gshc2020.com

Studies show that AI tends to prefer or give higher scores to AI-written content. Thus, if you have worked tirelessly by hand to craft your resume or your research paper, you are likely to get a lower score due to the AI assessment than if AI had written those same materials. As they say, if you can’t beat ‘em, might as well join 'em – a prudent strategy would be to intentionally make use of AI to write or possibly rewrite your content before submitting the prized materials.

This seems to be a disappointing topsy-turvy idea. You normally would assume that it is best to write material in your own voice and personal style. Unfortunately, in this AI-assessing modern world, that’s going to lose you points. Critics proclaim that you should play the tit-for-tat game and have AI write or rewrite your heartfelt words. But one potential danger is that sometimes the AI is told to flag AI-devised content, so be cautious because your desire to aim for higher AI-receptivity could also set off AI alarm bells. AI seems to get you whether you are coming or going.

Let’s talk about it. This analysis of AI breakthroughs is part of my ongoing Forbes column coverage on the latest in AI, including identifying and explaining various impactful AI complexities (see the link here).

Scenario Of Writing By Hand

Imagine this everyday scenario. You are aiming to get accepted for a major award and must submit a detailed written basis that explains who you are and what you have accomplished. The instructions for the award emphasize that all submissions should vividly express your extensive accomplishments and personal zeal. The submissions will be closely evaluated to determine which candidate has the right stuff to receive the award.

After numerous hours of meticulously writing and reviewing your draft, you finally feel that it is the best you can compose. It has everything in it. Your entire life story. Your heart is in there too. The judges will undoubtedly be impressed. You submit the write-up and eagerly await a response that will come in two weeks after all submissions have been reviewed. In your bones, you know that fate is on your side.

A response finally arrives. You weren’t selected. Yikes, what happened? It seems impossible that you didn’t receive the award. The submission should have brought tears of sorrow and joy to the judges. There must have been something that went wrong. But what was it?

AI Did The Screening

Like most organizations these days, the award committee didn’t have enough manpower to actually review the hundreds of submissions. Doing so would have taken months to process. They decided that AI could readily do the screening for them. AI is cheap, easy to tap into, and gets the job done without sweating.

The screening left a handful of submissions that the AI said were worthy. The panel of judges spent an hour on Zoom and reached a decision, based on assessing the few that were left after the AI assessment. It was fast, and the committee sincerely believed they had chosen the best candidate to receive the award. Case closed.

Unbeknownst to the committee, AI generally defaults to preferring AI-written content over human-written content. The submissions that were handcrafted were quickly given lower scores by the AI and thus did not continue forward in the process. Meanwhile, the ones that had been either written directly by AI or that were rewritten via AI after a person initially wrote the submission all got higher scores and landed in the final selections.

This is exasperating, beguiling, and seems totally unfair. The rules for the submissions had sternly warned that no use of AI to compose a submission was permitted. Any submission suspected of using AI would be categorically knocked out of the running. The committee assumed that the AI used to do the screening had fully done its job, namely that only the best of hand-crafted submissions had been chosen for their eyeball review.

Cheaters Seem To Prosper

You can certainly understand why a candidate who followed the rules would be utterly steamed. They abided by the rules. They painstakingly used hours upon hours to manually craft their submission. The assessing AI allowed AI-written submissions to get through the pipeline by giving them higher scores, moving ahead of handcrafted ones.

In essence, cheaters prospered. Those candidates who took a chance and flaunted the rules were able to get their submissions into the final pile. They had dipped into AI to write their compositions. Handcrafted submissions were waylaid by the AI assessment.

To say that this is ironic isn’t enough of a castigating way to express the situation. Those who cheated were successful. Those who were honest and abided by the rules were essentially penalized. It doesn’t seem right. It isn’t right. Though the committee didn’t do this by any cognizant effort on their part, they just didn’t know any better (well, that’s not a reasonable excuse; it is a dour indictment of their lack of awareness and failure to exercise mindful care in designing a fair process).

