Reddit deep dives can involve anything from TV show fan theories to DIY advice.

A new AI-assisted analysis now suggests that the 'front page of the internet' could help researchers spot potential side effects of GLP-1drugs used to manage weight and diabetes, such as Ozempic, Wegovy, Mounjaro, and Zepbound.

A team from the University of Pennsylvania analyzed more than 410,000 Reddit posts across a six-year span, looking for mentions of two active ingredients in widely used GLP-1-based drugs: "semaglutide" or "tirzepatide", or their brand names.

" Clinical trials are the gold standard, but by design, they are slow," says computer and information scientist Sharath Chandra Guntuku.

"This is not a replacement for trials, but it can move much faster, and that speed matters when a drug goes from niche to mainstream almost overnight."

When it came to potential side effects that doctors may not know about, two groups of complaints stood out: reproductive health issues (including irregular menstrual cycles) and temperature-related problems (such as chills and hot flashes).

The way the body's metabolism burns energy is known to impact temperature balance, so there is a real mechanism that potentially links the side effect to the drug in a causative way. But there is less research on how these drugs affect the menstrual cycle.

It's worth noting that these previously "unrecognized potential effects" were far from the most commonly reported on Reddit, and the researchers emphasize that they don't see their new analysis as a replacement for trials or clinical assessments.

However, the findings suggest that online boards and peer-to-peer conversations may be places where people feel able to discuss certain symptoms, some of which they might not mention to a doctor.

"Some of the side effects we found, like nausea, are well known, and that shows that the method is picking up a real signal," says Guntuku.

"The underreported symptoms are leads that came from patients themselves, unprompted, and clinicians could potentially pay attention to them."

This kind of large-scale data capture and interrogation is made possible by the latest AI models. Here, the researchers used GPT models from OpenAI to crunch through Reddit posts and find patterns.

That's no easy task, given the volume of text and the variety of ways people might talk about GLP-1 drugs and their side effects. These scans can be completed rapidly, identifying potential sources for investigation in future clinical trials.

With little else known about the Reddit posters involved, the research can't be definitive about whether drugs like Ozempic or Mounjaro are actually causing these symptoms.

However, these self-reported side effects are worth further investigation.

"These drugs are thought to work by engaging part of the brain called the hypothalamus, which helps regulate a wide variety of hormones," says psychologist Jena Shaw Tronieri.

"That doesn't mean the medications are necessarily causing these symptoms, but it could suggest that reports of menstrual changes and body temperature fluctuations are worth studying more systematically."

GLP-1 drugs are named after glucagon-like peptide-1, a natural hormone that the medications mimic. Specifically, the drugs limit appetite, slow down digestion, and trigger the release of insulin from the pancreas in response to high blood sugar levels.

While these treatments are associated with significant benefits in weight loss and diabetes management, research is ongoing into other potential consequences of GLP-1 use.

Those consequences potentially include protection against Alzheimer's, better cardiovascular health, and a greater risk of acute or chronic pancreatitis – so, quite the mix. We also know that regaining most of the weight after GLP-1 treatments is common.

The kind of online analysis done here may highlight problems that might otherwise be missed.

"The whole point of this kind of approach is that it can move quickly, and that's exactly when it's most valuable," says Guntuku.

Reddit tends to skew towards younger, male, US adults – but that doesn't mean it can't be useful for flagging problems that researchers need to know about.

Related: Ozempic-Like Drugs Can Help You Lose Weight, But There's a Catch

"Clinical trials generally identify the most dangerous side effects of drugs, but they can fail to find what symptoms patients are most concerned about," says computer and information scientist Lyle Ungar.

"Even though social media is not necessarily representative, a large collection of posts may reflect additional concerns."

The research has been published in Nature Health.

ScienceAlert stories are written, fact-checked, and edited by humans, never generated by AI. Don't miss a story, subscribe here.

The hottest giant planets in the galaxy should, in theory, have the fastest winds.

The hotter a planet is, the stronger its atmospheric currents should be – and a category of exoplanets known as hot Jupiters contains the hottest worlds we've ever found.

They orbit so insanely close to their host stars that some of them are literally evaporating from the heat…

Yet a new analysis of seven hot Jupiters reveals wind speeds that are practically sluggish, compared to what astronomers expected.

The best explanation for this surprise, according to a team led by astronomer Julia Seidel of Côte d'Azur Observatory in France, is that something is holding the winds back.

