fraud

Former head of crypto platform Celsius sentenced 12 years

  • Mashinsky, 50, pleaded guilty last December to securities fraud in a deal that reduced the level of charges he faced.
  • The founder and former CEO of bankrupt cryptocurrency trading platform Celsius, Alexander Mashinsky, was sentenced Thursday to 12 years in prison on fraud charges.
  • Mashinsky, 50, pleaded guilty last December to securities fraud in a deal that reduced the level of charges he faced.
The founder and former CEO of bankrupt cryptocurrency trading platform Celsius, Alexander Mashinsky, was sentenced Thursday to 12 years in prison on fraud charges.
Mashinsky, 50, pleaded guilty last December to securities fraud in a deal that reduced the level of charges he faced.
The sentence comes down nearly three years after the startup's collapse as a cryptocurrency platform, which offered customers the ability to invest in digital currencies, including its own coin, CEL.
According to the indictment, Celsius executives took more than $4 billion in customers' assets to finance the platform's operations, make unsecured loans and invest in high-risk items.
Mashinsky was also accused of manipulating the price of CEL by using customers' funds to purchase the currency, artificially inflating its price.
At its peak in late 2021, Celsius had more than one million clients and held more than $25 billion in assets.
But the company hit hard times in the spring of 2022 as the value of cryptocurrencies plummeted.
Facing deep customer withdrawals, Celsius on June 12, 2022 froze over $4.7 billion in customer accounts before filing for bankruptcy protection a month later.
A progress report published in March found that 93 percent of the frozen assets had been recovered and returned to former Celsius customers.
The 2022 cryptocurrency collapse affected a number of other startups in the field, including FTX, the second-largest crypto exchange that filed for bankruptcy in November 2022.
tu/elm/pno/jgc/sla

economy

Facing a broken economy, Ghana's tech-savvy teens turn to fraud

BY WINIFRED LARTEY

  • As dusk settles over Nima, Ghost logs off, pockets his burner phone, and steps into the fading light. 
  • In the dusty alleys of Nima, a shanty town in the heart of Ghana's capital, a 17-year-old called Ghost reclines on a faded plastic chair inside a dimly lit internet cafe.
  • As dusk settles over Nima, Ghost logs off, pockets his burner phone, and steps into the fading light. 
In the dusty alleys of Nima, a shanty town in the heart of Ghana's capital, a 17-year-old called Ghost reclines on a faded plastic chair inside a dimly lit internet cafe.
Outside, barefoot children chase a punctured football. Inside, Ghost's fingers dance across the keyboard, his eyes locked on WhatsApp as he engineers a phishing scam that could earn him thousands of cedis in just a few hours.
"I made GHC12,000 ($770) last month," Ghost told AFP, his voice low and calculated, describing an online store he set up on Instagram. 
"People bought phones and laptops. None of it existed."
Ghost, a pseudonym he gave AFP, is one of a growing number of Accra's teenagers turning to cybercrime to survive in a country mired in economic crisis, battling both youth unemployment and broken dreams. 
From mobile money fraud to investment scams, a murky digital underworld is sucking in minors, many working from their bedrooms or small kiosks operating in plain sight -- anywhere with a stable internet connection.

Too good to be true

Nima's 441 neighbourhood is a tightly packed, working-class community of corrugated iron housing and open drains. Opportunities are scarce -- but mobile phones are not. 
Ghana's Cyber Security Authority (CSA) has raised the alarm over a surge in cyberfraud, with financial losses tallying $282,776 between January and March 2025, nearly doubling the $154,241 recorded during the same period in 2024. 
Officials warn that youth-led scams, powered by social media and peer pressure, are driving the spike.
Phishing scams, brand impersonation and fake online shops dominate the scene, with teenagers posing on Snapchat and TikTok as vendors with offers that are too good to be true.
Mercy Adumoah, 20, was one such victim. 
"I saw a page on Snapchat selling heels. I needed a pair for an event, so I paid without thinking twice," she recounted.
After they received the money, the sellers blocked her account.
Experts say these crimes have become systemic in a country with a battered economy that is still recovering from a 2023 debt default. Inflation shot above 54 percent in 2022 and has remained above 20 percent into 2025.
At a junior high school near Nima, teacher Mohammed Inusah has witnessed a transformation.
"Some of my students have iPhones more expensive than my salary," he told AFP. "They flaunt cash, buy designer clothes and sneakers."
"The parents are either unaware or too afraid to confront them."

'I know it's wrong'

A soft-spoken boy with a mop of dreadlocks who gave the name Tricky said he got his start in scamming by copying scripts from online forums in Nigeria. 
Later, his cousin taught him mobile money fraud -- how to pose as an agent from the local telecom company to get access to people's accounts.
Tricky claims his biggest hit since in his two years of scamming was $500 -- twice the monthly salary of a public health nurse or a teacher. 
"I bought clothes, helped my mum pay rent," he said. "I know it's wrong, but tell me, what else can I do?"
The CSA has also flagged a dramatic rise in online investment fraud. Between January and August 2024 alone, 149 cases were recorded with losses nearing $128,534.
"I fell for one," admits 18-year-old "Bronzy", who went from victim to perpetrator. "A guy scammed me using a fake forex trading site. So I learnt the game."
He now runs a group on Telegram promising 20 percent weekly returns.
"People invest and I disappear," he said.
Abubakar Issaka, president of the Cyber Security Experts Association of Ghana, said the situation is only getting worse.
"The regulations exist... but enforcement is weak. The number of professionals is not growing fast enough to match the fraud cases," he said. 
Tracing perpetrators "is a challenge due to poor data integration" between telecoms operators and the national ID database, he added.
In some cases, the phone numbers used in frauds "belong to people who died years ago. Fraudsters are steps ahead."