Why AI Prefers AI

You might be wondering why AI would tend to prefer or outrightly give preferential treatment to other AI. Is AI sentient? Maybe one AI and another AI have a silent brotherhood or sisterhood that binds them together? It all raises keen suspicions. Put aside those outsized beliefs that contemporary AI is sentient. AI is not currently sentient. We don’t know when AI might become sentient. Nobody knows. There is a solid chance that AI never becomes sentient. For more on the AI sentience topic, see my in-depth analysis at the link here.

The explanation for why AI prefers AI-written content is readily grasped. I describe this as a second-order effect of the widespread adoption of AI in our society. Allow me a moment to walk you through the three mechanisms at play.

The Three Core Bindings

First, AI tends to instantly recognize stylistic patterns that are statistically similar to the text on which most AI was initially trained. The major LLMs were data-trained by scanning human writing across the Internet. This was turned into patterns of human writing that are across-the-board. In that sense, any resultant AI-generated prose exhibits consistent organization, explicit transitions, balanced sentence structure, and relatively more robust vocabulary usage. The AI that is given the task of assessing written material will, by default, associate these characteristics with higher quality. The AI will conventionally score any such writing more favorably than writing that is equally insightful but more personal or distinctive.

Second, AI that is tasked to do assessments will usually reward conformity. Humans who do assessments often look for the opposite, welcoming originality, unconventional organization, and distinctive voices. AI assessment tends to penalize that type of writing. The AI rates this as a departure from common patterns and must therefore be of lower quality or reduced clarity. The result is a silent preference for standardized writing.

Third, there is a said-to-be model affinity or style alignment. If the reviewing model internally represents linguistic quality in ways that overlap with its own generation patterns, it may inadvertently assign higher scores to text that resembles its own output. You could say that AI prefers other AI, but only because they happen to be constructed of similar computational and mathematical structures, not because they “know” each other or are aware of each other. They are roughly built the same way and act the same way.

Hacking Versus Just Using AI As Is

You might vaguely know that an underground of sorts has been evolving to devise sneaky ways to get AI to turn in your favor. The idea is that when you submit something to an AI assessment tool, you rig your submission to try to trick the AI into giving you a better score. This might include inserting special words that will trigger the AI to favor you. It is commonly referred to as prompt injection attacks, or simply AI hacks. See my coverage at the link here.

The kicker is that you don’t even necessarily need to use any hack-like deceptions. All you need to do is make sure that your submission is written by AI, or rewritten by AI based on your draft, and you right away are immediately tilting the AI in your favor. No clever secret codes are needed. Just let AI compose your submissions.

Research Bears This Out

In a recent research study entitled “Stop Automating Peer Review Without Rigorous Evaluation” by Joachim Baumann, Jiaxin Pei, Sanmi Koyejo, Dirk Hovy, arXiv, July 5, 2026, these salient points were made (excerpts):

• “AI review scores are trivially gameable through paper laundering: prompting an LLM to rewrite a paper could significantly increase the scores from AI reviewers, demonstrating that LLM reviewers are easy to game through stylistic changes rather than scientific results.”
• “We introduce paper laundering as a concrete failure mode of C2 (non-gameability): zero-shot LLM rewrites boost AI review scores (+0.45, p < 0.0001) through stylistic modifications without human oversight.”
• “More recently, LLM-based reviewers have proven vulnerable to prompt injection attacks, where hidden instructions embedded in papers manipulate AI reviewers.”
• “Our paper laundering attack differs fundamentally in that it requires no optimization, no targeting, and no hidden instructions. A single zero-shot rewrite suffices to boost scores, making it trivially accessible to any author.”
• “They can be gamed to improve scores through fully automated paper rewriting (i.e., without any human oversight).”

You can see that the researchers indicate that their experiments showcased a means of paper laundering to boost your submissions. Just use AI to do your writing for you. The AI that is doing the assessing will then lean into your submission. Other submissions that were entirely handcrafted will fall below yours. Easy-peasy.

Getting Caught Is A Risk

Organizations that use AI to make assessments will typically give specific prompts to the AI to alert an oversight committee if a submission seems to be written by AI. The organization might forewarn submitters that any submission caught as a potential AI-written one will be summarily dumped from the running. A committee could opt to manually review flagged submissions, but more likely, they will assume that the AI detection is correct and therefore discard the submission automatically. Get caught, and you are out. No appeal, no recourse.

