpenAI and Anthropic served as panelists at the Bipartisan Commission for Biodefense’s meeting on AI and biological threats.

But according to Twist Biosciences’s Diggans, the best way to defend against misuse of AI models to design harmful pathogens is to use AI models as defense. These defensive models can be used to detect attempted misuse before anything happens. DNA synthesis companies can employ these models to ensure orders for highly-engineered sequences are given the same scrutiny and evaluation as orders for naturally occurring sequences.

“[Gene synthesis] companies have to agree to have their order screened not just against a list of sequences but by an AI that people agree is smart enough to recognize and thwart an adversary who’s trying to build a deadly pathogen,” David Haussler, the scientific director of the UC Santa Cruz Genomics Institute and a signatory of the open letter, told me.

Using AI to protect against AI

The good news is that this work is already underway. Last year, I reported that OpenAI provided $30 million in seed funding to biodefense startup Valthos, which develops frontier AI systems to detect biological threats and create medical countermeasures. Valthos’s co-founder Kathleen McMahon signed the open letter.

In September 2025, the Coalition for Epidemic Preparedness Innovations (CEPI) and philanthropic nonprofit Sentinel Bio created the Pandemic Preparedness Engine AI platform (sometimes referred to simply as “the Engine”). They’re taking a biosecurity-by-design approach, considering biosecurity risks from the outset. “This includes a multi-layered approach to biosecurity: from protecting biosecurity-sensitive data needed to train the AI to carefully managing who has access to the Engine and monitoring how they use it,” Sarah Carter, a biosecurity consultant at CEPI, told me over email. 

Users of the Pandemic Preparedness Engine would use AI prompts to interact with the system, similar to how people use consumer platforms. User prompts could be monitored in real time by a specialized AI agent built to assess the risk of misuse potential or attempts to “jailbreak” an LLM to get it to generate prohibited content, such as the “recipe” for assembling a deadly virus. 

Still, even commercially available technologies may present problems of their own. This week, Anthropic launched Claude Fable 5, a version of its highly powerful and restricted Mythos model that the company has aimed to make safe for public use. Claude automatically stops use of Fable if it detects requests involving cybersecurity, biology, chemistry, or distillation (attempting to extract Claude’s capabilities to train competing AI models), shunting those requests to a less powerful model. Users have complained that trying to discuss biology for legitimate purposes with Fable 5 results in the model refusing to engage or defaulting to less capable models instead. The Fable example shows that it’s possible to overcorrect, limiting the potential upside of using AI for the life sciences.

“The major providers of LLMs are doing their best to prevent the models from answering questions that would enable somebody to do something dangerous,” Haussler told me. “[But] the end product of jailbreaking an LLM that’s capable of teaching you how to build a deadly virus is that you now have an LLM that’s capable of teaching anybody how to build a very dangerous virus. And we don’t want that to happen.”

It’s here that the letter’s signatories hope they can stop a still-simmering problem before it comes to a full boil. “Mandatory synthesis screening is that rare case where a threat is clearly visible and substantial prevention clearly achievable before any crisis has occurred,” said Richard Danzig, a natural security expert who served as the 71st Secretary of the Navy under former President Bill Clinton. “Opportunities to act in advance are unusual in this field. I think we should take this one.”

The story of global health over the last few centuries has generally been one of great progress — vastly longer lifespans, far fewer women dying in childbirth, many fewer children dying from miserable diseases like measles and smallpox. But there is one often overlooked feature of modernity that has brought a new and enormous degree of mortality and injury to everyday life, a risk that falls most heavily on the world’s poorest people. It kills about as many people as the world’s deadliest infectious disease — tuberculosis — and it’s the leading cause of death globally for people in the prime of their lives, aged 5 to 29. It is one of the defining technologies of modern life, one of the 20th century’s most dangerous gifts: the car. 

Around 1.19 million people globally are killed by road crashes every year, according to estimates from the World Health Organization (some estimates put the number higher), and many times more — likely between 20 and 50 million — are injured, sometimes leaving them with life-altering disabilities. More than 90 percent of those deaths occur in low- and middle-income nations, although these countries contain only around 60 percent of the world’s cars. 

This century, humanity has halved the mortality rate for children under five and reduced AIDS-related deaths from their peak by 70 percent. But the number of people killed by cars has remained roughly the same for the last 20 years. As motor vehicles spread around the world — the total fleet has doubled over the past 20 years — the burden of those deaths has shifted increasingly to lower-income countries. Despite all the progress we’ve made against ancient natural killers, we’re making little against a killer we engineered ourselves. 

That’s not for a lack of known solutions, but rather because there’s been comparatively little attention paid to car crash deaths as a real global health issue until relatively recently. Unlike deadly maladies that are purely bad, cars do add value to society. Perhaps as a result, even though wealthy countries have brought down per capita road fatalities over the course of decades, deaths by car have still often tended to be discounted by policymakers and the general public as the price of progress and economic growth. It’s “one of the few public health problems where society and decision makers still accept death and disability on such a large scale as inevitable,” the late Dinesh Mohan of the Indian Institute of Technology wrote in 2019. 

“You can become very depressed,” James Leather, director of transport at the Asian Development Bank, told me in a recent conversation at the International Transport Forum summit (an event sometimes called the Davos of transportation). “Why is no one taking this seriously?” 

Of course, it’s not that literally no one is taking it seriously, but rather that cars have long been an underrated threat to human well-being. But that is, perhaps, slowly beginning to change. 

Why cars kill so many people in countries with so few of them

I am sometimes known as a bit of a car hater, devoting a lot of my consciousness to thinking about how the United States got locked into car dependence. Our car-oriented development pattern is part of the reason the US has one of the highest road fatality rates of any wealthy country. (But, listen, I own a car too, and benefit greatly from it! I am American, after all.) 

US car fatality rates may be an outlier by wealthy-country standards, but most low- and middle-income nations face far greater risk. Haitians and Ethiopians are more than three times more likely to be killed by a car than an American; Kenyans, Bolivians, and Thais are more than twice as likely. 

That alone is worth dwelling on. If you live in the US, consider that you probably know at least several people who’ve been killed in a car crash or who have loved ones who have, and that this proximity to sudden, violent loss is felt even more acutely in most of the world. Road deaths account for around 1 percent of all deaths in the US; globally, that figure is about 2 percent, and in a typical middle-income country like Vietnam, it is more than 3 percent. 

That might sound a bit surprising — and feels all the more unfair — in light of the fact that poorer nations do not have anywhere close to as many cars as wealthy ones do, and their residents travel fewer miles by car than people in rich countries do. If cars kill so many Americans because we simply drive so much, in the developing world, the problem is almost the inverse: A minority of people who can afford it ride in private cars, while everyone else walks, bikes, or rides a motorcycle, scooter, or three-wheeled vehicle like an auto rickshaw. And those outside of an automobile — known as “vulnerable road users” — often share space in the road with cars and are at high risk of being hit. 