And the mechanism that could best explain that powerful braking effect is a magnetic field.

If the team's findings are validated, these laggardly winds could be the best evidence we've seen yet of magnetic activity on a world outside the Solar System.

"This breakthrough opens a completely new window on exoplanet research," Seidel says.

"It's the first time we can compare the magnetic environments of other worlds – a key step toward ultimately understanding which planets can stay alive, keep their water, and perhaps even, one day, host life as we know it."

Hot Jupiters are already some of the most fascinating exoplanets in the Milky Way. These worlds are in such proximity to their stars that, in the most extreme cases, their orbits are less than a day.

This means that two things are usually true for hot Jupiters. The first is that they are tidally locked, with one side permanently in daylight facing the star, and the other in permanent darkness facing away.

This produces a temperature contrast that should create some absolutely demented weather.

The second is that these worlds are usually heated to equilibrium temperatures of several thousand degrees, helping drive even stronger atmospheric circulation.

Now, we can't directly measure magnetic fields on exoplanets, but previous studies of individual hot Jupiters have shown that, by tracing vaporized iron in the atmosphere, wind speeds can be established.

Because we know that magnetic fields can act as a brake on electrically charged gases, the researchers thought they might be able to use hot Jupiter wind speeds as a proxy for magnetic field activity.

They used the MAROON-X instrument on the Gemini North telescope and the ESPRESSO instrument on ESO's Very Large Telescope to measure wind speeds across seven hot Jupiters.

Now, wind speeds on these worlds are still far beyond anything we might see in the Solar System. The researchers recorded howling gales at speeds between 2 and 7 kilometers (1.2 to 4.3 miles) per second. Jupiter's wind speeds – the fastest in the Solar System – only get as high as about 0.4 kilometers per second.

However, what makes the hot Jupiters interesting is the clear relationship between wind speed and temperature.

The researchers found that the hotter the exoplanet, the slower its winds.

There are some other explanations for slower-than-expected winds on hot Jupiters; but, the researchers argue, the other possibilities would still show the opposite trend, with wind speed increasing with temperature.

"This is totally counterintuitive because, all things being equal, hot planets have more energy to accelerate the winds!" says astronomer Vivien Parmentier of Côte d'Azur Observatory. "Something must happen that slows down the wind speeds for hotter objects."

This something, the researchers argue, is most likely to be magnetic fields… and, based on the trend in their observations, they were even able to infer the strength of the field producing the effect.

The hot Jupiters, they found, should have magnetic fields of only a few gauss, roughly comparable to Jupiter's.

Because it's a proxy measurement, further observations may be required to confirm the team's findings.

Related: Ludicrous Lemon-Shaped World Is Like Nothing We've Ever Seen

However, it's still a lovely result – one that shows just how far we've come in understanding alien worlds, moving away from the characteristics of individual planets to statistical-level analyses that start to reveal patterns.

"Here on Earth, we know the beauty of the northern and southern lights, where particles from the Sun hit our magnetic field and are guided toward the poles, colliding with gases in the atmosphere to produce colorful displays of green, pink, and purple," says astronomer Bibiana Prinoth, formerly of Lund University, Sweden, now at the ESO.

"I like to imagine that some of these worlds have a sky filled not only with stars, but with vast curtains of colorful light dancing across a planet that's half in perpetual day and half in endless night."

The research has been published in Nature Astronomy.

ScienceAlert stories are written, fact-checked, and edited by humans, never generated by AI. Don't miss a story, subscribe here.

One of the hardest things to do in physics is to generate true, provably unpredictable randomness.

That's because it's impossible to determine randomness based on the output alone.

Dice may have nicks and flaws that influence how they roll.

Computer random-number generators are usually driven by algorithms.

Even coin flips are governed by physical forces that, in theory, could be predicted.

The difficulty lies not in generating numbers that appear random, but in showing that no one could have possibly predicted the outcome – that the system isn't secretly affected by subtle hidden rules or biases.

Now, a team of physicists at ETH Zurich in Switzerland has overcome that challenge by leveraging one of the strangest phenomena in quantum mechanics: entanglement.

"The resulting sequence of zeros and ones is now really perfectly random, and we can even certify that," says physicist Renato Renner of ETH Zurich.

Randomness is crucial to modern security.

It's the core feature that makes passwords, authentication codes, and encryption keys harder to guess.