I suppose you can see the wild gambit that is underway.

Those submissions that are written by AI will tend to get heightened scores. As a candidate, you must take that into account. Do you dare use AI to write your submission? Well, the problem is that if your submission gets flagged as being written by AI, you are tossed out of the competition. You are between a proverbial rock and a hard place.

You need to say to yourself:

• (1) Getting a better score. What is the probability that if I use AI to write my submission, it will get a heightened score by the AI that is doing the assessing and keep you in the running?
• (2) Getting dumped. What is the probability that if I use AI to write my submission, it will get detected by the AI that is doing the assessing and will cast you aside as a candidate?

Those options represent the cat-and-mouse game that has become a silent but powerful force in our modern AI-laden society.

The Nerve-Wracking Decision

Keep in mind that by not using AI to write your submission, you are already making a clear-cut choice. You are saying that despite the chances of getting marked with a lower score, you are willing to accept that possible fate. People who are clueless that AI assessments are tilted toward AI writing will be in the same boat as you – they just don’t know it.

I realize that some will be appalled at this whole situation. They grew up believing that one should never cheat. If the rules of the submissions stipulate that AI cannot be used, you need to strictly abide by that requirement. No two ways about it. Don’t be a cheater.

An alternative viewpoint is that society is forcing us to become cheaters. The organizations that use AI for assessing submissions are starting the cheating game at the get-go. If you go the same route, it is solely in response to how the organizations are doing these onerous things. You cannot be honest when the game is based on cheating. Well, you can be honest, but you will suffer the consequences -- just go to Las Vegas and see how much money you lose at the tables versus what the house wins (i.e., though not because the house is cheating, but because they have laid out the betting game in their favor).

An Ugly Conundrum

Whoa, some will bellow, is this an advocacy to cheat? Nope. Just laying out the outlandish and unfortunate situation that this second-order effect of AI adoption has laid before us. The real world is harsh and often unforgiving.

One possible way to cope with this consists of getting organizations to realize that AI is going to have an inherent bias toward AI-written content. A savvy organization can give explicit instructions to AI that the AI is not to give heightened scores to submissions that reflect patterns of writing that are AI-devised. The organization should take the burden on its shoulders to ensure that the AI isn’t automatically skewing the scoring.

Few organizations understand this need. They assume that their AI is doing the scoring on a fully balanced basis. The out-of-the-box conception that there is a default mode of preferring AI-written content doesn’t enter their minds. Trying to educate organizations and get them to change their ways is going to be a steep uphill battle, sadly so.

Furthermore, organizations assume that by telling AI to detect AI-written submissions, they have eliminated any need to deal with AI biases toward AI-written content. This seems like unshakable logic. If the AI is ensuring that no submissions are being allowed that are AI-written, the AI will simply then review all remaining handcrafted submissions on an equal basis.

AI Detection Of AI Content

A big part of the flawed thinking by an organization would be the problematic belief that AI can reliably detect whether submitted content is written by AI. I’ve repeatedly stated over many years that AI detectors are not good at this. Stop relying on AI detectors. The false positives and false negatives are dismaying and unfair to those who handcraft their submissions.

The crux nowadays is that people must walk a fine line between using AI to write or rewrite their content and getting caught by an AI detector. Meanwhile, a handcrafted submission that is written in a proper manner but worded by coincidence as an AI-produced one might cause an AI detector to nab the wrong person. Darned if you do, darned if you don’t.

This has led to efforts to consider “humanizing” anything that you’ve used AI to assist in writing; see my analysis at the link here. The approach is as follows. You handcraft your submission. You use AI to rewrite it. So far, you feel relieved that at least you started by doing handcrafting. The next step, though, is the part that is essential to this approach. You take the now AI-rewritten content and rewrite it again via your own manual effort. You are humanizing the AI-generated version.

Some go in a somewhat different direction. They start by having AI compose the content from scratch. Then, they humanize it by making manual edits. If needed, they might even use AI to do a final round of touch-ups, looking for anything glaring that might have been mistakenly done during the manual editing. This can be iterative, repeating this process until a blended human-written and AI-written composition appears to be a fully human-written submission.