Cars themselves in developing nations are often more dangerous for their occupants than vehicles in rich countries are, too. Weaker car safety standards and a reliance on imported old cars mean that people sometimes travel in vehicles that lack safety features long taken for granted in rich countries, including airbags and frames designed to absorb the force from a crash. 

Amid all this, cars and other motorized vehicles are spreading rapidly in the Global South — much more quickly than that transition took place in North America and Europe — and doing so before governments have built safer roads, vehicle standards, adequate trauma care, or robust traffic regulations. Many nations lack comprehensive laws governing what the WHO considers the five key behaviors that shape road fatalities: high speeds, drunk driving, seatbelt use, helmet use for motorcyclists, and child restraints in cars.  

In Southeast Asian countries, which have seen a massive proliferation of motorized vehicles since 2010, “maybe the infrastructure was designed when you didn’t have so many cars, and now all of a sudden you have twice the number of cars that you did before,” Nhan Tran, the WHO’s head of violence and injury prevention, told me. Road crashes are a major burden on the medical systems of these countries and exact staggering economic costs, amounting to about 5 percent of national GDP in Vietnam, for example. 

Meanwhile, as the total number of global road fatalities has stayed roughly constant for the last few decades, the gap between poor and rich countries has widened. Between 2010 and 2021, high-income countries, particularly those in Europe, saw dramatic decreases in car crash deaths, while deaths in the vast majority of low-income nations (which are predominantly in sub-Saharan Africa) increased, according to the WHO’s most recent report on global road safety. Across lower-middle-income nations, like India, the aggregate number of deaths and the per capita fatality rate stayed roughly flat. 

I asked Leather whether there was an easy, no-brainer intervention that could make a big dent in these deaths. He pointed, among other things, to helmets — in the Philippines, where he lives, national law now requires that helmets be made available with every new motorcycle purchase, though for that to work, people of course actually have to use them.

“If you go to New Delhi today, nearly every motorcycle rider wears a certified full-faced helmet. This was achieved through strong enforcement,” Kavi Bhalla, a professor at the University of Chicago’s Department of Public Health Sciences and an expert on global road safety, told me in an email. “In contrast, most other cities in India don’t enforce the helmet law, have very low helmet use, and this leads to many unnecessary deaths.”  

Poor countries don’t need to wait their turn for safer roads

Twenty years ago, two US economists published what became one of the most influential papers in the field of global road safety, on the relationship between a nation’s wealth and its traffic fatality rate. As countries get richer, they argued, motor vehicle ownership rises, and per capita car deaths rise in tandem. Eventually, as countries become wealthier — and as safer roads, vehicles, and traffic policies catch up with motorization — fatality rates start to fall, as they did across much of the industrialized world beginning in the early 1970s. That tipping point, the authors found, comes at around $8,600 (in 1985 international dollars) of per capita GDP. 

But this “economic determinism,” as Bhalla has described it, might be the wrong way of looking at the problem. It contributes to a sense that traffic carnage is inevitable until a nation becomes rich. But we would never argue that maternal mortality or malaria deaths can’t be significantly mitigated in low-income countries; in fact, we already know they have been. Although Europe, the US, and other high-income nations have steadily reduced car death rates over the last 60 years, Bhalla told me “it is a mistake to think that this has much to do with these countries being rich.”

Instead, “safety improved in these countries once they established national road safety agencies, gave them the authority to regulate what happens on the roads, and gave them a dedicated funding stream,” he wrote to me. “These agencies did what you would expect agencies to do. They identified the most common traffic safety risks in the countries, undertook investigations on how best to address these, and then made investments for large scale interventions focused on safer designs of cars and roads, coordinated enforcement programs, and emergency medical systems. Low and middle income countries can and should do this now.”

The WHO and other global organizations, along with some philanthropies, have been working to speed along such work over the last few decades, but the results have so far been somewhat underwhelming. The United Nations had aimed to halve global road deaths from the baseline of roughly 1.2 million by 2020, a goal we didn’t come anywhere close to reaching. On the other hand, world population has greatly increased in the last few decades, so holding the absolute number of traffic deaths constant is still a meaningful achievement: From 2010 to 2021, the global per capita road fatality rate decreased by about 16 percent. And in that period, Tran said, road safety has at least gained a lot more visibility among political leaders and civil society as a badly neglected public health crisis. 

Having missed the 2020 target, the UN now aims to halve road deaths by 2030. But we will “definitely not” meet that goal either, Bhalla told me. 

A core reason the global road fatality crisis has been so maddeningly obstinate is that the root of the problem is complicated, contested, and depends on one’s perspective. “It’s not the same as when you’re talking about Covid or HIV, where there is a virus” that we want to eradicate, said Tran. “When you talk about road safety, what is the virus?” Is it dangerous individual behaviors — speeding, drunk driving, refusing to wear a seatbelt? Is it deteriorating roads or a lack of sidewalks? Is it humanity’s growing dependence on cars themselves? 

Tran, like many road safety advocates today, calls for an approach that focuses on the most upstream cause of car fatalities — the proliferation of cars — and champions good urban planning designed to prioritize transit, walking, and cycling over the movement of cars. That would make safety an inherent feature of the transportation network and obviate the need for what Tran calls “quick fixes” to poorly designed systems.

WHO director-general Tedros Adhanom Ghebreyesus echoed that message in the agency’s 2023 road safety report: “As motor vehicles proliferate, countries are doubling down on transport systems built for cars, not people, and not with safety at their core,” he wrote. 

There’s a lot of wisdom to this, as the American experience over the last century well shows. The US experiment in car dependence has burdened us with a road fatality rate that rivals nations much poorer than us. Urban planners now widely agree that that car-dependent paradigm was a mistake, but now that it’s built out, it’s hard to claw our way out of.

But that lesson also requires some humility: Even a car hater like me can acknowledge that for many people in poorer nations, automobility offers a measure of freedom that rich countries have taken for granted for many years. And it would be a mistake to see simple interventions that can save tens of thousands of lives, and that were instrumental in bringing down car fatalities in rich countries, as mere Band-Aids. We need both approaches. Just as humans did with once-devastating infectious diseases, we will have to learn to see a person killed for simply trying to get somewhere not as a tragic act of God, but as the result of forces within our control. 

In just 10 days over the summer of 1854, 500 people died of cholera in the Soho neighborhood of London. The city’s population had more than doubled to 2.3 million people in the first half of the 1800s, and its sewage system could not keep up. But the streams of human waste flowing into the street and seeping into the water supply were considered unconnected to the cholera crisis. The prevailing theory of the day was that bad air — miasma — caused illness.

The English physician John Snow thought differently. Five years before the outbreak he had suggested that the diarrheal disease was actually caused by a waterborne infection rather than miasma. He soon had a chance to test his theory, mapping the location of cholera-related deaths in Soho. Snow realized that the victims used one specific water pump on Broad Street, and he persuaded city officials to remove the pump’s handle to prevent anyone else from using it. With the source eliminated, the outbreak, which had already passed its peak, ended in days. 

