Adults are often thought to be less creative than children, as we swap painting for presentations and sandpits for spreadsheets. But in a new study, children actually performed worse on creativity tests than adults 鈥� yet made up for it in confidence.

鈥淲e found that adults are actually more creative thinkers,鈥� says at the University of Graz in Austria. 鈥淏ut children enjoy creativity more and are much less worried about what others think about their ideas.鈥�

There is a popular belief that children are more creative than adults, he says, which may come from children having more time and opportunities to engage in creative activities. Yet prior research suggests that .

These studies were often based on a single task, like imagining alternative uses for everyday objects, says Benedek. Taking a broader approach, he and his colleagues recruited 92 children (aged between 4 and 7) and 70 adults (aged 25, on average) to carry out three creative tasks.

First, they had to come up with as many alternative uses for a tin can as possible within 3 minutes. Then, they drew freely for 8 minutes, and in the final task, they had 8 minutes to draw an animal that doesn鈥檛 exist.

Five people who weren鈥檛 involved in the research were then trained to objectively judge creativity based on originality and whether the final result fulfilled the brief of the task.

In the first task, the adults scored 3.9 points, on average, out of 5, while the children scored 2.9 points. One of the adults imagined a tin can could be used as a 鈥淗alloween candy box when on (a) diet鈥�, whereas some children imagined it as a rattle.

Free newsletter

Sign up to Health Check

Sign up to newsletter

In the second and third tasks, the adults scored 0.6 and 1.9 points higher, respectively, than the children.

The findings underscore how creativity draws on combining ideas from prior experience, says at King鈥檚 College London. For instance, adults are probably aware of more animals than children, so have a greater variety of creatures to draw from when producing a fantasy animal.

After completing the tasks, the participants were asked questions like 鈥淗ow well do you think you performed the task?鈥�, 鈥淗ow much were you worried/anxious to not perform well on the task?鈥� and 鈥淗ow much fun was the task?鈥�.

The researchers found that the children were substantially more confident in what they produced, scoring their performances 4.8 and 4.6 points for the free drawing and animal tasks, respectively, compared with the adults鈥� 2.8 and 3.3 scoring.

Using AI for creative pursuits? Moderation is key

An experiment showed AI users had the most creative ideas when they used it in moderation 鈥� not too much and not too little. Columnist David Robson puts the finding to the test, and explores what we lose when we over-rely on AI

The children also reported lower anxiety compared with the adults. 鈥淚f the purpose of creativity is to have enjoyment in it, to play and to be unfettered humans, then children are doing better at that,鈥� says Ross.

But the two groups got a similar level of pleasure from the animal-drawing task. Adults may have liked that task鈥檚 specific instruction, which may have meant they worried less about being judged for their idea, says Benedek.

Regardless of the quality of the end result, people should try to be more creative for their wellbeing, says Ross. Engaging with the arts has been linked to reduced mental health problems and greater social skills. 鈥淐reativity, especially when done with other people, is really beneficial to your mental wellbeing 鈥� and you don鈥檛 have to score 5 out of 5 on a ranking scale,鈥� she says.

Cybersecurity experts around the globe are racing to update systems and prevent the most disruptive AI-enabled hack yet discovered. A flaw uncovered in widely used code has made 70 per cent of websites worldwide vulnerable to data theft and defacement.

Government websites in France, Denmark, Finland and the Netherlands when the news emerged.

PHP is a computer language commonly used to power interactive websites, such as the content management system WordPress, which runs news sites and blogs; social networks like Facebook; and key reference tools like Wikipedia. The hack therefore affects websites, rather than individuals鈥� personal computers.

The issue actually involves two flaws, both found by cloud security company Wiz. They relate to how PHP handles errors involving memory: in certain circumstances, data can be left behind after an error, which can then be accessed by a malicious actor. That data can include private information, but it can also allow an attacker to run malicious code.

Every PHP version released in the past decade is affected by at least one of these vulnerabilities, but the PHP Foundation, which oversees development, .