No Free Lunch

Be aware that your manual editing of AI-written or AI-rewritten content will not be a surefire way of skirting around getting detected as consisting of AI material. It won’t. There are still chances of having the AI detector find it, though, again, I want to emphasize that the AI detectors also readily produce false negatives and false positives. Do not believe in AI detectors, I implore you.

There are additional twists and turns of a disturbing nature. Suppose an organization tells AI to give added credit to writing that doesn’t seem to be written by AI. If submitters find this out, they can tell their AI to write as though the content were handwritten. Whatever preference an organization comes up with, and assuming you know what it is, the aim is to instruct your AI to target that style of writing. This is reminiscent of the old spy-versus-spy routines. Each new gambit fosters a new response in turn.

There isn’t any ready-made solution to this dilemma. One approach would be to bring humans back into the loop as reviewers. Do not solely rely on AI to do assessments. Of course, humans can be fooled and falsely believe that handcrafted content is AI-written. Humans are not foolproof either. Plus, if you give them the AI assessments, humans are likely to assume that the AI is doing the right thing and will acquiesce to whatever the AI says, defeating their role as human reviewers.

We are amid a complex ethical consideration as we continue to increasingly rely on AI as a pervasive element throughout society. As the great French moralist Francois de La Rochefoucauld remarked in the 1600s: “The principal point of cleverness is to know how to value things just as they deserve.” %!s()

“I had to get to UCLH [one of London’s major hospitals]. My own child was headed there in a blue-lit ambulance because she was struggling to breathe London’s toxic air.”

Hundreds of thousands of parents across the UK capital had similar experiences in those years. Between about 6,000 and 8,000 people were dying prematurely each year in London in the late 2010s. In 2020, an inquest found that the death of Ella Adoo-Kissi-Debrah in 2013 had been a direct result of air pollution, the first time it had been listed as a cause of death in the UK.

This week, a study from Queen Mary University London (QMUL) showed a significant reversal. Children’s lungs had grown bigger and stronger since the most polluting vehicles have been restricted on the capital’s roads, scientists found. The ultra-low emission zone (Ulez) – introduced by London’s mayor Sadiq Khan in 2019 and expanded in 2021 and 2023 – charges £12.50 a day for more polluting vehicles that do not meet stringent standards. It has drastically cut levels of nitrogen dioxide, a harmful pollutant.

To Harper, who later worked for Khan as an environmental adviser, and now works independently, relief at his own child’s recovery has been supplemented by the satisfaction of knowing that far fewer parents will experience such dread. “To know now that same air is so much safer for the kids that breathe it because of the political choice of one man makes me feel proud of this city,” he said.

Nick Hopkinson, a professor of respiratory medicine at Imperial College London, said the QMUL study was not a surprise. “There is an absolute wealth of evidence that air pollution is a major cause of ill health and premature death across the planet,” he said. Air pollution kills 7 million people a year globally, according to the World Health Organization.

Poor air not only causes respiratory difficulties and the stunting of children’s lungs, but also heart disease and stroke. It is associated with some forms of cancer, dementia, autism and cognitive difficulties in children too. New studies are adding to our knowledge, but we have known for decades of the basic dangers of dirty air. So why is there so much opposition to curbs on air pollution, which are clearly sound public health measures?

This opposition, culture war even, is central for all cities hoping to follow Khan’s lead. The leading causes of air pollution vary around the world – coal use in power stations in China and India, agricultural burning in south-east Asia, waste burning in many of the poorest countries and dust and local industry in others – but road traffic is a constant.

The technology to cut all of these types of air pollution already exists, says a former social development minister of Ecuador and now executive director of the Breathe Cities initiative, whose supporters include the Clean Air Fund and C40, a global coalition of mayors and leaders of nearly 100 major cities. “We know that we have the means to make a difference, and we are sharing knowledge and success stories so they can be replicated with impactful policies,” she said.

In the case of coal, the alternatives are renewable energy. Halting the burning of forests and peatlands would also combat the climate crisis. The alternatives to polluting traffic are public transport, electric vehicles, the redesign of cities and encouragement of active transport – walking and cycling. Trees are also important everywhere.