Though it took years for Snow’s theory to achieve widespread acceptance, his approach is central to modern epidemiology. Investigating the source of outbreaks can prevent new cases, but it also gives us a better understanding of diseases and helps manage public fear. Even when infections have stopped, outbreak investigations are useful to develop strategies for preventing — and, failing that, responding to — future outbreaks. 

Two recent outbreaks have demonstrated the necessity — and the challenges — of such investigations, almost two centuries after Snow’s pioneering work. The first was the hantavirus outbreak that dominated headlines last month. Then, on May 17, the World Health Organization (WHO) declared a public health emergency of international concern, the highest level of global health alert, in response to an outbreak of the deadly hemorrhagic disease Ebola in the Democratic Republic of the Congo (DRC), which, as of June 2, had killed 62 people, with 363 confirmed cases. It’s the 17th Ebola outbreak in the DRC and one of the largest on record. It has spread to neighboring Uganda, where, as of June 4, there are 16 confirmed cases, one confirmed death, and one probable case and likely death. 

The first confirmed case, a healthcare worker in Bunia, DRC, died on April 24, but the outbreak may have been spreading undetected since as early as January. Investigators haven’t identified patient zero — the index case — and still don’t know how this outbreak began. Abdou Sebushishe, a doctor working with the International Medical Corps in Goma, DRC, told CBS News that up to 20 percent of current patients are themselves healthcare workers. He estimated that it may be more than six months before the outbreak could be controlled, given that the disease is outpacing the current response.

Part of the challenge is that the current outbreak is caused by the Bundibugyo strain of Ebola, which is relatively uncommon and has a genome about 30 percent different from the Ebola viruses that usually spark outbreaks. Testing for more common variants didn’t pick up the Bundibugyo virus right away, and ongoing conflict in the DRC contributed to the delay and continues to make contact tracing difficult. Unlike other strains, the Bundibugyo virus has no approved therapeutics or vaccines.  

In the past, researchers have had some success identifying the index case of Ebola outbreaks. Investigators managed to identify the first patient of the 2014-2016 West Africa Ebola epidemic — the largest and deadliest in history, with more than 15,000 confirmed cases and 11,000 deaths — as a toddler in the west African nation of Guinea. What’s harder to definitively determine is how the boy, who died in December 2013 before the outbreak had been identified, contracted it. It’s possible that he came into contact with an Ebola-infected fruit bat or its droppings while playing in a hollow tree, but scientists can’t say for sure.

Investigating outbreak origins is inherently fraught and can lead to the international fingerpointing that characterized much of the Covid-19 pandemic. But it’s not primarily about assigning blame. Instead, knowing where and how outbreaks began informs how we respond to them, halt transmission, communicate to the public, and prevent them from happening again. It can identify high-risk regions and influence how public health officials monitor a disease. As the recent Ebola and hantavirus outbreaks demonstrate, however, that effort is often complicated by a host of factors, and the resulting uncertainty makes it that much harder to manage public health concerns efficiently and well. 

The curious case of Legionnaires’ disease in New York City

Our epidemiological tools have come a long way since John Snow used hand-drawn maps to identify the source of the Soho cholera outbreak. The value of these new tools lies in the information they generate — which is crucial to fighting outbreaks. 

Take the case of New York City’s biggest — and deadliest — outbreak of Legionnaires’ disease (LD), a bacterial infection that causes a severe pneumonia and has a fatality rate of 10 percent. By the time public health investigators detected it in the summer of 2015, dozens had already been hospitalized. It was the second-largest LD outbreak in US history, infecting 138 people and killing 16. 

The initial epidemiologic investigation started with contact tracing to find the source of the disease, but the results didn’t suggest any shared exposures. Cooling towers, which provide water for air conditioning systems in the form of an inhalable mist, had been involved in previous LD outbreaks, but officials didn’t know how many cooling towers there were in the city or how well-maintained they were. 

Investigators ultimately located and tested 55 cooling towers in the South Bronx, where cases were clustered, for Legionella. They identified the source: a single cooling tower atop the Opera House Hotel. The hotel disinfected the tower, and New York’s City Council passed new regulations requiring every building in the city with a cooling tower to register it with the health department, test it every 90 days, and remediate it if Legionella was found. 

Within a year, the health department inspected almost 80 percent of the city’s towers — detection and disinfection that would have never been conducted otherwise. No large LD outbreaks emerged — until inspections declined in 2025. “Regulations do not enforce themselves,” Jay Varma, a physician and epidemiologist who served as incident manager for the 2015 New York outbreak, wrote last year in Healthbeat. “The Covid pandemic has sparked a strong backlash against government authority, and austerity budgets are now starving public health agencies. Infections may be inevitable, but outbreaks are a choice.”

Cholera and LD are waterborne, but Ebola and hantavirus, which first cross over to humans from animal reservoirs, present a different challenge. 

The challenge of hantavirus and Ebola

“The end of the world, the beginning of everything” is the motto of Ushuaia, Argentina, the southernmost city on the planet, where tourists flock to watch birds and embark on cruise ships. It’s the main gateway to Antarctica, making up 90 percent of all cruise departures to the continent. 

It’s here that a Dutch couple may have contracted the Andes virus, the only strain of hantavirus known to spread from person to person, before sparking an outbreak on the MV Hondius. The Argentinian government’s prevailing theory is that the couple got infected while birdwatching at a landfill in Ushuaia before the cruise, coming into contact with the rodents that carry the Andes strain. 

Well, maybe not. 

“The current theory of a couple birdwatching in southern Argentina may not be plausible, because the [long-tailed pygmy] rice rat that is responsible for spreading the Andes strain of the virus is usually found in northern Argentina or Chile, and we know the birdwatching at the landfill occurred in the southern part of Argentina,” Omer Awan, a physician and public health expert, told me over email. There have been no recorded cases of hantavirus in Tierra del Fuego province, where Ushuaia is located, before. 

“Understanding the origins of the outbreak will be helpful in guiding interventions like rodent control, isolation protocols, and…how the rare Andes strain of Hantavirus is transmitted,” Awan said. “[And] identifying the source of the [2026] ebola outbreak can influence response strategy and how public health officials monitor the virus.”

Delayed detection and human movement — especially for illnesses like hantavirus and Ebola that can incubate over the course of weeks — make tracing the source of an outbreak difficult, even in the best of circumstances. We still don’t know the original source of the first Ebola outbreak in 1976, which occurred in two simultaneous waves. Debates still rage over whether Covid-19 emerged naturally through zoonotic spillover — the virus jumping from an animal host to humans — or if it potentially escaped from a lab in an accident. We know that the hantavirus and Ebola outbreaks are natural in origin, but there are still international efforts to shift the “blame” from Argentina to neighboring Chile, especially with economic interests on the line.

Such spillover events have only become more likely as humans destroy ecosystems and infringe on animal habitats. Climate change exacerbates existing infectious disease risk. “Because of our choices as a society, there’s a one-in-five chance that another pandemic will occur in the next decade that will kill at least 25 million people,” Neil Vora, the executive director of Preventing Pandemics at the Source coalition, wrote in Time Magazine. 

