Indian boxing star Mary Kom opened up on the heartache she faced returning to competition after having children as she captured her first Commonwealth Games title on Saturday at the age of 35.
Mother-of-three Kom, the subject of a Bollywood film in 2014, won a unanimous points decision over Northern Ireland's Kristina O'Hara in their light-flyweight final on Australia's Gold Coast.
It is Kom's first Commonwealth medal of any colour and burnishes a heaving trophy cabinet that already includes an Olympic bronze from London 2012 and five amateur world championship titles.
The trail-blazer, known in India as "Magnificent Mary", celebrated with a victory parade on her coach's shoulders and she had tears in her eyes afterwards.
Kom, devastated not to gain a wildcard for the 2016 Rio Olympics, gave mixed signals about whether she would now target the Tokyo 2020 Games and a first Olympic gold.
"That is difficult to say, but I will try," said Kom, who is already older than most of her opponents.
Part of the problem is that at the Olympics there is no light-flyweight category, meaning Kom would have to step up to flyweight, which would mean fighting bigger and stronger opponents.
"If I am still super-fit in another year I will try," she added of another tilt at a maiden Olympic gold.
"If I am not enough fit, maybe I will retire."
Kom, who went from a poor village life to sports stardom and celebrity, has three sons -- twin boys aged 10 and another of five.
They are not boxing fans.
"It's very hard," said Kom, reflecting on returning to the boxing ring after the birth of her youngest child.
"My husband and parents are very supportive, but if I wasn't strong that would not be possible.
"I'm still strong from here and here," said Kom, pointing to her heart and her head.
Several sportswomen have returned to competition at these Commonwealth Games following the high-profile case of tennis star Serena Williams.
Double Olympic shot put champion Valerie Adams gave birth last October and won silver on Friday, while Vanuatu beach volleyball player Miller Pata won bronze in front of her seven-month-old toddler, Tommy.
Kom, who has spoken repeatedly at the Games in Australia about the problems facing women and girls in India, said being away for long periods for competition and to train was challenging emotionally.
"Willpower, mentally strong," she said, asked how she coped with it.
"That's really required being a woman, especially leaving the family, leaving the children, they are very small."
THE UK government said on Monday that more than £1 billion has been paid to self-employed managers of Post Office branches who were affected by faults in the Horizon accounting software.
The update comes a few weeks after Alan Bates, the former subpostmaster who led the campaign for justice, criticised the compensation process, calling it “quasi-kangaroo courts”.
The Department for Business and Trade (DBT) said it had received 11,208 claims in total. Of these, 7,569 have been settled, while 3,709 are still pending.
Between 1999 and 2015, the Post Office prosecuted over 900 subpostmasters based on errors in Horizon, a software developed by Fujitsu. The system incorrectly showed shortfalls in branch accounts.
Many subpostmasters were forced to repay the shortfalls and later went bankrupt. Some were imprisoned and faced social stigma.
At least four people took their own lives, and several others died before they were exonerated.
In 2019, the High Court ruled that computer errors, not criminal behaviour, had led to the missing funds.
Alan Bates, who was knighted by King Charles III for his efforts to expose the issue, has criticised how the DBT is handling the assessment of claims.
"The department sits in judgement of the claims and alters the goal posts as and when it chooses," he told The Sunday Times last month.
Public attention around the case grew in January 2024 following a television drama about the subpostmasters’ experiences, which sparked widespread public reaction.
Following that, Fujitsu’s European director Paul Patterson appeared before a parliamentary committee and apologised for the firm’s role in prosecutions based on incorrect data. He said the company was “truly sorry” for “this appalling miscarriage of justice”.
Post Office Minister Gareth Thomas said the government had prioritised faster payments since taking office in July 2024.
"We are settling cases every day and getting compensation out more quickly for the most complex cases, but the job isn't done until every postmaster has received fair and just redress," he said.
By clicking the 'Subscribe’, you agree to receive our newsletter, marketing communications and industry
partners/sponsors sharing promotional product information via email and print communication from Garavi Gujarat
Publications Ltd and subsidiaries. You have the right to withdraw your consent at any time by clicking the
unsubscribe link in our emails. We will use your email address to personalize our communications and send you
relevant offers. Your data will be stored up to 30 days after unsubscribing.
Contact us at data@amg.biz to see how we manage and store your data.
Apple researchers evaluated several prominent generative AI systems
A new research paper from Apple has exposed serious shortcomings in the reasoning abilities of some of today’s most advanced artificial intelligence (AI) systems. Despite being marketed as powerful tools capable of solving complex problems, the study shows that these models still struggle with basic logical tasks, raising questions about the real capabilities of large language and reasoning models.
AI models fail child-level logic tests
Apple researchers evaluated several prominent generative AI systems, including ChatGPT, Claude, and DeepSeek, using classic problem-solving tasks. One of the tests was the well-known Tower of Hanoi puzzle, which requires moving discs across pegs while following specific rules.