No end in sight

Victims suffer anxiety and financial ruin. Scammers are not untouched. Ghost admits he's often afraid. 
"Sometimes, I can't sleep. I wonder if the police will knock. But when I see my friends living large, I feel like I must keep going."
As dusk settles over Nima, Ghost logs off, pockets his burner phone, and steps into the fading light. 
He didn't make any money that day. His world is one of quick, fickle wins and quiet fear. But with Ghana's economy showing no signs of rapid recovery, many more are likely to follow him into the shadows.
strs/nro/cw/srg/phz

AI

AI tool uses selfies to predict biological age and cancer survival

BY ISSAM AHMED

  • It was then tested on 6,196 cancer patients treated in the United States and the Netherlands, using photos snapped just before radiotherapy.
  • Doctors often start exams with the so-called "eyeball test" -- a snap judgment about whether the patient appears older or younger than their age, which can influence key medical decisions. 
  • It was then tested on 6,196 cancer patients treated in the United States and the Netherlands, using photos snapped just before radiotherapy.
Doctors often start exams with the so-called "eyeball test" -- a snap judgment about whether the patient appears older or younger than their age, which can influence key medical decisions. 
That intuitive assessment may soon get an AI upgrade.
FaceAge, a deep learning algorithm described Thursday in The  Lancet Digital Health, converts a simple headshot into a number that more accurately reflects a person's biological age rather than the birthday on their chart.
Trained on tens of thousands of photographs, it pegged cancer patients on average as biologically five years older than healthy peers. The study's authors say it could help doctors decide who can safely tolerate punishing treatments, and who might fare better with a gentler approach.
"We hypothesize that FaceAge could be used as a biomarker in cancer care to quantify a patient's biological age and help a doctor make these tough decisions," said co-senior author Raymond Mak, an oncologist at Mass Brigham Health, a Harvard-affiliated health system in Boston.
Consider two hypothetical patients: a spry 75‑year‑old whose biological age clocks in at 65, and a frail 60‑year‑old whose biology reads 70. Aggressive radiation might be appropriate for the former but risky for the latter. 
The same logic could help guide decisions about heart surgery, hip replacements or end-of-life care.  

Sharper lens on frailty

Growing evidence shows humans age at different rates, shaped by genes, stress, exercise, and habits like smoking or drinking. While pricey genetic tests can reveal how DNA wears over time, FaceAge promises insight using only a selfie.
The model was trained on 58,851 portraits of presumed-healthy adults over 60, culled from public datasets. 
It was then tested on 6,196 cancer patients treated in the United States and the Netherlands, using photos snapped just before radiotherapy. Patients with malignancies looked on average 4.79 years older biologically than their chronological age.
Among cancer patients, a higher FaceAge score strongly predicted worse survival -- even after accounting for actual age, sex, and tumor type -- and the hazard rose steeply for anyone whose biological reading tipped past 85.
Intriguingly, FaceAge appears to weigh the signs of aging differently than humans do. For example, being gray-haired or balding matters less than subtle changes in facial muscle tone. 
FaceAge boosted doctors' accuracy, too. Eight physicians were asked to examine headshots of terminal cancer patients and guess who would die within six months. Their success rate barely beat chance; with FaceAge data in hand, predictions improved sharply.
The model even affirmed a favorite internet meme, estimating actor Paul Rudd's biological age as 43 in a photo taken when he was 50.

Bias and ethics guardrails

AI tools have faced scrutiny for under‑serving non-white people. Mak said preliminary checks revealed no significant racial bias in FaceAge's predictions, but the group is training a second‑generation model on 20,000 patients.
They're also probing how factors like makeup, cosmetic surgery or room lighting variations could fool the system.
Ethics debates loom large. An AI that can read biological age from a selfie could prove a boon for clinicians, but also tempting for life insurers or employers seeking to gauge risk. 
"It is for sure something that needs attention, to assure that these technologies are used only in the benefit for the patient," said Hugo Aerts, the study's co-lead who directs MGB's AI in medicine program. 
Another dilemma: What happens when the mirror talks back? Learning that your body is biologically older than you thought may spur healthy changes -- or sow anxiety.
The researchers are planning to open a public-facing FaceAge portal where people can upload their own pictures to enroll in a research study to further validate the algorithm. Commercial versions aimed at clinicians may follow, but only after more validation.
ia/arp

Peru

Extremely online new pope unafraid to talk politics

  • Robert Prevost, now better known to the world as Pope Leo, quickly took to X to take a theological swipe at the vice president. 
  • From pillorying the US vice president to denouncing the death penalty, Pope Leo has proven unafraid to tackle prickly political issues on social media -- making him the first "extremely online" pontiff. 
  • Robert Prevost, now better known to the world as Pope Leo, quickly took to X to take a theological swipe at the vice president. 
From pillorying the US vice president to denouncing the death penalty, Pope Leo has proven unafraid to tackle prickly political issues on social media -- making him the first "extremely online" pontiff. 
When JD Vance suggested that Christians should love their family, neighbors, community and fellow citizens -- in that order -- one very notable Christian took umbrage. 
Robert Prevost, now better known to the world as Pope Leo, quickly took to X to take a theological swipe at the vice president. 
"JD Vance is wrong: Jesus doesn't ask us to rank our love for others," he wrote, reposting a columnist's searing opinion piece and prompting tens of thousands of likes and a slew of barbed comments. 
Pope Benedict may have been the first to tweet under the handle @Pontifex in 2012, but Pope Leo is undoubtedly the first to take the Chair of Saint Peter with the baggage of a long social media history. 
In 14 years since his X account was created, he has posted more than 400 times, opining on a range of hot-button issues: racism, sexual abuse by the clergy, Covid-19, the police murder of George Floyd and Russia's invasion of Ukraine. 
Perhaps unsurprisingly for an American who spent decades in Peru and took up its citizenship, immigration is a topic close to his heart. 
The new pope has notably amplified criticism of US President Donald Trump's immigration policies, reposting a 2017 article which called refugee bans "a dark hour of US history" and an abandonment of "American values."
He has repeatedly taken Vance to task, challenging a vice president whose religious views show all the zeal of a recent convert to Catholicism. 
It is clear from his ample online commentary, interviews and video blogs that retweets are almost always endorsements. 
In 2020, days after African American Floyd was suffocated to death under a police officer's knee, he implored fellow members of the clergy to speak up. 
"We need to hear more from leaders in the Church, to reject racism and seek justice," he posted. 
He has also demanded more action of the church in ousting members of the clergy who sexually abused children.
"If you are a victim of sexual abuse by a priest, report it," he told Peruvian paper La Republica this month. 
"We reject cover-ups and secrecy; that causes a lot of harm. We have to help people who have suffered due to wrongdoing." 
Embracing another contentious issue, in 2014 he wrote that it was "time to end the death penalty" and has repeated that point over the years in interviews, masses and in public remarks. 
"We have to be pro-life at all times" he once told assembled Peruvian journalists in his fluent and modestly accented Spanish.  
Still, he is also unafraid to post a joke, including a suggestion that while many people are intelligent, most are asymptomatic. 
Like many of us, the tempo of his social media posts appeared to increase during pandemic lockdowns. 
It is unclear if he will extend that social media chattiness from inside the Apostolic Palace. 
arb/sla