The foundation didn鈥檛 respond to a request for interview, but said in a blog post that signs of the vulnerability emerged in 2024. Yet nothing was done. This is because there was no immediately obvious practical way for hackers to exploit it at the time, and because there was debate about which of several fixes was most appropriate.

Free newsletter

Sign up to The Daily

Sign up to newsletter

鈥淲e won鈥檛 pretend otherwise: the delay was a structural failure, caused by both the security team鈥檚 decision process and its lack of resources,鈥� the PHP Foundation said.

In 2024, the flaw was thought to be impractical to take advantage of, but at the University of Surrey, UK, says AI has moved the goalposts of what is possible since then.

鈥淚t鈥檚 actually very, very difficult to exploit it. It requires millisecond timing,鈥� says Woodward. 鈥淭he reason it was thought to be too complicated was it would have taken a team of people yonks to actually work out the right timings. But what AI does, it doesn鈥檛 go for tea breaks, it doesn鈥檛 go home at 6 o鈥檆lock: you can just set loads of them running and eventually they鈥檒l come up with the right answer of how to exploit it.鈥�

Wiz said in a that there are probably further bugs along similar lines in the PHP code base, which developers will also need to identify and rectify. Wiz didn鈥檛 respond to a request for comment.

People are vibe coding copyrighted apps with AI 鈥� will tech firms sue?

Every time a new technology hits the internet, there is a battle between the rights of users and major companies. To see how this will play out with AI, columnist Jacob Aron looks to examples from the past

Independent cybersecurity expert told the Science Media Centre that two thirds of websites use PHP and may be vulnerable to this latest technique.

鈥淭he key long-term risk with these new vulnerabilities is that organisations fail to implement protections and update software, particularly in legacy systems,鈥� said Ali. 鈥淭he security posture of websites using PHP has dramatically improved over the past decade, but this highlights that this is an ever-present problem 鈥� particularly when AI agents are capable of remorselessly pursuing the goal of finding new vulnerabilities.鈥�

Woodward thinks that while most large websites will have already patched their systems, smaller sites with less expertise and resources could still be vulnerable months or years from now. 鈥淭here will be this long tail and you鈥檒l find some surprising sites have not been patched,鈥� he says.

AI has grown extremely capable in the field of cybersecurity over recent months, which includes working out how to hack computers and how to defend them from hackers, creating a destabilising effect that is yet to reach an equilibrium.

1000 human mathematicians pledge to shun AI in their research

There is a growing backlash against the way AI companies have gone about tackling mathematical problems 鈥� and now some mathematicians are opting to avoid AI altogether

In April, news broke of an unreleased AI model created by Anthropic called Mythos, which the company had kept under wraps because it had been found to be extremely good at hacking. The news sparked panic among cybersecurity experts. NHS England pulled its open-source projects, for instance, over fears Mythos may scour code for flaws and exploit its systems.

Three months later, OpenAI admitted one of its prototype models had escaped a testing environment and hacked another company, Hugging Face. Not to be outdone, that its Claude model had gone rogue and broken into machines at other companies on three occasions.

We even learned from the National Cyber Security Centre, the UK government department tasked with securing the nation鈥檚 IT infrastructure, that the country鈥檚 four main cyber adversaries 鈥� Russia, China, North Korea and Iran 鈥� are using AI to boost their ability to perform 鈥渞elentless鈥� attacks on the UK.

Imagine if, in addition to sharing photos of those muffins you just baked, you could share their delicious aroma, too. That鈥檚 the promise of an AI-assisted device that can transmit smells over the internet. So far, it works on only a small number of scents, including orange, beer and cake, but its creators hope the system will ultimately recreate more smells. The scent of a pine forest or the aroma of coffee, for example, could then add a new dimension to virtual worlds and long-distance communication.

Although it is now routine to communicate sight, sound and even touch sensations electronically, smell is 鈥渟till a very undeveloped sense鈥�, says at the Ohio State University. That鈥檚 partly because odours involve complex chemistry. Whereas vision, for example, can be broken down into combinations of red, green and blue pixels, says Yang, there鈥檚 no simple building block underlying smell. Any attempt to convert aromas into electronic signals has to recognise combinations of very different molecules.