Determining the source of outbreaks is even more difficult — and politically perilous — in the post-Covid era. The US and Argentina have pulled out of WHO. Global health funding cuts, on the part of the US as well as other countries, have weakened our biosurveillance architecture and ability to effectively respond to infectious disease. 

Compared to Covid, the scale of the 2026 Bundibugyo and hantavirus outbreaks are small. It’s still proving hard to get answers. That’s going to be a serious problem whenever the next pandemic arrives — and it is a matter of when, not if. 

An evolving threat landscape

Although we face escalating spillover risks from habitat destruction and climate change, we can’t count on the next global infectious disease threat being naturally occurring in origin when it does come. 

“It’s very clear that artificial intelligence capabilities are advancing incredibly rapidly,” Jaime Yassif, senior advisor for global biological policy and programs at the Nuclear Threat Initiative (NTI), told me. “[That could] make it easier for novice actors to engineer pathogens that we [already] know about or for sophisticated actors to engineer novel pathogens that are more dangerous than what’s found in nature.”

If there is an outbreak of uncertain origin — where it’s unclear if it’s natural, accidental, or deliberate — we lack robust international mechanisms that can investigate the source and quickly arrive at a conclusion. That would make it harder to address the source proactively, whether that means stopping future natural spillover events, preventing lab accidents, or holding bad actors to account. 

Public health professionals would need to take additional precautions if there was a risk of a deliberate outbreak, as we saw with the 2001 anthrax attacks, where letters laced with Bacillus anthracis were sent in the mail, infecting 17 people and killing five. A naturally-occurring anthrax exposure would have required a different response, since a bioterrorism investigation has to contend with the additional challenge of determining criminal responsibility. 

And as we’ve seen with the debates around Covid-19 origins, suspicion that something was caused by human activity can be incredibly corrosive to international trust, making necessary geopolitical cooperation in the face of outbreaks significantly harder. 

NTI identified that preparedness gap and proposed a Joint Assessment Mechanism to identify the source of outbreaks of uncertain origin. It would be housed in the UN Secretary-General’s Mechanism for Investigation of Alleged Use of Chemical and Biological Weapons (UNSGM) in order to pull together different components of the UN system and bridge security and public health. 

That project (which I supported and advocated when I worked at NTI from 2022 to 2024) is currently on pause. “We still think it’s a vital gap and really important, but we just couldn’t get the political will to move it forward in the system, notwithstanding the significant support for it internationally in various quarters,” Yassif said.

We are simply unprepared domestically and internationally to prevent, detect, and respond to global infectious disease threats. Emerging infectious disease outbreaks threaten us all, and we are nowhere near where we should be in order to protect vulnerable populations and countries around the world. While the current Ebola and hantavirus outbreaks are very unlikely to become pandemics on the scale of Covid-19, they’re still dangerous and deadly. Unless we can determine where and how they began, we’ll be ill-equipped to stop them from recurring. And next time, things could be far worse.

This story was originally published in The Highlight, Vox’s member-exclusive magazine. To get access to member-exclusive stories every month, become a Vox Member today.

For years, the “Make America Healthy Again” movement was driven by moms.

Concerned about the safety of childhood vaccines and about chemicals in the food their kids were eating, they helped propel Donald Trump to the White House — and Robert F. Kennedy Jr. to the role of the nation’s top health influencer — with a message centered on fear for the next generation. 

Now, that next generation is here.

The latest MAHA advocates to gain public attention are women in their teens or early 20s. Lexi Vrachalus, 20, posts videos of her seed-oil-free, sugar-free meals, snacks, and shopping trips. In a post around Easter, she made her own Peeps with maple syrup and beef gelatin.

Her message: “You can take back health into your own hands,” she told me. “You have the power to heal your body.”

She and other influencers, like the young filmmaker Grace Price and clean-living maven Ava Noe, are creating videos with a younger sensibility than their forebears — think baking sourdough for siblings rather than talking about kids’ vaccines. And their version of MAHA (that’s Make America Healthy Again, for the uninitiated) is breaking through to American teens.

“I get questions from my younger audience like, how can I encourage my parents to eat healthy?” Vrachalus said. “Or, how can I eat healthy when all my parents do is buy junk food?”

On the surface, there’s nothing wrong with young people trying to eat healthy. But educators and misinformation experts are worried about what comes next: Among adults, MAHA influencer culture has served as a funnel for a host of beliefs and behaviors that start with skepticism, veer into suspicion of all authority, and end up with actively dangerous behavior, including a resistance to vaccines that has led to outbreaks of disease.

“There’s this focus on healthy foods and environmental concerns, but running under the surface of some of those more superficial connections is this idea that there’s this cabal,” said Whitney Phillips, a professor of information politics and media ethics at the University of Oregon. “There’s this kind of conspiratorial thinking that ‘they,’ coded as liberal, are lying to you.”

So far, polling shows that young people are less likely to identify with MAHA than Americans in their 30s and 40s. But MAHA-inflected wellness videos are reaching more teens, and there’s evidence that more young people are falling for health misinformation that they see online. 

In a 2024 survey by the News Literacy Project, 80 percent of teens said they saw conspiracy theories on social media platforms, and a majority of those teens said they were inclined to believe one or more of those theories. The second most common type of conspiracy theory mentioned by teens in the survey (after “aliens & UFOs”) was content around Covid-19 and public health issues.

The rise of young MAHA influencers has educators and other experts asking what they can do to help Gen Z and Gen Alpha Americans — a group already deeply distrustful of institutions and authorities — distinguish reality from toxic misinformation. If teachers, families, and policymakers hope to thread that needle, they’ll have to do more than just respond to false claims point-by-point — they’ll need to address the sources of discontent and disaffection that may be pushing young people toward MAHA in the first place. 

The new face of MAHA

If you had to picture the MAHA coalition, you might think about a group of millennial and Gen X moms, banding together over their opposition to vaccine mandates and food additives. Or maybe you’d call to mind someone like Andrew Huberman or Joe Rogan, male podcasters in their 50s extolling the virtues of supplements and protein to an audience of “Huberman husbands.”

And sure, that’s today’s MAHA — a recent Politico poll found that those most likely to identify with the movement were Americans in their 30s and 40s. 

But tomorrow’s MAHA is coming, and the teen girls and young women emerging as MAHA influencers show us what it might look like. 

Vrachalus, for example, has more than 170,000 followers on Instagram — not as many as established voices like Vani Hari with follower counts in the millions, but a respectable reach for a creator, especially one so young. Vrachalus recently made a video with Kennedy, the Health and Human Services secretary, to promote the new federal dietary guidelines. 

When she was diagnosed with anorexia at 13, a dietitian told her she’d have to eat “junk food” in order to get better, Vrachalus told me. Instead, “I started to research, and I realized that basically everything in the grocery store is ultraprocessed junk food,” she said. 

Today, “I heal my body using real food that God created and designed us to eat,” she said.

Ava Noe, 18, has about 26,000 followers on her Instagram account, @cleanlivingwithava. She hopes to show young people that they “don’t have to be a certain age to take their health into their own hands,” she told me. “It’s never too early to start maximizing your health.”