technology

India tells X to block over 8,000 accounts

BY ANUJ CHOPRA

  • "X has received executive orders from the Indian government requiring X to block over 8,000 accounts in India, subject to potential penalties including significant fines and imprisonment of the company's local employees," the site's global government affairs team said in a statement.
  • India has ordered X to block more than 8,000 accounts, the platform said Thursday, adding that it was reluctantly complying with what it described as government-imposed "censorship."
  • "X has received executive orders from the Indian government requiring X to block over 8,000 accounts in India, subject to potential penalties including significant fines and imprisonment of the company's local employees," the site's global government affairs team said in a statement.
India has ordered X to block more than 8,000 accounts, the platform said Thursday, adding that it was reluctantly complying with what it described as government-imposed "censorship."
The move appears to be part of India's sweeping crackdown targeting social media accounts of Pakistani politicians, celebrities and media organizations amid heightened tensions and deadly confrontations between the nuclear-armed neighbors.
The order, which X said includes demands to block international news organizations and other prominent users, comes a day after Meta banned a prominent Muslim news page on Instagram in India at New Delhi's request.
"X has received executive orders from the Indian government requiring X to block over 8,000 accounts in India, subject to potential penalties including significant fines and imprisonment of the company's local employees," the site's global government affairs team said in a statement.
It added that in most cases, the government had not specified which posts from the accounts violated Indian laws, and in many others, it provided no evidence or justification for the blocks.
The Elon Musk-owned platform said it disagreed with the demands but it had begun the process to withhold the specified accounts in India.
"Blocking entire accounts is not only unnecessary, it amounts to censorship of existing and future content, and is contrary to the fundamental right of free speech," the statement said.
"This is not an easy decision, however keeping the platform accessible in India is vital to Indians' ability to access information."
The move comes amid fierce fighting between India and Pakistan, two weeks after New Delhi blamed Islamabad for backing a deadly attack on tourists in the Indian-run side of the disputed Muslim-majority region of Kashmir.
Pakistan rejects the charge.
At least 48 people have been killed on both sides of the border in escalating violence since India launched air strikes on Wednesday that it said targeted "terrorist camps."
Both countries accused each other on Thursday of carrying out waves of drone attacks.
X said it could not make the Indian executive orders public due to legal restrictions, but it encouraged the impacted users to seek "appropriate relief from the courts."
It did not name the affected users, but in recent days the Indian media has reported that the country has blocked the X accounts of Pakistani politician Bilawal Bhutto Zardari and Pakistan's former prime minister and cricket captain Imran Khan.
India has also banned more than a dozen Pakistani YouTube channels for allegedly spreading "provocative" content, including Pakistani news outlets.
Pakistani Bollywood movie regulars Fawad Khan and Atif Aslam were also off limits in India, as well as a wide range of cricketers -- including star batters Babar Azam and Mohammad Rizwan and retired players Shahid Afridi and Wasim Akram.
Rising hostilities between the South Asian neighbors have unleashed an avalanche of online misinformation, with social media users circulating everything from deepfake videos to outdated images from unrelated conflicts, falsely linking them to the ongoing fighting.
ac/jbr

Gates

Bill Gates speeds up giving away fortune, blasts Musk

BY JOHN BIERS

  • The organization, which had more than $71 billion in assets at the end of 2023, has been credited with helping to reshape the world of global public health.
  • Microsoft co-founder Bill Gates announced Thursday an accelerated timeframe for giving away his fortune as he touted artificial intelligence as a game-changer to boost public health and save lives globally.
  • The organization, which had more than $71 billion in assets at the end of 2023, has been credited with helping to reshape the world of global public health.
Microsoft co-founder Bill Gates announced Thursday an accelerated timeframe for giving away his fortune as he touted artificial intelligence as a game-changer to boost public health and save lives globally.
Under a new timetable, the Gates Foundation will spend more than $200 billion over the next 20 years, shutting down in 2045. The organization had originally planned to close 20 years after Gates' death.
The announcement came as Gates took aim at another billionaire tech titan, Elon Musk.
The Tesla CEO pushed through draconian cuts to the US Agency for International Development because Musk "didn't go to a party that weekend," Gates told the New York Times in an apparent dig at Musk's lifestyle.
Gates is listed as the 13th on the Forbes "real-time" billionaire list, with a net worth of $112.6 billion. Musk is first with $383.2 billion.
Gates, 69, published a chart showing his net worth plummeting 99 percent over the next 20 years in a blog post announcing the shift, describing a doubling of the pace of giving.
"People will say a lot of things about me when I die, but I am determined that 'he died rich' will not be one of them," Gates wrote.
The Bill and Melinda Gates Foundation launched in 2000, the same year Bill Gates stepped down as CEO of Microsoft. In 2024, Melinda French Gates exited the foundation three years after the couple's divorce.
The organization, which had more than $71 billion in assets at the end of 2023, has been credited with helping to reshape the world of global public health.
It lists five offices throughout Africa, in addition to locations in the United States, Europe, China, India and the Middle East.
Gates cited progress in health efforts including campaigns to eradicate polio and the creation of a new vaccine for rotavirus that has helped reduce the number of children who die from diarrhea each year by 75 percent.
Separate from the Gates Foundation, the Microsoft founder said he plans to continue to provide funding for initiatives to expand access to affordable energy and for breakthrough research into Alzheimer’s disease.