While the puzzle is simple enough for a bright child to solve, most AI models failed when asked to handle scenarios involving more than seven discs. Accuracy fell below 80% with seven discs, and performance dropped even further with eight. According to co-lead author Iman Mirzadeh, the issue wasn't just solving the puzzle — it was that the models couldn’t follow a logical thought process even when given the solution algorithm.
“They fail to reason in a step-by-step, structured way,” he said, noting that the models’ approach was neither logical nor intelligent.
The myth of scaling exposed
The results challenge one of the AI industry’s most commonly held beliefs: that simply scaling models — making them larger and feeding them more data — will lead to better performance. Apple’s research provides strong evidence that this is not always true.
Gary Marcus, a well-known AI researcher and commentator, called the findings a reality check. Venture capitalist Josh Wolfe even coined a new verb, “to GaryMarcus”, meaning to critically debunk exaggerated claims about AI. The Apple study, Wolfe argued, had done exactly that by revealing the real limits of model reasoning.
Marcus has long argued that AI systems, particularly those based on neural networks, can only generalise within the data they’ve seen before. Once asked to work beyond that training distribution, they often break down — a pattern clearly confirmed in Apple’s tests.
AI is not yet a substitute for human logic
To be clear, even humans make errors on the more complex versions of the Tower of Hanoi. However, AI systems were supposed to improve on this, not replicate human flaws. As Marcus points out, artificial general intelligence (AGI) should combine human creativity with machine-level precision. But instead of outperforming people in logic and reliability, today’s large models still make basic errors.
Most AI models failed when asked to handle scenarios involving more than seven discsiStock
Apple’s results also support concerns raised by Arizona State University’s Subbarao Kambhampati, who has cautioned against assuming AI models reason like humans. In reality, they often skip steps or fail to understand the underlying principles of a problem, despite producing convincing-sounding answers.
Caution urged for businesses and society
The implications are significant for businesses looking to integrate AI into their operations. While models such as GPT-4, Claude, and others perform well in areas like writing, coding, and brainstorming, they remain unreliable for high-stakes decision-making. As Marcus points out, these systems can’t yet outperform classical algorithms in areas like database management, protein folding, or strategic games like chess.
This unpredictability limits how much society can rely on generative AI. While the technology will continue to be useful in supporting human tasks, it is far from being a replacement for human judgement or traditional rule-based systems in critical contexts.
The illusion of intelligence
Perhaps most concerning is how easily these models can appear more capable than they are. If an AI performs well on an easy test, users may assume it can handle more complex problems too. But Apple’s study shows this confidence can be misplaced. The same model that solves a four-disc puzzle may completely fail when asked to solve one with eight.
This illusion of intelligence could lead to overtrust in AI systems — something experts warn must be avoided if the technology is to be used responsibly.
Rethinking the future of AI
Despite the findings, Marcus remains optimistic about AI’s future, just not in its current form. He believes that hybrid approaches, combining classical logic with modern computing power, could eventually produce more reliable systems. But he is sceptical that current LLM-based systems are the answer.
The Apple paper shows that hype around generative AI has outpaced its real-world abilities. Until AI can reason in a consistent, logical manner — not just produce convincing text — it will remain limited in scope.
As researchers and developers reflect on these findings, one thing is clear: the path to truly intelligent machines will require more than just scaling up. It will demand smarter, better-designed models that prioritise reliability over illusion.
Keep ReadingShow less
Kendrick Lamar on stage after winning Best Video for Not Like Us at the 2025 BET Awards
There’s no question who ruled the night at the 2025 BET Awards. Kendrick Lamar walked away with five trophies, including Album of the Year for GNX and Video of the Year for Not Like Us. The ceremony, held at the Peacock Theatre in Los Angeles, paid tribute to Black excellence in music, acting, and sports, and Kendrick made sure his voice and message were front and centre.
Kendrick Lamar accepts Best Album for GNX at the BET Awards in Los AngelesGetty Images
Lamar, Doechii, and a moment that mattered
The GNX rapper also won Best Male Hip-Hop Artist and shared the Video Director of the Year award with Dave Free. Accepting his award, Lamar reflected on his long-standing relationship with BET, saying, “They’ve always kept the culture at the core and put me in the middle of it.”
— (@)
Rising rapper Doechii also made headlines, not just for winning Best Female Hip-Hop Artist but for calling out former US President Donald Trump. In a fiery speech, she criticised the use of military force to break up protests in LA following ICE raids in Latino communities. “Every time we stand up for our rights, the military is sent in. What kind of leadership is that?” she asked the crowd.
A night of icons and throwbacks
The night wasn’t just about current stars. The BET Ultimate Icon Award honoured Mariah Carey, Snoop Dogg, Jamie Foxx, and gospel star Kirk Franklin for their decades of influence in music, entertainment, and community action. Stevie Wonder presented Foxx’s award as he opened up about his recovery from a 2023 stroke. “When I saw the ‘In Memoriam’ segment, I thought that could’ve been me,” he said.