For Noe, that looks like anything from searching for “clean” food at the grocery store to medically controversial practices like making her younger siblings use fluoride-free toothpaste. 

Meanwhile, some older MAHA influencers feature their young children as a way to get their message out to families. Gretchen Adler, for example, a creator with over 500,000 Instagram followers, recently posted a video in which her 9-year-old daughter explains why she makes her own gummy candy from orange juice and gelatin. Storebought gummies, she says, are “pure trash.”

“I’ll always say to show this to your child,” Adler says of her daughter’s appearances on her feed. “That’s the way that we can inspire these people or these young children, is when they see another child that they can relate to.”

The anti-seed-oil to anti-vax pipeline

Young people may be an especially receptive audience for the message that they can take their health into their own hands.

Gen Z Americans “feel very disillusioned by organizations in society and institutions, including, of course, medical institutions in the wake of Covid,” said Melissa Deckman, CEO of Public Religion Research Institute and author of The Politics of Gen Z. 

They are more likely than their elders to rely on friends and family or social media for health advice, and less likely to rely on doctors. They also distrust news outlets that could give them fact-checked information about health claims.

At the same time, young people are concerned about their health, experts say. “I have seen students become more inclined towards trying to think about wellness because they need to, because they’re not doing well,” said Phillips, who has taught university students for 18 years. “College students used to be some of the most carefree people in the world, and that just isn’t what is true anymore.”

The result, some say, is a population especially primed to listen to MAHA messaging delivered by influencers their own age. “These are beautiful young people that are promoting it, and they’re thinking, old people don’t know what they’re talking about. Here’s this cute 22-year-old who’s explaining this to me,” said Bertha Vazquez, who runs Generation Skeptics, a program that trains teachers to respond to misinformation.

However, experts worry that some MAHA content could be harmful, not helpful, for young people’s health. Such content often promotes the idea that consumers need to be vigilant about their food to avoid “toxins,” or that products can be divided into “real food” and “not-real food.” 

“That black-and-white thinking is very dangerous for people that have vulnerability to eating disorders,” Zoë Bisbing, a psychotherapist specializing in disordered eating, told me. 

Vrachalus isn’t convinced that opposing processed food promotes disordered eating. “Our great-great-grandparents, they didn’t have Oreos, they didn’t have ice cream,” she said. “I just don’t think that our great-great-grandparents had eating disorders because they didn’t have fake food.”

But eating disorders aren’t the only concern experts have raised. Some fear that exposure to MAHA content could push young people toward harmful behaviors that Kennedy and other movement leaders have supported, from using beef tallow as sunscreen to avoiding vaccines or chemotherapy. 

 “When they do get a dangerous virus, or they do get cancer, or they do have a child, the big concern is that they will not get the vaccines and the standard care,” Vazquez said.

Vrachalus and Noe don’t talk as much about vaccines or avoiding modern medicine as older MAHA and MAHA-adjacent influencers do. “I’m not anti-Western medicine at all,” Vrachalus told me. “If I break my arm, I’m going to the doctor tomorrow.”

But previous generations of MAHA and wellness influencers have cast doubt on proven treatments from the measles vaccine to chemotherapy, sometimes while pushing dietary supplements that are unproven and unregulated, or foods like raw milk that can cause serious illness.

Some young people are already subscribing to this kind of thinking — 18-year-old Shelby Gwinn, who is studying to be a dietitian, recently told the New York Times that “all pills do is cover up a problem instead of getting to the root cause,” and that today she takes 30 supplements to manage her eczema. “I do think the government should step in if a food company is putting absolute trash or chemicals in their food products,” she said — “but then again, I don’t trust the government.”

There’s a long history of wellness movements shading into conspiracy theory, Phillips told me. This anti-government, anti-medicine thinking began to creep into many wellness spaces, including yoga studios, around the time of the pandemic. 

“The messaging is basically this idea that you can’t trust doctors, you can’t trust the medical establishment,” Phillips explained. “They are trying to poison you.”

Getting young people to trust science again

In a polarized political landscape in which many young people are disillusioned with traditional news sources, conspiracy theories can be especially difficult to counter. That’s doubly true since so many young people really have been failed by their doctors, their government, and their world. 

“There are so many different ways that institutions have really genuinely let people down,” Phillips said. “But being able to make those kinds of critiques is different than this kind of conspiratorial attitude towards institutions.”

Teaching young people to think critically about information, whether it comes from an authority figure or a content creator their own age, may involve separating that information from politics. 

Melanie Trecek-King, a biology professor at Massasoit Community College and founder of the website Thinking Is Power, likes to start with European witch trials. She helps her students evaluate the beliefs about witchcraft that led to these trials, the evidence presented against accused “witches,” and the harms — including torture and executions — that false beliefs caused. 

By choosing examples from the past that aren’t personal for students today, she avoids putting them on the defensive. And once they’ve learned the process of evaluating information and evidence, “then they will make the connection in the real world,” she told me.

Not everyone can take a class like Trecek-King’s. But educators say it’s crucial for science communicators to meet young people where they are, whether that means posting on platforms like Instagram and TikTok or responding to questions about health without judgment or shaming.

“We have to be going to the places where people are,” said Jessica Knurick, a science communicator and content creator who has a PhD in nutrition science. Too often, scientific and medical experts take the attitude that “you should just listen to us because we’re us, instead of talking to people on a human level and understanding where their concerns are,” Knurick said. 

Getting expert information to where teens and other young people can see it will require changing professional norms that discourage doctors and tenure-track scientists from being on social media, Knurick said. It will also require finding ways to compensate experts for their time in a social-media economy that doesn’t always reward sober fact-checking.

But more science communicators and groups that serve young people are rising to the challenge. And it’s possible that young people’s tendency to question everything can be part of the solution.

“These MAHA influencers, they’re so confident in their claims, and you’ll never see a scientist like that,” Vazquez said. “Science is never about 100 percent certainty.”

That’s something educators can teach students, Vazquez said: “If someone’s so cocksure of themselves, then that’s a red flag.”

One of the clearest success stories in US healthcare over the past 20 years has been the dramatic decline in the number of Americans without health insurance. In 2010, the year the Affordable Care Act was enacted, 16 percent of the population lacked coverage. By 2025, according to estimates from the US government, that figure was cut nearly in half, to 8.3 percent.

The increase in coverage hasn’t been a panacea; even people with an insurance card can struggle to afford their medical bills or to secure a doctor’s appointment. But with the US standing alone among its international peers in its failure to offer universal healthcare, it represented significant progress toward ensuring every American had a basic level of access to routine medical services.

Now, however, those gains are about to be reversed.

Last year, when drafting their One Big Beautiful Bill, Republicans had a chance to strike a blow against the ACA — a law they’d vilified for years — 15 years after its passage and eight years after failing to repeal the law in President Donald Trump’s first term. They established work requirements to target the people covered by the ACA’s Medicaid expansion and allowed subsidies that had helped millions of people to buy private coverage on the ACA marketplaces to lapse.