Not a 'forever' foundation

In the blog post, Gates credited the writings of 19th-century US steel magnate Andrew Carnegie, whose foundation is still around.
But Gates told the New York Times he had no designs on creating a "forever" foundation out of "some weird legacy thing," preferring to pump out billions more to take advantage of emerging technologies.
"The tools are so phenomenal," he said of the potential for AI in global health.
"All the intelligence will be in the AI, and so you will have a personal doctor that's as good as somebody who has a full-time dedicated doctor -- that’s actually better than even what rich countries have," Gates told the New York Times.
While private foundations can do a lot, Gates described the government role as essential, ruing deep budget cuts by the United States, Britain, France and other countries.
"It's unclear whether the world’s richest countries will continue to stand up for its poorest people. But the one thing we can guarantee is that, in all of our work, the Gates Foundation will support efforts to help people and countries pull themselves out of poverty," he wrote. 
The moves have included the assault on USAID by Musk's "Department of Government Efficiency" in Donald Trump's presidential administration. 
Gates called the cuts "stunning," far more severe than expected.
Musk is "the one who cut the USAID budget," Gates told the New York Times. "He put it in the wood chipper."
In an interview with the Financial Times, Gates ridiculed Musk's apparent confusion of Gaza Province in Mozambique with Gaza in the Middle East as the Trump administration targeted programs.
"The picture of the world's richest man killing the world's poorest children is not a pretty one," Gates said of Musk in an interview with the Financial Times.
jmb/bgs

company

Global cult following keeps Le Creuset simmering

BY ETIENNE BALMER

  • - Social media success - Le Creuset has 575 retail outlets in the world, with online sales having received a boost from a home cooking craze during the Covid pandemic.
  • They feature in the Duchess of Sussex's "With Love, Meghan" Netflix cooking show.
  • - Social media success - Le Creuset has 575 retail outlets in the world, with online sales having received a boost from a home cooking craze during the Covid pandemic.
They feature in the Duchess of Sussex's "With Love, Meghan" Netflix cooking show. They've been spotted in rapper Snoop Dogg's kitchen. Top chefs can't live without them, and TikTok posts draw millions.
For a maker of pots and pans, Le Creuset has had an astonishing global run and cult-like following that nobody could have predicted when the company first set out to produce staple kitchenware in Fresnoy-le-Grand, a modest village in northern France, in 1925.
Two Belgian entrepreneurs built what, a century later, is still Le Creuset's home factory in the village of barely 3,000 inhabitants, home of the company's trademark enamelled cast-iron cookware.
The flagship Dutch oven model, now available in about 100 colours, started out exclusively in flaming orange, which still makes Le Creuset pots instantly recognisable.
With a price tag in the region of 250 euros ($280) for basic cast-iron models -- rising fast for elaborate models or special editions -- Le Creusets are high-end designer creations with a reputation for indestructibility.

'Crucible'

All the company's cast-iron cookware is still exclusively made in the Fresnoy-le-Grand factory, the centrepieces of which are two giant electric furnaces -- also called "creuset", which is French for "crucible".
The furnaces heat molten cast iron to 1,550 degrees Celsius (2,822 Fahrenheit), the melting point for this iron and carbon alloy.
The blindingly bright liquid, hotter than lava, is then poured into a transfer recipient, which is automatically carried along a rail.
The cast iron is poured quickly into sand moulds shaped by metal patterns to make raw products. The remaining cast iron and sand are recycled back into the manufacturing process.
After being ground by robots and stripped by being exposed to bombardment with tiny steel beads, the utensils are glazed with enamel -- a mixture of glass, quartz, clay, water and colorants -- before vitrification at nearly 800C.
The resulting variety of shapes and colours presents an industrial challenge, but "really embodies the strength and DNA of the brand," said Frederic Salle, manager of the site.
Le Creuset now sells 95 percent of its production abroad, in more than 80 countries, but keeps a tight lid on financial data, which the privately held company is not obliged to disclose.
Things weren't always upbeat. When Paul van Zuydam, a Briton with a South African background, bought Le Creuset in 1988, customers had gone cool on the brand.
But Van Zuydam, who is still Le Creuset's president, pushed the company's international expansion, established it at the high segment of the market and diversified production sites for non-cast iron products to foreign countries, including China and Thailand.