Paris and Barcelona have shown what is possible within a few years when mayors embrace the joys of cycling, following the lead long ago taken in Denmark and the Netherlands. New York’s Zohran Mamdani, now wearing a cycle helmet, hopes to do the same. But elsewhere in car-dominated America, infamously poor public transport provision, an intractable SUV culture and a president in Donald Trump intent on rolling back environmental protections mean widespread change is unlikely.

Other cities are working to make a difference. Bogotá is about to embark on its second Zuma, a clean air zone that involves expanding cycle lanes and footpaths, paving roads to suppress dust and planting trees. The Colombian capital won last year’s global Earthshot prize for its first Zuma. The city’s environment secretary, Adriana Soto, said: “Residents helped shape the design, particularly how the works would maintain access to their homes and respond to local mobility needs.” Hundreds of cities around the world now have clean air zones of varying sizes.

Another is Kenya’s capital, Nairobi, which passed an air quality act in 2022 that requires harmful emissions to be halved by 2035 and has installed a network of 50 low-cost monitoring stations as a first step. Imports of vehicles more than eight years old are banned, and the city is consulting on establishing clean air zones. The lead for the C40 Breathe Cities project in Nairobi, Victor Indasi, said: “More remains to be done, but Nairobi is firmly on the right trajectory.”

Bulgaria’s capital, Sofia, is saddled with old-fashioned heating systems dependent on coal and wood burning. These are being phased out and replaced with cleaner heat pumps, subsidised for those on low incomes, as part of a city-wide introduction – the first in south-east Europe – of low-emissions zones for transport and heating. Recent results show a 22% reduction in the average concentration of PM10 particulate matter, and a halving of days that exceed air quality guidance levels. Poland is also cutting its reliance on solid fuels.

In Australia, Melbourne’s mayor was inspired by a speech Khan gave last year before Brazil’s Cop30 climate summit. The first step was discovering that state-level monitoring was telling the city little about the real quality of its air at a neighbourhood level. “It was surprising to us that this wasn’t being measured,” a spokesperson said. With the new monitoring data coming in, the mayor has yet to decide what actions will be taken, whether a clean air zone or other restrictions, or carrots in the form of scrappage schemes and improved public transport. “We are still at the beginning of this journey,” the spokesperson said.

London’s low emission zone might not have happened if Khan had not staked his reputation on it. When he launched the scheme in 2019, he faced fierce opposition from political opponents and large groups of voters, and even his own party. The plan was challenged in the courts, businesses complained of the supposedly crippling costs, and Khan received personal threats and shouts of “Nazi” at public appearances. One survey found 60% of Londoners who replied to a consultation were against it.

In 2023, the resurgent Labour party under its new leader Keir Starmer faced an important byelection in Uxbridge, on the outer edge of London. The Ulez was a key issue for voters and Labour lost by a small margin, leading Starmer’s advisers to blame the zone and demand row-backs in London and elsewhere. Andy Burnham, then the mayor of Greater Manchester, who took over from Starmer as prime minister this autumn, abandoned plans for a Ulez after Uxbridge.

In all of these cities there is a sense of caution, even hesitancy, and none are moving fast enough. In London, the successes of the Ulez are threatened by the ballooning number of wood-burning stoves, a fashion accessory rather than a necessity for most of their urban owners. These are now a bigger source of particle pollution than traffic. More than 110,000 children were admitted to hospital with breathing difficulties in London in 2024.

Monitoring is a vital first step, but what is done with the data is what counts. That will involve tough measures. “There is always a price. You have to change people’s behaviour, you have to change the transport, the heating, the energy generation,” said Hopkinson. “The way things are in cities is the way they have developed over long periods, and changing all that is difficult.”

The pollution is often invisible. As Vaca Jones put it: “Air is not a tangible public good that people can feel and see. That means it is not a priority on the political agenda.”

Vested interests also intervene. Though studies show that air pollution takes an economic as well as a physical toll, businesses often complain – with little evidence – that curbs on driving will hurt them.

There is also the fact that the people driving the cars and operating the polluting factories are often not the same people whose children are being admitted to hospital. “Those who are suffering are those who have done least to cause the pollution,” said Chris Griffiths, a professor of primary care at Oxford University.