As a result, millions of Americans are dropping their health insurance this year, and millions more are expected to lose their coverage in the years to come.

The uninsured rate has spiked before, but it’s usually the byproduct of an economic crisis; people lose their jobs, and they lose their coverage. What makes the current turmoil different is that it is entirely a matter of policy choices. 

Now, millions of Americans will pay the price.

“I don’t think there’s any historical precedent for the rollback in federal support for health coverage coming with the cuts in Medicaid plus the expiration of enhanced ACA premium subsidies,” Larry Levitt, executive vice president for health policy at the healthcare think tank KFF, told me. “The expected effects of OBBBA on coverage are self-inflicted and dwarf even the historical losses due to changes in the economy.”

ACA marketplace enrollment is projected to shrink dramatically in 2026

One of the major ways that the ACA expanded health insurance coverage was by setting up insurance marketplaces where individuals and families could purchase private health plans with the help of government subsidies.

Enrollment in those marketplaces has ballooned — particularly since 2021, when Democrats in Congress approved an expansion of the ACA’s financial aid that made more people eligible for assistance. Prior to 2021, there had been a strict cutoff at 400 percent of the federal poverty level (about $64,000 for an individual in 2026, or $132,000 for a family of four). Anybody who made a higher income was ineligible for aid. After 2021, anybody could qualify for ACA subsidies, and their insurance premiums were capped at a percentage of their income. (The subsidies were initially authorized for two years and, then, were extended to 2026 through the Inflation Reduction Act.)

It seemed to have plugged one of the obvious holes in the healthcare law: While many people below 400 percent of the poverty level had enjoyed both mandated comprehensive coverage and new government subsidies that offset any increases in costs, people above that threshold had been subjected to significant premium hikes since the ACA passed. Now, they were able to access the same subsidies, and sign-ups boomed. Marketplace enrollment grew from 9.8 million Americans in 2019 to 22.3 million in 2025. 

But, to keep down the cost of their legislation and get it passed with a narrow Senate majority, Democrats allowed the new subsidies to expire in 2026. Then, Trump won the 2024 presidential election, and Republicans took control of Congress. The GOP decided not to extend the subsidies, despite some bipartisan efforts to pull together a plan. When people went to sign up for their health insurance for 2026, many of them no longer had access to financial aid. I spoke last year with some of those people. One family was preparing to allow one parent and child to become uninsured so they could afford a health plan for the other parent who has an autoimmune disease. A young man with asthma also expected to go without coverage after his previous plan ($100 per month and no deductible) was no longer available, and the cheapest replacement he could find was $282 per month with a $10,000 deductible. He told me he was banking on being able to pay for his medication out of pocket or getting it through a charity service.

So, we knew some people would drop their insurance as a result of the expired subsidies, but it was hard to be sure how many. Now, we’re starting to get hard data, and it does not look good. Based on KFF’s preliminary analysis of enrollment data and premium payments, about 4.7 million fewer people will actually end up being enrolled in an ACA marketplace plan in 2026 compared to 2025 — a 21 percent drop in a single year.

Work requirements are going to knock millions of people off Medicaid

The ACA’s other major coverage provision was the expansion of Medicaid eligibility to any American with an income at or below 133 percent of the poverty level (about $21,000 for an individual in 2026, or $44,000 for a family of four). It replaced the preexisting patchwork system for eligibility that created significant differences across states — in particular, millions of childless adults, some of whom were living in deep poverty but had been left out of the program in many states before the ACA, now qualified for Medicaid. 

As of June 2025, more than 16 million Americans who became newly eligible for Medicaid through the ACA had been enrolled in the program, making up nearly a quarter of all Medicaid enrollees.

Republicans in Congress had been sharply critical of Medicaid expansion, even as many GOP-led states adopted it, and 2025’s OBBBA gave them a chance to roll it back. They approved, for the first time, national work requirements for Medicaid, targeted to expansion-eligible enrollees, and made several other technical changes to constrain states’ Medicaid financing. People on the program will be required to work or perform other approved activities for at least 80 hours per month or show they should be exempted from the requirement. Otherwise, they could lose their benefits.

And based on what we know from historical precedent, many of the coverage losses won’t be because people are actually ineligible for Medicaid, but because of the administrative burden of complying with these new requirements, even if you are working, or if you are someone — like a pregnant person — who is supposed to be exempted. Arkansas is the only state to implement Medicaid work requirements prior to the OBBBA, and only a fraction of the people required to submit work activities to the state actually did so; many of the people who lost coverage lost it because they failed to turn in paperwork. 

The Medicaid population is, by nature, hard to reach. This group is lower-income and might work irregular hours, move around more, or have less access to the internet. It’s easy for people to fall through the cracks.

The OBBBA’s requirements go into effect nationally in January 2027 (after this year’s midterm elections), but some states are instituting them early. Nebraska implemented work requirements on May 1, Montana and Arkansas are starting theirs on July 1, and Iowa will adopt the requirements on December 1. Then, starting on January 1, 2027, they will apply in every state.

The coverage losses are difficult to project, and they could take time to accrue, but they are expected to be sizable. The nonprofit research group RAND estimated Medicaid enrollment will drop by 7.6 million people by 2034. 

And they, much like those people dropping ACA coverage, will lose more than just their insurance card. Health insurance, even with its shortcomings, does a lot to help people. Americans with health insurance accrue less medical debt. They are more likely to go to routine medical appointments and receive routine screenings. Prior research on Medicaid expansion’s effects has estimated that it saved tens of thousands of lives.

In other words, the coming increase in the uninsured rate will do more than change some percentage points on a spreadsheet; it will make it harder for millions of Americans to stay healthy and stay alive.

Shortly after brandishing his infamous chainsaw on a conservative conference stage last February, Elon Musk attended a Cabinet meeting where, giggling slyly, he admitted to having “accidentally canceled” Ebola prevention in his haste to obliterate the US Agency for International Development (USAID).

“We restored the Ebola prevention immediately,” he added coolly at the time, “and there was no interruption.” That claim has since proven to be disastrously, profoundly untrue. 

On May 17, the World Health Organization declared a rapidly spreading Ebola outbreak in the Democratic Republic of the Congo and Uganda a “public health emergency of international concern,” only the ninth-ever time the agency has made that designation. In the weeks since, at least 220 people have died of the highly fatal virus, and more than 900 suspected cases have been identified so far. It is already the third-largest Ebola outbreak on record.

And yet, that toll is likely a tremendous undercount because, as the New York Times reported from the ground this week, “only a trickle of tests are being processed every day” in the cities most affected by the outbreak. “The virus is far ahead of us,” Ahmed Mahat, a manager with International Medical Corps, told the Times. “And it’s spreading fast.”

In fact, publicly known cases are rising exponentially faster than in any prior outbreak, including the largest ever, West Africa’s catastrophic outbreak in 2014, and the second-largest in 2018. By the time this outbreak was declared, hundreds of people had already been infected.