Social media success

Le Creuset has 575 retail outlets in the world, with online sales having received a boost from a home cooking craze during the Covid pandemic.
"The brand is doing very well pretty much everywhere in the world," said Marie Gigot, managing director for France, Belgium, the Netherlands and Luxembourg.
Like for many global companies, US President Donald Trump's tariff threats are a concern, she acknowledged. "The situation changes every day, so we follow it very closely."
But US buyers wealthy enough to purchase Le Creuset products in the past will probably not be put off by any tariff hikes, said Nick Stene, head of home and garden research at Euromonitor, a market research company.
"Homes that can afford to invest in the higher price points, especially over $300 for luxury cookware, are the last households to feel the pain when buying power is under pressure," he told AFP.
Le Creuset has been "one of the strongest performers" in the homeware category, which has seen around 4.5-percent annual growth since 2019, he said.
One major factor of success has been social media, where proud owners like to showcase their Le Creuset to prove they can afford it, but also that they "know how to use it properly", accompanied by hashtags like #LeCreuSlay, he said.
"There is nothing quite as efficient as having your customers also act as your ambassadors and marketing team," added Stene.
etb/jh/as/lth/jhb

politics

Digital voting breeds distrust among overseas Filipino workers

BY PURPLE ROMERO

  • The fiasco has also left election watchdogs and migrant groups sceptical that the switch to online voting will boost turnout as intended.
  • A new online voting system aimed at boosting turnout among the Philippines' millions of overseas workers ahead of Monday's mid-term elections has been marked by confusion and fears of disenfranchisement.
  • The fiasco has also left election watchdogs and migrant groups sceptical that the switch to online voting will boost turnout as intended.
A new online voting system aimed at boosting turnout among the Philippines' millions of overseas workers ahead of Monday's mid-term elections has been marked by confusion and fears of disenfranchisement.
Thousands of overseas Filipino workers, or OFWs, have already cast their ballots in the race dominated by a bitter feud between President Ferdinand Marcos and his impeached vice president Sara Duterte.
While official turnout figures are not yet publicly available, data from the Commission on Elections (Comelec) show at least 134,000 of the 1.22 million registered overseas voters have signed up for the new online system, which opened April 13.
But Jun Burlasa III, a Filipino working in Singapore, says he will not vote again if he has to do it online. 
"I'd rather do manual," the 50-year-old told AFP this week, describing the new system as "confusing and suspicious".
At issue is a digital QR code generated after voting that leads users to a page asking them to verify their ballot has been submitted correctly. Below that is a box containing a jumble of computer code and candidate names.
Burlasa said many of the names visible were candidates for whom he had not voted.
Similar stories about the anxiety-inducing webpage have proliferated across social media, including Facebook posts that have reached thousands.
Eman Villanueva, a Hong Kong-based activist with migrant rights group BAYAN, said he was unsure his vote had been properly counted. 
"There is absolutely no way for the voters to know if the votes that went through really reflected our choices," he said. 
In previous overseas elections, voters could review the names they selected after the fact, but Comelec told AFP the QR code was never supposed to serve that purpose.
The landing page was only intended to verify a ballot's receipt, the commission said, adding that the name of every candidate running in the election should appear.
"We are definitely considering the feedback and studying how to incorporate them in future elections," Ian Geonanga, Comelec's director of overseas voting, told AFP.
Election watchdogs, however, say the commission failed to properly explain the new system and warn of the confusion risks disenfranchising voters. 
"It's a natural reaction of people that if you're not familiar with the system, then you won't trust it the first instance," said Ona Caritos, executive director of the nonprofit Legal Network for Truthful Elections (Lente). 

ChatGPT, disinfo and 2028

Since April 14, 1.5 million people have watched a video in which a Philippines-based engineer named Jaydee San Juan quizzes ChatGPT about the names visible on the ballot verification page.
"It's highly likely showing the candidates that were selected/voted for using that ballot ID," the AI chatbot replied.
Comelec, however, got the opposite answer when conducting the ChatGPT experiment itself, Geonanga told AFP.
The election commission's efforts to quell fears about the new system, meanwhile, have been misrepresented to sow more disinformation.
AFP fact-checkers recently debunked a video edited to make it appear Geonanga was saying online ballots were "designed" to rig the election's results.
The fiasco has also left election watchdogs and migrant groups sceptical that the switch to online voting will boost turnout as intended.
Danilo Arao, convenor of voting watchdog Kontra Daya, said even a small change to the ballot's design might have helped assuage fears he believes could lead to "widespread disenfranchisement".
Lente's Caritos said losing trust in the online voting system could impact OFWs' participation in the 2028 presidential election.
“We don't want that, because if election results are not trusted by our voters, then it would go into the legitimacy of the government," she said. "It's a domino effect."
pr-jan/cwl/df/dhw

technology

Meta blocks access to Muslim news page in India

BY ANUJ CHOPRA

  • The @Muslim account is among the most followed Muslim news sources on Instagram.
  • Meta has banned a prominent Muslim news page on Instagram in India at the government's request, the account's founder said Wednesday, denouncing the move as "censorship" as hostilities escalate between India and Pakistan.
  • The @Muslim account is among the most followed Muslim news sources on Instagram.
Meta has banned a prominent Muslim news page on Instagram in India at the government's request, the account's founder said Wednesday, denouncing the move as "censorship" as hostilities escalate between India and Pakistan.
Instagram users in India trying to access posts from the handle @Muslim -- a page with 6.7 million followers -- were met with a message stating: "Account not available in India. This is because we complied with a legal request to restrict this content."
There was no immediate reaction from the Indian government on the ban, which comes after access was blocked to the social media accounts of Pakistani actors and cricketers.
"I received hundreds of messages, emails and comments from our followers in India, that they cannot access our account," Ameer Al-Khatahtbeh, the news account's founder and editor-in-chief, said in a statement. 
"Meta has blocked the @Muslim account by legal request of the Indian government. This is censorship."
Meta declined to comment. A spokesman for the tech giant directed AFP to a company webpage outlining its policy for restricting content when governments believe material on its platforms goes "against local law."
The development, first reported by the US tech journalist Taylor Lorenz' outlet User Magazine, comes in the wake of the worst violence between nuclear-armed India and Pakistan in two decades.
Both countries have exchanged heavy artillery fire along their contested frontier, after New Delhi launched deadly missile strikes on its arch-rival.
At least 43 deaths were reported in the fighting, which came two weeks after New Delhi blamed Islamabad for backing a deadly attack on tourists in the Indian-run side of the disputed Muslim-majority region of Kashmir.
Pakistan rejects the charge and has warned it will "avenge" those killed by Indian air strikes.
The @Muslim account is among the most followed Muslim news sources on Instagram. Khatahtbeh apologized to followers in India, adding: "When platforms and countries try to silence media, it tells us that we are doing our job in holding those in power accountable."
"We will continue to document the truth and stand out firmly for justice," he added, while calling on Meta to reinstate the account in India.
India has also banned more than a dozen Pakistani YouTube channels for allegedly spreading "provocative" content, including Pakistani news outlets.
In recent days, access to the Instagram account of Pakistan's former prime minister and cricket captain Imran Khan has also been blocked in India.
Pakistani Bollywood movie regulars Fawad Khan and Atif Aslam were also off limits in India, as well as a wide range of cricketers -- including star batters Babar Azam and Mohammad Rizwan and retired players Shahid Afridi and Wasim Akram.
Rising hostilities between the South Asian neighbors have also unleashed an avalanche of online misinformation, with social media users circulating everything from deepfake videos to outdated images from unrelated conflicts, falsely linking them to the Indian strikes.
On Wednesday, US President Donald Trump called for India and Pakistan to immediately halt their fighting, and offered to help end the violence.
bur-ac/sla