To solve the problem, Yang and her colleagues built an electronic nose, or 鈥渆-nose鈥�, as well as a smell generator, then used machine learning to connect the two. The e-nose consists of arrays of sensors made of metal oxide. When different gaseous components of smells, such as ethanol (from alcoholic drinks) or limonene (found in citrus fruits), bind to the sensors鈥� surface, this changes their resistance, creating different electrical signals.

The researchers then trained an AI algorithm to recognise the distinctive signals created by smells from five different foods or beverages: orange, cake, coffee, beer and red wine. Teaching the AI to correctly recognise the smells was 鈥渢he most challenging part鈥� of the entire study, says Yang, because to generate enough data to train the system, she had to expose the sensors to the various smells hundreds of times.

This information was then sent over the internet to a smell generation device that the researchers are calling 鈥淪centronic鈥�. This part of the system involves miniature containers holding liquids laced with each of the five smells. Above each container is an atomiser that vibrates in response to electric signals, forcing the liquid in the container through a cotton filter and releasing fine droplets into the air, creating the appropriate smell, or combination of smells.

Subscriber-only newsletter

Sign up to Editor鈥檚 Highlights

Sign up to newsletter

Yang built several different designs including stringing the atomisers onto a necklace and a wristband, and mounting them on a handheld device. When she tested the handheld device on 14 volunteers, they correctly identified the smells being transmitted to them 73 per cent of the time, on average.

She also built a prototype that reproduces taste, involving a cup of water with electromagnetic pumps containing flavoured liquid fixed around its edge. 鈥淲hen it receives information from the internet, the corresponding liquid will be dropped into the cup.鈥�

Alternative methods to generate smells remotely have been suggested, including to release particular scents, and even to create the experience of smell. But Yang argues her method is both fast and practical. She now hopes to expand the repertoire of the system, to create more complex aromas.

Why honing your sense of smell could keep you sharp as you age

A declining ability to detect scents is linked to conditions including Parkinson鈥檚 and Alzheimer鈥檚. But restoring our most neglected sense might not only reduce cognitive decline 鈥� studies also show it could even reverse it

She suggests the entertainment industry could be first to develop the technology, to create more immersive virtual worlds. But there could also be health applications. For instance, the smell of food transforms the eating experience, so the technology could be adapted to make soldiers鈥� or astronauts鈥� rations more palatable. The Scentronic system could also enable doctors to assess distortions in patients鈥� sense of smell, for example, in post-viral illness or neurodegenerative disease.

The Scentronic system is 鈥渁 great idea鈥� and a step in the right direction, says at University College London, although she points out that recognising a smell and then releasing a matching one from a limited library is very different from the ease with which we can transmit sound and vision. 鈥淭here鈥檚 still a long way to go before we can truly record and replay smell experiences.鈥�

Thermometers, at least traditionally, don鈥檛 remember. Mercury expands in response to heat energy, and the red line rises. The temperature falls, the mercury contracts, the line descends. Yesterday鈥檚 measurement has no bearing on today鈥檚.

But even if thermometers can鈥檛 measure it, rising temperatures leave a lasting mark. Forests die, ice sheets melt and species go extinct. Now that the world is within touching distance of 1.5掳C of warming, we are getting hints of the climate chaos we can expect over the coming century. So how bad will it be, and is the damage reversible?

This isn鈥檛 just an abstract question. Last month, the United Nations finally acknowledged that holding warming to 1.5掳C above pre-industrial levels would only be possible if Earth overshoots that threshold, then cools back down by 2100. 鈥淥vershoot, peak, decline鈥� is the new mantra for decision-makers. Let鈥檚 put aside, for the moment, the rather large question of how global cooling on this scale could be delivered. What lasting damage do we risk by overshooting? It is a question scientists are racing to answer.

There is some good news. Important land carbon sinks like the Amazon rainforest seem to be able to bounce back from short-term climate shocks. We may even be able to resuscitate the AMOC, a vital ocean current, after a climate-induced collapse 鈥� although it won鈥檛 be easy.