Mariah Carey celebrates her Ultimate Icon Award with a rare live performanceGetty Images
Mariah Carey, presented her award by Busta Rhymes, lit up the stage with a brief but heartfelt speech. “If you're going to win one of these, it might as well be the Ultimate Icon,” she joked, hinting at the ups and downs of her legendary career.
With Lamar and SZA heading back on tour and voices like Doechii’s rising louder, the 2025 BET Awards proved it is a stage for change.
Keep ReadingShow less
Harris Dickinson debuts as Rhode’s first male model in new Glazing Mist campaign
Rhode, Hailey Bieber’s rapidly growing skincare label, has chosen Babygirl actor Harris Dickinson as the face of its new product, the Glazing Mist. Known for his understated performances and growing fashion presence, Dickinson is the first male model to front a campaign for the brand. The new mist, priced at £24 (₹2,520), goes live on 24 June exclusively at rhodeskin.com.
A surprising choice that feels just right
The campaign, shot in stark black-and-white, features close-up visuals of Dickinson misting his face with the Glazing Mist, designed to refresh skin and deliver a natural glow. With ingredients like ectoin, ceramides, and magnolia extract, the formula offers both hydration and barrier support. It has also earned a stamp of approval from the National Eczema Association.
Hailey Bieber explained her choice simply: she watched Babygirl, loved Dickinson’s performance, and sensed he could bring the kind of quiet edge she wanted. “I didn’t want it to feel too ‘on-the-nose’ with the film,” she said. “It had to be chic and sporty.” The internet, however, couldn’t resist the connection to Dickinson’s now-infamous milk scene in the film, and fans were quick to point out how the dewy mist echoed that very vibe.
Rhode expands, without changing its DNA
Since its 2022 launch, Rhode has stood out for sleek, limited product drops and a focused digital presence. Its recent £790 million (₹82,950 crore) acquisition by e.l.f. Beauty has only amplified interest in where the brand goes next. While introducing a male face might suggest a shift, Bieber was clear: this isn’t about launching a men’s line but widening the brand’s appeal. “I want everyone to see themselves in Rhode,” she said.
Dickinson’s inclusion reflects a subtle repositioning. Best known for roles in Beach Rats, Triangle of Sadness, and now Babygirl, he has also been a regular in fashion, working with Dior and Loewe.
With this campaign, Rhode continues to evolve clearly, without losing the minimalist identity that made it stand out in the first place.
Keep ReadingShow less
HBO’s Harry Potter reboot slammed for casting Italian actress as Indian character Parvati Patil
HBO’s upcoming Harry Potter reboot is already under fire, this time for casting Italian actress Alessia Leoni as Parvati Patil, a character rooted in Indian culture. The announcement has triggered widespread backlash from fans who accuse the makers of side-lining authentic South Asian representation in favour of what many see as surface-level diversity.
Parvati, originally portrayed by Shefali Chowdhury in the films, is one of the few explicitly Indian characters in the Harry Potter universe. The decision to cast someone with no clear Indian heritage has led to heated discussions across platforms like Reddit and X, with fans calling out what they perceive as performative inclusivity.
Fans question why Indian talent was overlooked again
Many online comments have shared similar frustrations: that casting a brown-skinned actor is not the same as casting someone of Indian origin. “They just picked someone who looks brown and called it a day,” one Redditor commented. Another added, “Parvati is not just a name, it’s a direct reference to Hindu culture. It deserved more thought.”
This is not the first controversy to hit the reboot. Earlier announcements like casting Paapa Essiedu as Snape and Arabella Stanton as Hermione also drew criticism from those who felt the series was playing fast and loose with established character identities. Supporters of accurate representation argue that while the intent may be diversity, the execution often feels tokenistic.
— (@)
Adding fuel to the fire is the casting history of the original films, where British-Bangladeshi actors Shefali Chowdhury and Afshan Azad played the Patil twins after being discovered through grassroots-level auditions. That sense of grounded authenticity, fans argue, is missing in the current reboot.
HBO's broader casting choices also under scrutiny
Alongside Leoni, HBO announced several other cast members: Katherine Parkinson as Molly Weasley, Lox Pratt as Draco Malfoy, Johnny Flynn as Lucius Malfoy, Leo Earley as Seamus Finnigan, Sienna Moosah as Lavender Brown, Bel Powley and Daniel Rigby as the Dursleys, and Bertie Carvel as Cornelius Fudge.
— (@)
The series, hoped to be a faithful adaptation of J.K. Rowling’s books, is scheduled to premiere in 2026. But with these early casting calls, many longtime fans feel HBO is missing the mark on cultural nuance—and all this before a single spell has been cast on screen.