tech

Web archivists scrambling to save US public data from deletion

  • EDGI, a consortium of academics and volunteers, began safeguarding public climate and environmental data after Trump's first election in 2016.
  • As President Donald Trump's administration purges public records since storming back to power, experts and volunteers are preserving thousands of web pages and government sites devoted to climate change, health or LGBTQ rights and other issues.
  • EDGI, a consortium of academics and volunteers, began safeguarding public climate and environmental data after Trump's first election in 2016.
As President Donald Trump's administration purges public records since storming back to power, experts and volunteers are preserving thousands of web pages and government sites devoted to climate change, health or LGBTQ rights and other issues.
Resources on AIDS prevention and care, weather records, references to ethnic or gender minorities: numerous databases were destroyed or modified after Trump signed an executive order in January declaring diversity, equality and inclusion programs and policies within the federal governmentto be illegal.
More than 3,000 pages from the Centers for Disease Control and Prevention site were taken down and more than 1,000 from the Justice Department's website, Paul Schroeder, president of the Council of Professional Associations on Federal Statistics, told AFP.

404 error

Some websites have disappeared altogether, such as that of the US development agency USAID, which has been effectively shuttered as Trump slashes US aid to poor countries. 
And the National Children's Health Survey page displays a "404 error" message.
Federal agencies must now avoid hundreds of words such as "woman," "disability," "racism", "climate crisis" and "pollution" in their communications, the New York Times reported.
"The focus has been on removing language related to environmental (or) climate justice on websites, as well as removing data and tools related to environmental (or) climate justice," Eric Nost, a geographer at Canada's University of Guelph and member of the Environmental Data and Governance Initiative (EDGI) told AFP. 
"This Trump administration moved more quickly and with a greater scope than the previous Trump administration," he said.
EDGI, a consortium of academics and volunteers, began safeguarding public climate and environmental data after Trump's first election in 2016.
Among the tools used are the WayBack Machine from the non-profit Internet Archive, or Perma.cc, developed by the Library Innovation Lab at Harvard Law School.
These systems, which long predate Trump's election, help "courts and law journals preserve the web pages they cite to," said Jack Cushman, director of the Library Innovation Lab.
Long used by journalists, researchers and NGOs, web archiving enables a page to be preserved, even if it were to disappear from the internet or be modified later.
This data is then stored on servers in a large digital library, allowing anyone to consult it freely.

Volunteer work

Archiving initiatives have multiplied, expanded and coordinated since Trump's return to the White House.
The Data Rescue Project (DRP) brought together several organizations to save as much data as possible.
"We were concerned about data being deleted. We wanted to try to see what we could do to rescue them," Lynda Kellam, a university librarian and DRP organizer, told AFP.
She first launched the project as an online Google doc in February -- a simple word-processing tool listing downloaded PDF files, original dataset titles and archived links.
It is now maintained by volunteers "who are working after work" to keep it running, said Kellam.
"We are all volunteers, even myself. We have other jobs so that has been challenging," Kellam added.
The data collection work, largely carried out by associations and university libraries, is threatened by a lack of resources.
"Funding is the key issue... as the library and archives community rushes to take on a larger preservation challenges than ever before," Cushman said.
"We need to fund coordinators for the ongoing effort, new tools, and new homes for the data."
Harvard is also battling the ire of the Trump administration, which has cut federal grants to the prestigious university and threatened its tax-exempt status after it refused to comply with the president's demands to accept government oversight.
"Data is the modern lighthouse, helping us plan our lives: it shows where we are so we can plan where we're going," Cushman said.
"Businesses, individuals, and governments will suffer greatly from any failure to collect and share reliable data on weather and climate, health, justice, housing, employment, and so on."
ecb/tq/dhw/dw