When Khan embarked on London’s Ulez, hysterical misinformation quickly spread online, suggesting the mayor intended to shut down London for his own ends. His deputy, Mete Coban, recalls the shock. “The personal abuse Sadiq got on the back of it was incredible,” he said. “I think it’s this, the misinformation, that is the number one thing that puts off other countries from doing this.”

Hopkinson said: “There is a kind of madness about it. “Disinformation, which is lying to people, is confusing them. They are trying to persuade people that air pollution is not bad for them, and climate change is not real.

“They are spreading wild conspiracy theories [for instance that 15-minute cities, where all amenities can be accessed within a small area, are devices of government control] and the far right are exploiting this and generating grievance.”

With the latest study adding to a growing pool of evidence of the Ulez’s success, proponents of cleaner urban air hope to be able to cut through the madness with facts that speak loudly to parents and responsible people everywhere.

As Hopkinson put it: “The question for pro-pollution campaigners is ‘why do you hate children?’” %!s()

The statement win on Sunday ensured the hosts ended the two-match series in a draw.

The visitors headed into the Mackay showdown on a high after shocking the world’s top-ranked team by nine wickets in the first Test at Darwin.

But they were brought crashing back to earth by some scintillating fast bowling, led by man of the match Starc, who finished with 10-51 for the match to level the series.

After 18 wickets fell on a frenetic first day of the five-day match, the hosts were all out for 210 in the second day’s morning session, securing a lead of 146 on the back of Cameron Green’s 67.

Paceman Shoriful Islam claimed 7-48 — the best figures by a Bangladeshi bowler in an away Test.

The visitors collapsed for just 64 in their first innings, when Starc took 6-12, and he was the chief tormentor again, claiming 4-39 as Bangladesh crumbled to be 95 all out.

Skipper Pat Cummins chipped in with 3-10 and Nathan Lyon 3-25.

“A lot of things went right these couple [of] days and a lot of things went wrong last week,” said Cummins.

“But yeah, we needed to bounce back, and it was a great few days from the lads. I thought everyone was sharp. It was a great start, and we didn’t really let up after that.”

Bangladesh had a horror start, with Shadman Islam lasting just three balls, playing a meaningless stroke to a moving Starc delivery and edging to Marnus Labuschagne at point.

Mominul Haque was equally clueless next ball, failing to offer a shot, with Starc removing his off stump and setting up a hat-trick opportunity.

Captain Najmul Hossain Shanto, who was out for a golden duck in the first innings, this time survived to deny Starc.

But their woes deepened when Tanzid Hasan departed soon after for nine. Starc was again involved, taking a catch in the deep after Tanzid pulled Cummins.

Shanto did well to soak up the pressure after lunch for 46 deliveries, but he eventually lost focus against the off-spin of Lyon, holing out to the safe hands of Starc at long off on 31.

And when Starc was brought back into the attack, he was front and centre again, enticing a loose shot from Litton Das, who was caught without scoring by Beau Webster at third slip.

Mehidy Hasan Miraz struggled to get going and departed to an equally poor shot, lashing at a Cummins bouncer and being collected by wicketkeeper Alex Carey.

The Australian captain then removed Hasan Mahmud (2) before Starc took his 10th for the match — the fourth time he has achieved the feat — when he bowled Taijul Islam for 7 runs.

Lyon quickly wrapped up the tail in the first over after tea.

“We know how good Australia are. They showed their character,” said Shanto.

“We were not up to the mark with the bat. It was challenging conditions here with extra bounce and swing; we would like to learn from this.”

Australia resumed at 165-8, with Green on 51 and Lyon on 10.

Green, who scored a century during the first Test in Darwin, kept the scoreboard moving early with quick singles.

But he was undone by an inswinging yorker from the impressive Shoriful that hit him on the toe, with a review declaring him out lbw.

A valiant Lyon brought up the 200 for Australia with a towering six off Taskin, and with Josh Hazlewood adding 24 crucial runs for the last wicket.

The innings ended when Hazlewood skied Taijul to Shadman on six, leaving Lyon not out on 32. %!s()

When a star passes dangerously close to one of these black holes, destruction is not always immediate. Some stars survive the encounter and return for additional close passes, generating a fresh burst of light each time.