It's the reason password generators will produce a string of meaninglessly jumbled characters rather than something like YourFirstPet123.

But the stakes extend far beyond a Flickr password to international security.

Recent examples of security weaknesses include the 2024 PuTTY vulnerability, in which one of the world's most widely used SSH clients had a flaw in its random-number generation for cryptographic signatures.

And don't forget the 2025 AMD Zen 5 RDSEED bug, in which a hardware random-number instruction would generate predictable values while falsely reporting success.

If a code is not perfectly random, it's easier for attackers to guess.

"Any conventional electronic device, like a phone or a computer, is completely deterministic," Renner told Adam Kovac at Scientific American, "so it's actually very difficult for a computer or any other electronic device to generate a random value."

To try to find a solution to this problem, the researchers turned to a quantum experiment known as the Bell test.

They created a pair of entangled quantum bits, or qubits, separated by 30 meters (98 feet) and cooled to temperatures close to absolute zero.

Entangled particles are those that, when measured, show similarities that cannot be explained by classical physics alone.

Measurements performed on the qubits produced correlations so strong that they could not be explained by ordinary hidden rules or pre-programmed behavior.

This achievement required major technical improvements to both the stability and speed of the experiment, allowing the team to perform more than a billion Bell-test trials over roughly nine hours.

Previous quantum random-number generators could produce highly random outputs, but they still relied on trusted hardware and perfectly random starting conditions.

The ETH Zurich team instead demonstrated something called randomness amplification, deliberately starting with imperfect randomness – taking randomness that may contain subtle flaws or biases and transforming it into randomness that can be certified as perfectly unpredictable.

"Crucially," they write in their paper, "randomness amplification has been proven to be impossible by purely classical means."

The result is a system capable of generating certifiably perfect randomness, even when starting with flawed or imperfect randomness.

Related: Crystals Have Been Used to Generate Truly Random Numbers For The Very First Time

And it's also device independent, which means the randomness does not depend on trusting the hardware itself, but on the quantum behavior observed in the experiment.

In the long term, the researchers say that their system could perform the same function atomic clocks perform for timekeeping – a physically certified source of randomness against which others can be measured and set.

"The technical improvements allowed us, for the first time, to create random numbers that will remain perfectly random for all eternity – no matter what analytical methods are used to assess their randomness," Renner says.

The research has been published in Nature.

ScienceAlert stories are written, fact-checked, and edited by humans, never generated by AI. Don't miss a story, subscribe here.

For millions of people, statins are a daily shield against heart disease.

But around 10 percent of those who take statins to lower cholesterol experience a mysterious, painful side effect that causes many to discontinue these potentially life-saving medicines.

Scientists have recently found one possible reason why.

Research from Columbia University and the University of Rochester in the US revealed a potential culprit: a tiny calcium gate inside muscle cells that statins may force open.

The resulting calcium leak can damage muscle tissue, offering a new explanation for at least some cases of statin-associated muscle symptoms (SAMS).

"I've had patients who've been prescribed statins, and they refused to take them because of the side effects," said lead author Andrew Marks, a cardiologist at the Columbia University Vagelos College of Physicians and Surgeons.

Statins work by blocking an enzyme that's required for the biosynthesis of cholesterol in the liver.

As a result, levels of 'bad' LDL cholesterol are reduced in the blood, helping to prevent one of America's top killers: cardiovascular diseases like atherosclerosis, the buildup of fatty deposits in blood vessels.

But statins also affect "off-target" molecules, including a protein called ryanodine receptor 1 (RyR1). RyR1 is a mushroom-shaped channel, or gate, located on the sarcoplasmic reticulum, a web-like structure that surrounds muscle fibers.

RyR1 acts like a bouncer at a club, opening or closing the door to let calcium ions flow into the muscles. This calcium flow is an essential process that mediates muscle contractions.

Using mice as models, the researchers observed the precise way statins bind to RyR1, using an imaging technique called cryo-electron microscopy (cryo-EM).

Cryo-EM involves flash-freezing biological samples and then blasting them with electron beams. The deflection pattern of the electrons reveals tiny structures, allowing scientists to create highly detailed 3D images of things like proteins and view their constituent molecules.

Yet cholesterol-lowering drugs like simvastatin may keep these gates open, allowing calcium ions to leak into muscle cells, which can either directly damage muscles or trigger enzymes that degrade them.

As a result, statin users may experience persistent pain, weakness, tenderness, and cramps.