AI

Google shares plunge after Apple executive's court testimony

BY ALEX PIGMAN

  • Investors were further unsettled when Cue suggested Apple might soon offer AI alternatives as default search options on its devices, heightening concerns that Google's advertising revenue could face serious threats from AI competitors.
  • Shares in Google parent Alphabet plunged more than eight percent on Wednesday after Apple executive Eddy Cue testified in federal court that Google's search traffic on Apple devices declined last month for the first time in over two decades.
  • Investors were further unsettled when Cue suggested Apple might soon offer AI alternatives as default search options on its devices, heightening concerns that Google's advertising revenue could face serious threats from AI competitors.
Shares in Google parent Alphabet plunged more than eight percent on Wednesday after Apple executive Eddy Cue testified in federal court that Google's search traffic on Apple devices declined last month for the first time in over two decades.
Cue, Apple's senior vice president of services, told the Washington antitrust trial that Google was losing ground to AI alternatives like ChatGPT and Perplexity.
His revelation that this decline "has never happened in 22 years" sent shockwaves through Wall Street, wiping more than $170 billion from Google's market capitalization in a single trading session.
The testimony came during a pivotal trial where District Judge Amit Mehta will determine remedies for Google's previously ruled illegal search monopoly. 
The case, ongoing since 2020, has exposed Google's practice of paying Apple tens of billions dollars annually to remain the default search engine on Safari browsers and Apple smartphones.
Investors were further unsettled when Cue suggested Apple might soon offer AI alternatives as default search options on its devices, heightening concerns that Google's advertising revenue could face serious threats from AI competitors.
With the three-week trial set to conclude Friday, government attorneys are pushing Judge Mehta to order Google to divest its Chrome browser.
They argue that AI technologies will only strengthen Google's dominance by leveraging its vast data resources across products like Maps, YouTube, and Chrome to stifle competition.
However, Cue's testimony bolstered Google's defense that AI is already disrupting its search dominance, with chatbots now posing legitimate threats to its business model.

'Losing sleep'

When Judge Mehta issues his ruling in August, he could end Google's default search agreements with Apple and others -- a prospect that Cue told the court he was "losing sleep" over, with potential revenue losses impacting Apple's product development and operating system investment.
Alternatively, Mehta might order Google to share its search data with competitors, which CEO Sundar Pichai warned would effectively amount to a "de facto divestiture of search."
As a counter offer, Google proposes a more limited remedy that would allow it to continue paying for default placement of its search engine, but with an annual renegotiations and greater freedom for smartphone manufacturers to choose which Google apps to install on their devices.
The Google case represents just one of five major tech antitrust actions currently pursued by the US government, with Meta facing similar scrutiny in the same courthouse.
Google recently lost a separate case regarding its ad technology business and may face additional divestitures, while Apple and Amazon are also expected to confront antitrust challenges in US courts.
arp/jbr 

technology

OpenAI offers to help countries build AI systems

  • The OpenAI for Countries initiative was launched under the auspices of a Stargate drive announced by US President Donald Trump to invest up to $500 billion in AI infrastructure in the United States.
  • OpenAI on Wednesday announced an initiative to help countries build their own artificial intelligence (AI) infrastructures, with the US government a partner in projects.
  • The OpenAI for Countries initiative was launched under the auspices of a Stargate drive announced by US President Donald Trump to invest up to $500 billion in AI infrastructure in the United States.
OpenAI on Wednesday announced an initiative to help countries build their own artificial intelligence (AI) infrastructures, with the US government a partner in projects.
The San Francisco tech firm's move to put its technology at the heart of national AI platforms around the world comes as it faces competition from Chinese rival DeepSeek.
DeepSeek's success in delivering powerful AI models at a lower cost has rattled Silicon Valley and multiplied calls for US big tech to protect its dominance of the emerging technology.
"It's clear to everyone now that this kind of infrastructure is going to be the backbone of future economic growth and national development," OpenAI said in a blog post.
"This is a moment when we need to act to support countries around the world that would prefer to build on democratic AI rails, and provide a clear alternative to authoritarian versions of AI that would deploy it to consolidate power."
The OpenAI for Countries initiative was launched under the auspices of a Stargate drive announced by US President Donald Trump to invest up to $500 billion in AI infrastructure in the United States.
"We've heard from many countries asking for help in building out similar AI infrastructure," OpenAI said.
"In response to these interested governments, OpenAI is offering a new kind of partnership for the Intelligence Age."
OpenAI, in "coordination" with the US government, will help countries build datacenters and provide customized versions of its ChatGPT AI tailored for local languages and cultures to improve healthcare, education and public services, according to the tech firm.
Projects are to involve "local as well as OpenAI capital".
Partner countries would invest in the broader Stargate Project to expand "US-led AI leadership," OpenAI said.
gc/arp

AI

Alphabet's share price plunges on traffic drop testimony

  • Cue told the court this would have a significant impact on Apple's ability to invest in new products and services. arp/dw
  • Shares in Google parent Alphabet plunged more than seven percent on Wednesday after an Apple executive told a federal court that the search engine's traffic fell on Apple products last month.
  • Cue told the court this would have a significant impact on Apple's ability to invest in new products and services. arp/dw
Shares in Google parent Alphabet plunged more than seven percent on Wednesday after an Apple executive told a federal court that the search engine's traffic fell on Apple products last month.
Eddy Cue, Apple's senior vice president of services, testified at an antitrust trial in Washington that Google search volume was losing traffic to AI alternatives such as ChatGPT or Perplexity, according to US media reports.
"That has never happened in 20 years," legal news outlet MLex quoted him as saying.
Cue was giving testimony in a trial in which US Judge Amit Mehta will determine how Google must address his landmark ruling last year that it operates an illegal monopoly in online search.
The Apple executive's remarks saw Google's market capitalization wiped of $140 billion since the close of trading on Wall Street on Tuesday.
The marathon court case has revealed that Google pays Apple tens of billions of dollars every year in a revenue sharing agreement in which Google's search engine is set as the default on Apple's Safari browser.
Markets were also rattled by Cue's comment that "over the coming year we will add other (AI) choices to the search engine choice in the browser, because I think those products are getting better and better," he said, according to MLex.
The testimony backed Google's argument that the emergence of AI has begun a new era in how people get information online, with its search engine now facing new rivalry from AI chatbots.
US government attorneys have urged Judge Mehta to force Google to sell off its Chrome browser, arguing that artificial intelligence will actually only ramp up the tech giant's online search dominance.
Another option is that the judge, in a decision expected in August, will order an end to the payouts from Google to Apple and others for the default position on devices.
Cue told the court this would have a significant impact on Apple's ability to invest in new products and services.
arp/dw