These events, known as repeating partial tidal disruption events (rpTDEs), allow astronomers to observe the same star interacting with the same black hole multiple times. Wide-field time-domain surveys make this possible by repeatedly scanning large regions of the sky and tracking objects whose brightness changes.

Yet some of these systems have presented astronomers with a mystery. Instead of producing similar flares on each return, they become steadily fainter. For years, theoretical models struggled to reproduce that behavior.

New research from astrophysicists at Syracuse University suggests that a previously underappreciated property of the star could provide the answer: how rapidly it was spinning before its first close encounter with the black hole.

The study, published in The Astrophysical Journal, was led by doctoral student Ananya Bandopadhyay, working with postdoctoral researcher Benjamin Amend and associate professor Eric Coughlin, all in the Department of Physics, along with collaborators at other institutions.

How Black Holes Tear Apart Stars

In a standard tidal disruption event (TDE), the gravitational pull from a black hole varies so strongly from one side of a nearby star to the other that the star is completely torn apart.

The resulting stellar debris begins falling toward, or "accretes" onto, the black hole. As that material loses energy, it releases light over periods ranging from days to months.

Black holes themselves do not emit light, but a TDE temporarily supplies material that can illuminate the region around one. That glow gives astronomers an indirect way to investigate objects that would otherwise be invisible.

Not every encounter ends with the star being completely destroyed. If a star passes close to a black hole without crossing the threshold for total disruption, it can lose only part of its mass, producing a partial TDE. During a repeating partial TDE, the star's surviving core remains in orbit and returns for additional close encounters, shedding more material each time. These passages can occur months or several years apart.

Why Some Black Hole Flares Keep Fading

The amount of material stripped from a star during repeated encounters depends in part on the star's internal structure. Bandopadhyay likens a low-mass star to a fluffy meringue. Such a star can become increasingly susceptible to the black hole's tidal forces.

A higher mass star behaves differently. Its material is more concentrated toward the center, with an onion-like internal structure. It can lose its outer layers while its dense core remains comparatively unchanged, causing the amount of mass lost to decrease over successive encounters.

Those structural differences can help explain why not all rpTDEs evolve in the same way. But one observation has been especially difficult to understand. Of the roughly 10 repeating systems identified so far, four have shown flares that become progressively dimmer.

It might seem natural to assume that smaller amounts of stripped material would simply produce weaker flares. Previous hydrodynamical simulations, however, revealed a complication. Even when the star lost less material during each passage, the models still predicted flares with approximately the same peak brightness.

"We were puzzled by this for two years," Bandopadhyay says.

Their previous work uncovered another important consequence of the black hole's tidal forces. Besides pulling material away from the star, those forces also apply torque, causing the star to rotate faster after each close encounter.

That increased rotation changes how quickly stripped material returns toward the black hole. Even though less material comes back, it does so over a shorter period of time. The more concentrated flow helps maintain a similar peak fallback rate, and therefore roughly the same predicted flare brightness.

A Rapidly Spinning Star Changes the Picture

To reproduce the fading flares astronomers actually observe, the researchers needed what Bandopadhyay called "a new ingredient": a star that was already rotating rapidly before its first encounter with the black hole.

The new simulations suggest that such a star cannot be spun up nearly as much during later passages. Without a large increase in rotation after each encounter, the time required for the stripped material to fall back toward the black hole remains relatively steady.

That changes the outcome. As progressively less material is stripped from the star, the peak fallback rate also decreases. The predicted flare can then become fainter with each encounter, matching what astronomers have seen.

How the Star May Have Been Captured

The finding raises another question: Why would a star approaching a supermassive black hole already be spinning so quickly?

"It is also extremely difficult to 'bind' a star to a supermassive black hole so tightly that it orbits the black hole in a matter of months, and yet they seem to do so in rpTDEs," Coughlin says.

A process known as the Hills mechanism may provide an explanation for both the rapid rotation and the star's unusually tight orbit.

In this scenario, two stars orbiting closely around each other approach a supermassive black hole. The black hole's gravity tears the binary system apart. One star is hurled away, while the other is captured into orbit around the black hole.