The issue is exacerbated in individuals with RyR1 mutations, who may also experience episodes of malignant hyperthermia (a severe overheating triggered by medication) or weakness in the diaphragm that leads to reduced lung function and respiratory disorders.

In rare but potentially life-threatening cases, the side effects of statins can induce rhabdomyolysis, a serious syndrome in which muscle tissues rupture and leak into the bloodstream, culminating in kidney failure.

The equally gruesome autoimmune-mediated necrotizing myositis may also rarely occur, a condition in which the immune system turns against its own tissues and kills muscle tissue.

The leaky calcium gate explanation may not apply to all cases of SAMS, but now that we understand this mechanism, it could help identify people at risk of statin intolerance.

Around 40 million adults take statins in the US alone, and approximately 10 percent of treated individuals experience SAMS.

"It's the most common reason patients quit statins, and it's a very real problem that needs a solution," said Marks.

Related: US Cardiologists Have Published New Guidelines For Managing Cholesterol

The researchers highlight two promising options. The first is to redesign statins so they don't bind to RyR1 but still inhibit cholesterol production in the liver.

Alternatively, when the researchers treated statin-intolerant mice with Rycal, an experimental class of drug used to treat patients with rare muscle diseases, they were able to close the leaky RyR1 calcium gates and prevent simvastatin-induced muscle weakness.

"It is unlikely that this explanation applies to everyone who experiences muscular side effects with statins," Marks explained.

"But even if it explains a small subset, that's a lot of people we could help if we can resolve the issue."

This research was published in the Journal of Clinical Investigation.

An earlier version of this article was published in February 2026.

ScienceAlert stories are written, fact-checked, and edited by humans, never generated by AI. Don't miss a story, subscribe here.

A life that leads to dementia can take many paths, but there are some common risk factors that make a diagnosis more likely.

An extensive new study shows, however, that among older adults, the effect of those risk factors differs.

Some risk factors seem to hit women's cognition harder than men's, and accruing multiple risk factors over a lifetime seems to worsen women's brain function more so than men's.

In other words, certain risk factors may not mean the same thing in a woman's brain as it does in a man's, which has implications for dementia research and treatments.

It's well known that dementia is more common in women than in men, and longer life expectancy does not fully explain the gap.

There's something else going on that these findings could also help us understand.

"Our study suggests that women may be at greater risk of dementia because they experience a greater number of risk factors, and because these risk factors reduce cognition to a greater degree than [in] men," the researchers write.

Study co-authors Megan Fitzhugh and Judy Pa, two neuroscientists from the University of California, San Diego, say their results provide further evidence that dementia risk needs to be assessed and managed in a personalized way.

"Looking beyond which risk factors are most common, we found that some have a disproportionately larger impact on women's cognition," says Fitzhugh.

"This suggests that prevention efforts may be more effective if they are tailored not just to risk factor prevalence, but to how strongly each factor affects cognition in women versus men."

The researchers analyzed health data from 17,182 individuals aged 40 or older, examining 13 risk factors linked to dementia.

Depression, physical inactivity, and sleep problems were more common in women, compared to men.

Higher rates of hearing loss, diabetes, and heavy alcohol use, on the other hand, were more often reported by men.

Some risk factors were linked to greater reductions in cognitive scores in women – an indication that they affect women's brains more negatively than men's.

These factors included high blood pressure, hearing loss, and diabetes.

Higher BMI was also associated with poorer cognitive performance in women in their 50s and 60s, but not at older ages.

Amid that picture of cognitive decline, there were some positive trends suggesting that certain factors might help preserve cognitive function in women compared to men.

"Two risk factors, years of education and total cholesterol, showed positive associations with cognition, such that higher levels were correlated with greater cognition," write the researchers in their published paper.

Given their links to cognitive performance, it's possible these factors may be especially important to investigate in women's dementia risk.

However, this observational study can't prove cause and effect. A longer-term analysis could provide stronger evidence that these risk factors were contributing to the cognitive test scores.

While 'women' and 'men' are recognized as gender categories, this study uses these terms to refer to individuals' self-reported biological sex.

"It is important to distinguish between sex differences in the prevalence of risk factors and their impact on cognition, because prevalence and impact may not correspond," writes the team.

"Targeting only the most prevalent risk factors within each sex may overlook certain risk factors that more markedly influence cognitive decline."