When you stop looking, you can’t see

Why did this outbreak spread so quickly? Part of it was the virus itself, a rare Bundibugyo strain of Ebola, which is harder to diagnose and for which there are no vaccines or treatments. (At least, not yet.) Another reason is that this outbreak began in a remote province of eastern Congo, an active war zone, where what health systems exist have been ravaged by decades of armed conflict. 

As if the odds weren’t already stacked enough, however, this outbreak broke out under the heavy shadow of US foreign aid cuts that, among other calamities, gutted the world’s Ebola detection and response apparatus last year. Despite Musk’s earlier assurances, US-funded programs to detect new Ebola cases and dispatch a response were indeed frozen under the Trump administration, according to Stat. US cuts also indirectly contributed to the outbreak by weakening local health systems and stockpiles. 

Altogether, the US Department of Health and Human Services disbursed about $10 million to Congo last year, down from $33 million the year prior, Stat noted. USAID sent $693 million in aid to Congo last year, down from nearly $1.2 billion in 2024. 

Cuts to disease surveillance meant that this virus took longer to identify than it should have. And with cuts to local health systems, it’s now much harder to come by the tests, nurses, doctors, and protective equipment needed to stop the spread. 

“It’s so bad. It’s so bad,” Jean Kaseya, director-general of the Africa Centre for Disease Control and Prevention, told Devex. The Africa CDC’s role in quelling outbreaks has become even more important as wealthy countries have retreated from the global health stage, but it is impossible to fill all of the medical surveillance gaps left by the US withdrawal of support, he said. “No one can give you the magnitude of this outbreak.”

Bleeding out

The US has done some course correction since the outbreak began. Last week, the State Department pledged $23 million in emergency funding for Congo and Uganda, plus the deployment of a disaster response team and enhanced involvement from the CDC, which says it’s been actively coordinating with local health agencies. At least some lost funding should have also begun flowing back to both countries through their bilateral aid deals with the US. 

But when you lose a limb to a chainsaw — even a “chainsaw of bureaucracy” like the one Musk dragged across a stage — you can’t expect a bandaid to make up for the damage. Beyond the money, the US withdrawal from the WHO and other policy decisions have had a deeply destabilizing effect on global health systems, which no doubt helped bungle this outbreak response. In many cases, the disease experts and researchers who were once in charge are simply not there anymore. 

Given the outbreak’s virulence so far, things will probably get significantly worse before they get better. While the majority of cases have occurred in Congo so far, Robert Redfield, former head of the CDC, predicted last week that the virus could soon spread to neighboring countries like Tanzania and South Sudan. Researchers have rapidly begun development on a new vaccine for the deadly virus, but even in a very best-case scenario, it will take months to roll out. In the meantime, health workers will continue to play catch-up to a virus that now has a massive head start.

As Nicholas Enrich, the former top global health official for USAID, told the New York Times last week: “In a time when hours matter, we’re delayed by weeks.”

Arsenic-contaminated groundwater affects more than 230 million people living in 108 countries. About 180 million of these people live in the Indian subcontinent (which includes Bangladesh, Nepal, and Pakistan, in addition to India) and Southeast Asia. The Indian state of Bihar, which borders Nepal, has several regions with extremely high levels of naturally occurring arsenic in their groundwater.

In Bihar, silt from the Himalayas containing arsenic and other heavy metals is routinely deposited in floodplains and seeps into the groundwater below. This phenomenon puts up to 21 million residents in Bihar at risk of consuming arsenic-contaminated water each day. Arsenic is a carcinogen that has also been linked to diabetes, pulmonary disease, cardiovascular disease, and infant mortality.

Though Bihar has close to 600 groundwater treatment units designed to filter out arsenic, a recent study of 98 units found that 90% of them were installed in parts of the state where groundwater arsenic levels were within the World Health Organization’s permissible limits (below 10 parts per billion)—which means almost all the communities that need these units the most still do not have access to them. The research was published in Groundwater for Sustainable Development.

“Some of the areas with these units had reported a higher prevalence of gallbladder cancer, which is associated with arsenic poisoning. But we found that it was the food that was the main source of arsenic exposure, not groundwater,” said Arun Kumar, a study author and senior scientist at Mahavir Cancer Sansthan & Research Centre in Patna, the state’s capital city. “In the last decade, we have observed drastic changes in groundwater arsenic levels in Bihar. Along with that, the cancer burden has also reduced in some parts of the state.”

In another city, Buxar, Kumar and his colleagues observed levels of arsenic of up to 1,900 parts per billion in the groundwater in 2015. But when the researchers retested that region’s water samples last year, the arsenic levels had gone down to 100–200 parts per billion.

“We hypothesize that because Bihar is prone to earthquakes, the seismic activity might have changed the properties of sediments and silt in groundwater. And perhaps, at some stage, those regions with the groundwater treatment units had experienced arsenic contamination,” added Kumar. “It is still a mystery to us” why the levels changed so drastically.

Ditching Groundwater for River Water

Kumar acknowledged that in the past few years, there has been a mushrooming of public and private groundwater arsenic treatment units in regions located within 10 kilometers (6.2 miles) of the Ganges River in Bihar. The majority of the 98 units included in the study were installed by the state government from 2016 onward. The researchers observed that privately owned units underwent regular maintenance, unlike many of the government-run units.

The corresponding author of the study, Laura Richards, a professor of water resources and geochemistry at the University of Manchester, explained that regions close to the Ganges River may have been given higher priority mainly because they are situated along major roads and highways, making them easier to access than inland Bihar.

“Much of the previous large-scale groundwater testing conducted in Bihar was limited to the 6-mile stretch on either side of the Ganges River. The issue with that is that the regions selected for arsenic remediation units were likely based on nonrepresentative spatial sampling of the state, and those locations might not have necessarily covered all areas with arsenic contamination in the groundwater,” said Richards. “Arsenic distribution across the state is really quite heterogeneous.”

The researchers further found that in 10% of the locations where groundwater arsenic treatment units were installed by the state government, high levels of fluoride posed a greater public health risk than arsenic, suggesting that governmental policies were rolled out without site-specific water quality monitoring and testing.

In addition to arsenic and fluoride, the groundwater in different parts of Bihar has high levels of manganese and iron. Currently, the state has more than 3,000 groundwater treatment units for arsenic, fluoride, and iron. However, Kumar said a better solution would be to look to other sources for drinking water and to ensure water treatment centers are properly maintained.

“People would be a lot safer if they stopped consuming groundwater altogether,” Kumar said. “This is why the state government has started treating and supplying water from the Ganges River to villages. They have already started doing it in two districts and plan on expanding the supply of river water.”

“Alluvial or sand-rich aquifers are the main culprits of arsenic-contaminated water in Indian terrains,” said M. Santosh, a professor at the China University of Geosciences in Beijing who was not involved in this study. “This study clearly shows how we can rectify remedial measures on a local level. We should encourage more such studies on how to tackle this problem.”

—Anuradha Varanasi, Science Writer

Citation: Varanasi, A. (2026), In Bihar, groundwater treatment units were installed in regions that didn’t need them, Eos, 107, https://doi.org/10.1029/2026EO260168. Published on 21 May 2026.