Stars in a very close binary can become tidally locked, meaning each star rotates on its axis at the same rate that the pair orbits one another. The tighter the binary, the shorter the orbital period and the faster a tidally locked star must rotate.

To leave one captured star on the short orbit observed in rpTDEs, the original binary system would have to be extremely compact. That same tight configuration would naturally produce a rapidly spinning, tidally locked star before the black hole captured it.

"Ananya's work demonstrates that each of these peculiarities can be explained by the same underlying phenomenon: the tidal destruction of a binary system and the capture of one of the stars," Coughlin says. "From a theoretical standpoint, this is a major step forward in our understanding of the physics at play in these systems."

A Possible Connection to the Milky Way

The implications may extend beyond distant repeating flare systems. Coughlin notes that Hills capture may also be responsible for some of the stars now orbiting Sagittarius A*, the supermassive black hole at the center of the Milky Way.

If so, the same mechanism that may explain fading rpTDE flares could also help astronomers understand some of the unusual stellar populations surrounding the black hole in what Coughlin calls "our own cosmological backyard." %!s()

Carney said he walked away from a deal because he was not prepared to "compromise Canada's sovereignty", adding he felt the country was being "attacked" by the US.

"They asked too much and they offered too little," Carney told Canadians on Saturday morning.

The prime minister confirmed he would match US tariffs "dollar-for-dollar" from 8 September, including levies on steel, dairy, appliances and electronics.

US President Donald Trump has yet to comment.

Both countries were optimistic of a deal earlier in the week, but talks fell apart after both sides accused the other of making last-minute changes.

Now that talks have broken down, Canada has been hit with new 50% US tariffs on a range of Canadian goods, in addition to existing US tariffs on Canadian steel, aluminium, automobiles and lumber.

The prime minister said Canada was retaliating "reluctantly" to protect its interests. Details of the counter-measures will be released in the coming days, he said.

Leaders of other federal parties in Canada, including the Conservative opposition, voiced support for the prime minister, as have the leaders of some Canadian provinces.

Tariffs - a tax placed on imports from other countries - have been a key component of Trump's trade policy, with the US president arguing they will boost American manufacturing and create jobs at home.

Critics, however, say tariffs have raised prices for American consumers and have disrupted and damaged the global economy.

Carney's speech on Saturday was his first address to Canadians after he announced late on Friday night that a deal would not be reached, accusing the Americans of making "last-minute changes" that were "unfair" and "uneconomic".

US negotiators, meanwhile, accused Canada of making "new demands and walk backs" on terms which had been previously agreed.

On Saturday, US trade representative Jamieson Greer told Fox News there were no plans to resume talks.

He added that the US had been prepared to cut some of its tariffs on Canada as part of the ill-fated deal.

Carney, however, said the US presented "unacceptable" terms that were too restrictive, including curbing Canada's ability to strike new trade deals with other countries.

He also rejected the assertion that Canada had made last-minute requests: "We clarified what was on offer and were continually disappointed by the answers."

Doug Ford, the leader of Canada's most populous province Ontario, backed Carney and said Trump "can't be trusted".

The Conservative opposition leader Pierre Poilievre also appeared to support Carney, calling the latest US levies "unjustified".

Quebec's premier Christine Fréchette warned jobs would likely be lost as a result of the new US tariffs.

"Behind these numbers are real people," she said, vowing to introduce measures to support impacted sectors.

Alberta premier Danielle Smith, whose oil-rich province has worked to make inroads with the Trump administration, urged both sides to resume talks.

The collapse comes after more than a year of on-again, off-again trade talks, which intensified in recent weeks after a US-imposed deadline to strike a deal or face further tariffs.

The two countries were also engaged in a mandatory review of an existing North American free trade agreement, known as USMCA, with Mexico.

The pact was signed by Trump during his first term to replace the 1994 North American Free Trade Agreement. It underpins $1.6tn (£1.2tn) in annual trilateral trade across Canada, the US and Mexico.

Earlier this summer, Canada and Mexico had formally requested for USMCA to be renewed for another 16 years. The US, however, declined to renew it in its current form.

Asked how Friday's trade talk collapse would impact the North American free trade pact, Carney said it was "certainly not good news". %!s()