The new findings fit with previous research suggesting that dementia risk factors may affect men and women in different ways, though these studies have tended to look at only one risk factor at a time.

Alzheimer's disease now affects an estimated one in nine US adults aged 65 and older, and two-thirds of those affected are women.

While the figures for dementia may be bleak, there are real and practical ways to reduce risk, while work on treatments continues.

The researchers emphasize that all these risk factors are potentially modifiable. That means they're real targets that people and their doctors can try to address, whether it's drinking less, moving more, or seeking help for depression.

Next steps here could include research to determine why women's cognition may be more vulnerable to certain factors.

Hormonal changes around menopause may be involved, but the mechanisms remain unclear.

Related: One Vital Bodily Function Could Link Many Dementia Risk Factors

"These differences highlight the importance of considering sex as a key variable in dementia research," says Pa.

The research has been published in Biology of Sex Differences.

ScienceAlert stories are written, fact-checked, and edited by humans, never generated by AI. Don't miss a story, subscribe here.

Night shift work is not for the weak.

Staying awake from dusk through to dawn, as many nurses, doctors, and emergency responders do, seems to take a toll on the body and mind.

But does it have an impact on the brain?

Neuroscientists in Singapore have now found evidence that shift work is tied to brain volume losses in key parts of the brain.

If shift work is stopped, however, those reductions are partially recovered within two and a half years, on average.

What those losses and gains actually mean for human health or behavior is unclear.

A secondary analysis revealed a negative correlation between volume loss and cognitive performance: Increasing volume loss was associated with poorer performance on some, but not all, cognitive tests.

But the effect size is "very small", the authors warn, so the results "should be interpreted cautiously."

That said, there's an important clue in the details. The brain regions that showed significant volume losses also help govern our sleep cycles.

What's more, they are involved in many of the symptoms of shift work, like poorer emotional regulation and memory performance.

The study is the largest of its kind and finds a change in brain volume where most previous analyses of shift work have not.

It analyzed MRI and long-term health data from 14,198 middle- to older-age adults with no medical issues who took part in the UK BioBank.

Among 2,122 shift workers, the researchers noticed a symmetrical pattern of modest volume loss in the right thalamus, which is part of the brain's information relay 'hub' and is closely involved in memory retrieval.

They also noticed modest volume loss in the left amygdala, which regulates emotional responses.

This was after accounting for age, sex, chronotype, and skull volume, among other factors, in their analysis.

"The selective thalamic and amygdalar volume loss observed in healthy shift workers may represent an early, subclinical marker of neural vulnerability linked to chronic circadian disruption," the team concludes, led by neuroscientist Thomas Welton.

"These regions are central to sleep-wake regulation, emotion, and attention, functions that are commonly affected in shift work-related fatigue and mood disturbance."

Challenges with regulating emotions are often tied to poor sleep, and shift workers are known to face higher risks of both sleep disorders and mental health problems.

Researchers have long speculated that a disrupted circadian rhythm is to blame.

Other factors that may contribute include a lack of sunlight or changes to eating times.

But just because some parts of the brain are shrinking does not mean they are necessarily dying. The brain is a flexible organ that can rewire itself to meet the challenges of the time.

Perhaps that is what it is doing for shift workers; maybe their brains are somehow compensating in a way that allows them to work through the night.

"It is possible," the authors note, "that individuals who fail to acquire these brain changes are unable to tolerate shift work and are therefore biased toward non-shift working roles."

The study took place only among older adults, which means it's not clear how the brains of younger workers may cope with the demands of shift work.

Further studies are needed to fully understand how different people respond and are affected.

Today, full-time shift workers make up about 10 to 17 percent of the US population, but by some estimates, roughly a quarter of the adult workforce currently labors during non-traditional hours.

Related: Sleepless Nights Could Drive Half a Million Cases of Dementia in The US Each Year

If this work repeatedly disrupts the body's natural circadian rhythm, it could have a long-term and measurable impact on the brain, but we won't know until those changes are studied further.

"In the "era of longevity", it is critical to understand the relationship between shift work and structure of the middle-older aged brain," Welton and colleagues write.

"The apparent reversibility of these [observed] structural effects within two years of ceasing shift work highlights a potential therapeutic window for prevention and recovery," they add.

The study is published in NeuroImage.

ScienceAlert stories are written, fact-checked, and edited by humans, never generated by AI. Don't miss a story, subscribe here.