Text © 2026. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.

For the first time in the 21st century, the United States has approved a new sunscreen ingredient. Well, new to us.

It’s called bemotrizinol, also known as BEMT, and it’s been available in Europe and Asia for years. But the peculiar way that sunscreen is regulated in the United States — as an over-the-counter drug rather than a cosmetic — had long prevented it from coming to American store shelves. 

In 2020, however, Congress ordered the Food and Drug Administration (FDA) to overhaul its sunscreen approval process, and in 2024, DSM Nutritionals, which manufactures a bemotrizinol-based sunscreen, asked the FDA for approval. After a review of relevant safety and efficacy data, bemotrizinol has become the first new sunscreen ingredient to be approved for sale in the US since the late 1990s. The Environmental Working Group, which has lobbied for bemotrizinol’s approval since 2019, called its approval “a monumental victory for health and wellness.” 

Dr. Adewole Adamson, who is a dermatologist and assistant professor of internal medicine at the University of Texas at Austin, agreed that this is a win for consumers. “We haven’t been able to really have any innovation in US-based sunscreens since last millennium,” he told me. 

Sunscreen use has ticked downward in the US, at the same time that concerns about sunscreen seeping into your body and causing adverse health effects have risen. BEMT’s boosters hope it can change that trend by promising broad protection, a more aesthetically appealing application, and less risk of it permeating your skin.

Sunscreen is already complicated, as Vox’s Allie Volpe covered in her 7 burning questions about sunscreen explainer. Now there’s a new ingredient to consider. Here’s what you should know.

Sunscreen and the sunscreen backlash, briefly explained

The sun emits a spectrum of ultraviolet rays, including two types — UVA and UVB — that can burn your skin if you are exposed for too long without protection. 

That is why experts advise consumers to make sure they are buying “broad spectrum” sunscreen, which means it provides protection against both kinds of UV rays. Those products usually combine several different agents (or “filters”) that protect against different parts of that spectrum. 

“Some filters only cover part of the spectrum, so you have to combine a bunch of them in order to get that broad-based coverage,” Adamson said. 

Sunscreens are either “mineral” or “chemical.” Both types are equally effective if used correctly, assuming they have the same sun protection factor, or SPF, but each come with their own trade-offs. Mineral sunscreens leave unsightly white residue, while chemical sunscreens have faced widespread safety concerns in recent years.

The major shift came in 2019, when the FDA announced an overhaul in its safety assessment of some of the most popular sunscreen ingredients, sparking a backlash against chemical sunscreen in particular. The agency said that the two ingredients primarily used in mineral sunscreens — zinc oxide and titanium dioxide — were generally regarded as safe for human use. Two ingredients (aminobenzoic acid and trolamine salicylate) were said to be unsafe, and more than a half-dozen other ingredients used in chemical sunscreens were left a question mark due to “insufficient data.” New research soon followed that suggested that the ingredients in chemical sunscreens could seep into your blood and body in concerning concentrations, raising the specter of uncertain long-term health effects.

In response to the new findings and the doubts they raised about such a widely used product, anti-sunscreen advocacy spread, bolstered by the broader wellness and MAHA movements. As the Washington Post described, some people online boasted of stopping their sunscreen use — despite its clear effectiveness in preventing skin cancer, which kills thousands of people in the US every year — and promoted DIY formulas featuring, for example, oil and butter. (They do not confer the same protection.) Some influencers have even argued for the health benefits of more sun exposure.

One consumer analysis found that the percentage of Americans who believed sunscreen is toxic grew from 17 percent in 2021 to 24 percent in 2025. And, at the same time, the share of people who reporting using sunscreen at all has slightly declined.

Bemotrizonal is broad spectrum and could be more cosmetically appealing

Into that messy context comes a new sunscreen ingredient. 

A big part of what makes bemotrizinol appealing is that it provides protection against both types of dangerous ultraviolet rays on its own. And not only does it provide that broad level of protection, Adamson said, but it could also be more “cosmetically elegant,” as he put it. It won’t leave those white streaks that mineral sunscreens do, which could encourage more people to actually put it on. 

The shift toward mineral sunscreen in the wake of the chemical sunscreen panic has brought one unfortunate side effect: that white film on the skin of beachgoers and baseball game fans across the country. If you have ever applied zinc-centric sunscreen, you probably know the look (and that heavy feeling of the cream on the skin).

Chemical sunscreens can be annoying for people with sensitive skin, but by and large, people seem to prefer those products because they look better when wearing them. BEMT could make it easier for manufacturers to produce sunscreens that provide that broad level of coverage while being aesthetically more pleasing.

BEMT also comes with fewer safety concerns

The other hope is that bemotrizinol products will ameliorate some of the safety concerns that have driven sunscreen skepticism since 2019, when the one-two punch of the FDA’s announcement that most ingredients had “insufficient” data to judge their safety, followed by a worrying study, damaged the reputation of chemical sunscreens for the better part of a decade.

The study, published in JAMA in May 2019, showed several popular chemical sunscreen ingredients appeared to penetrate a user’s body in volumes sufficient enough that they should trigger new safety studies. The authors noted that some of the ingredients had previously been found in human breast milk and other bodily fluids. The findings raised real concerns, thus the FDA’s policy shift — but those concerns also took on a life of their own in the health and wellness social media ecosystem, stoking doubts about sunscreen overall.

“That freaked everyone out. And everyone was like, ‘I don’t want to do chemical sunscreens. They’re terrible. They’re getting [in] your blood. They’re endocrine disruptors.’ All of that kind of fearmongering,” Adamson said. “This ingredient doesn’t seem to do that.”

He pointed me to preliminary evidence from clinical trials that indicates BEMT does not generally lead to the same kind of concentration in human plasma. The drug has already been in use in other countries for decades and has accrued a strong safety record. But the FDA’s policy of regulating sunscreen as an over-the-counter drug, rather than as a cosmetic, sets a higher standard for approval, which meant that it took more than 25 years for BEMT to finally cross the Atlantic from Europe to the US.

BEMT will be coming soon to stores near you

DSM Nutritionals will have exclusive rights for 18 months to sell their proprietary BEMT formulation Parsol Shield in the United States; after that, other companies will be able to sell sunscreens with it in them too. Going forward, consumers can check for bemotrizinol or BEMT on the ingredients list.

Whether or not you opt for BEMT, here is the thing to keep in mind about protection when you’re buying this or any sunscreen: SPF, or sun protection factor. Experts say that the ideal is between SPF 30 and SPF 50, which blocks 98 percent of the sun’s rays. Just remember that SPF above 50 adds minimal additional protection, and doesn’t mean you can spend longer in the sun without reapplying.

Advocates hope BEMT can revive people’s faith in sunscreen which, despite the recent controversies, remains a lifesaving product. Skin cancer is still the most commonly diagnosed cancer in the United States, with more than 200,000 new cases expected this year. “American consumers deserve access to the best available sun protection,” Alexa Friedman, senior scientist at EWG, said in a statement. “Today they’re finally getting closer to it.”