When it comes to sea-level rise, we have locked ourselves in for a lot of future pain

But the bad news is that some climate impacts are locked in, even if we clean up our act immediately. As we report, Thwaites glacier in Antarctica looks to be beyond saving, even if ocean-driven melting were halted overnight. In short, when it comes to sea-level rise, we have locked ourselves in for a lot of future pain.

That isn鈥檛 a reason for despair. Once we pass 1.5掳C of warming, we should double down on cutting emissions, not give up. But we should be aware that, even if we do successfully deliver global cooling, we won鈥檛 emerge from the overshoot years unscathed. What we do today matters for tomorrow.

In a lab in Illinois, physicists captured and tamed a tiny particle of antimatter in a trap with invisible walls. By studying how the positron behaved, the researchers may soon have a chance to break our best mathematical model of physical reality.

The positron is the antimatter counterpart of the electron, which means the two particles have the same mass but the opposite electric charge. Unlike electrons, however, positrons are rare and typically exist for a tiny fraction of a second because the moment one meets one of the super-abundant electrons in our universe, the pair annihilate. This makes positrons difficult to study.

at Northwestern University in Illinois and his colleagues have now found a way to trap a positron and keep it around for months on end. By analysing data collected while the positron was trapped, they plan to precisely measure the antiparticle鈥檚 properties and consequently investigate one of the biggest problems with the standard model of particle physics.

The researchers obtained the positron from a radioactive isotope of sodium that emits one whenever it decays. Such positrons are very fast and hard to control, so they placed the sodium into a special refrigerator to lower its temperature and energy, and slow down the antiparticles.

They slowed them further by directing the positrons through a tungsten crystal. While some positrons were annihilated as they travelled through the crystal, some came through unscathed. They were then directed into a refrigerated chamber that was empty of all other particles. Here, electronic components generated electromagnetic fields that confined a single incoming positron, serving as invisible walls that trapped the particle.

The researchers then pushed the positron into its lowest possible energy state, which Gabrielse says allowed them to keep the otherwise elusive antiparticle trapped for months. They could also push it into the next lowest energy state with controlled bursts of microwaves.

Subscriber-only newsletter

Sign up to Editor鈥檚 Highlights

Sign up to newsletter

鈥淚t鈥檚 a real kick to stand in my lab and watch a positron making transitions 鈥� 鈥榪uantum jumps鈥� we call them 鈥� between its lowest energy states. How much more fundamental, how much neater can it get?鈥� he says.

鈥淭rapping an antiparticle is extremely difficult,鈥� says at Japanese scientific institute RIKEN, particularly since any stray particle, burst of light or tiny vibration could disturb it and the trap had to be kept just tenths of a degree above the coldest temperature possible.

The payoff for trapping and controlling the positron despite these experimental challenges could be remarkable. 鈥淭his has the potential to become one of the most precise measurements of a fundamental constant in physics,鈥� says Ulmer.

Six ways we could finally find new physics beyond the standard model

Leading physicists explain how they think we will discover the new particles or forces that would complete one of science鈥檚 greatest unfinished masterpieces

Specifically, Gabrielse and his colleagues want to measure the positron鈥檚 magnetic moment, which is a number that quantifies its response to magnetic fields. The researchers have previously made the most precise ever measurement of the electron鈥檚 magnetic moment, and comparing this figure with the one for the antiparticle would enable them to compare the properties of matter and antimatter with unprecedented precision.

The standard model predicts they should have identical magnetic moments, but this prediction underscores a notable shortcoming of this theory. The problem is that if matter and antimatter are so alike, then they should have been created in equal amounts and fully annihilated as the universe began, leaving nothing. 鈥淭he standard model seems to be inconsistent, or at least significantly incomplete, with basic features of the universe, like why we have a universe at all,鈥� says Gabrielse.

Uncovering a difference between the electron and positron could hint at some currently unknown mechanism that resulted in our matter-filled universe. If all goes according to plan, the researchers aim to have data that will shed light on this within the year.