internet

US jury awards WhatsApp $168 mn in NSO Group cyberespionage suit

  • A jury on Tuesday found that NSO should pay WhatsApp $444,719 in compensatory damages and another $167,254,000 in punitive damages intended to discourage repeating the behavior that landed it in court.
  • A US jury on Tuesday handed WhatsApp a major victory in its cyberespionage suit against NSO Group, ordering the Israel-based firm to pay some $168 million in damages.
  • A jury on Tuesday found that NSO should pay WhatsApp $444,719 in compensatory damages and another $167,254,000 in punitive damages intended to discourage repeating the behavior that landed it in court.
A US jury on Tuesday handed WhatsApp a major victory in its cyberespionage suit against NSO Group, ordering the Israel-based firm to pay some $168 million in damages.
Meta-owned WhatsApp sued NSO in late 2019 in federal court in Northern California, accusing it of planting Pegasus spy software on the smartphones of targets using the messaging app.
"This trial put spyware executives on the stand and exposed exactly how their surveillance-for-hire system –- shrouded in so much secrecy –- operates," Meta said in a blog post.
"Put simply, NSO's Pegasus works to covertly compromise people's phones with spyware capable of hoovering up information from any app installed on the device."
Pegasus software also enables smartphone cameras or microphones to be remotely turned on without letting users know, according to Meta.
WhatsApp accused NSO of cyberespionage targeting journalists, lawyers, human rights activists and others on the Facebook-owned messaging service.
A jury on Tuesday found that NSO should pay WhatsApp $444,719 in compensatory damages and another $167,254,000 in punitive damages intended to discourage repeating the behavior that landed it in court.
"We will carefully examine the verdict's details and pursue appropriate legal remedies, including further proceedings and an appeal," NSO vice president for global communication Gil Lainer said in response to an AFP inquiry.
"We firmly believe that our technology plays a critical role in preventing serious crime and terrorism and is deployed responsibly by authorized government agencies."
Evidence presented at the trial said NSO had spyware installation methods to exploit the technology of companies other than Meta, spending tens of millions of dollars annually on ways to install malicious code through messaging, browsers and operating systems, according to Meta.
In 2016, Apple rushed out a security update after researchers said prominent Emirati rights activist Ahmed Mansoor was targeted by UAE authorities using Pegasus spyware.
The software has been pinpointed by independent experts as likely being used in a number of countries with poor human rights records.
"Given how much information people access on their devices, including through private end–to-end encrypted apps like WhatsApp, Signal and others, we will continue going after spyware vendors indiscriminately targeting people around the world," Meta said in the blog post.
"These malicious technologies are a threat to the entire ecosystem and it'll take all of us to defend against it."
The legal complaint said the attackers "reverse-engineered the WhatsApp app and developed a program to enable them to emulate legitimate WhatsApp network traffic in order to transmit malicious code" to take over the devices.
Infecting smartphones or other gadgets being used for WhatsApp messages meant the content of messages encrypted during transmission could be accessed after they were unscrambled for recipients.
Founded in 2010 by Israelis Shalev Hulio and Omri Lavie, NSO Group is based in the Israeli seaside hi-tech hub of Herzliya, near Tel Aviv.
gc/dw

merger

Food delivery app DoorDash agrees to buy peer Deliveroo

BY ALEXANDRA BACON

  • - European expansion -  DoorDash, the largest food delivery app in the United States, entered the European market in 2021 with the purchase of Finland-based Wolt for $8.1 billion. 
  • US food delivery app DoorDash has agreed to buy UK-based rival Deliveroo for £2.9 billion ($3.9 billion) in a deal that will expand its global reach.
  • - European expansion -  DoorDash, the largest food delivery app in the United States, entered the European market in 2021 with the purchase of Finland-based Wolt for $8.1 billion. 
US food delivery app DoorDash has agreed to buy UK-based rival Deliveroo for £2.9 billion ($3.9 billion) in a deal that will expand its global reach.
The deal, announced by the two companies Tuesday, will create a delivery service present in more than 40 countries, serving around 50 million monthly-active users.
The combined group "will bring together DoorDash's strong operating playbook with Deliveroo's local expertise to invest in innovation and execution at an even higher level", DoorDash chief executive Tony Xu said in a statement.
London-listed Deliveroo posted its first annual profit in March following sizeable full-year losses owing to high investment costs since American Will Shu founded the company in 2013.
The company's initial public offering in 2021 had been London's biggest stock market launch for a decade, valuing the group at £7.6 billion.
The offer from DoorDash is worth £1.80, less than half Deliveroo's IPO price of £3.90.
Shares in Deliveroo rose two percent to £1.75 Tuesday on London's second-tier FTSE 250 index, after already jumping in response to news of the takeover proposal last week.
Deliveroo experienced a surge in demand during the Covid-19 pandemic from lockdown-hit customers but increased competition led it to scale back global operations.
Most recently, it exited Hong Kong amid growing competition in the Chinese city, following its exit from Australia and the Netherlands. 
The deal announced Tuesday is expected to be completed in the last three months of 2025, subject to regulatory approval and the approval of Deliveroo shareholders.
- European expansion - 
DoorDash, the largest food delivery app in the United States, entered the European market in 2021 with the purchase of Finland-based Wolt for $8.1 billion. 
It is now looking to further expand its reach, with Deliveroo operating in the UK, Belgium, France, Ireland, Italy, Kuwait, Qatar, Singapore and the United Arab Emirates.
For Deliveroo, it marks "the beginning of a transformative new chapter", said Shu, also the company's chief executive.
It is the latest deal in the food delivery market, after Dutch investment group Prosus announced plans in February to buy Just Eat Takeaway.com for 4.1 billion euros.
San Francisco-based DoorDash said it has no plans to relocate Deliveroo's London headquarters. 
It added that it does not anticipate making any changes that would impact the contracts of delivery drivers, or "riders".
As big players in the gig economy, food delivery apps have faced controversy over the status of their self-employed riders. 
In late 2023, the UK Supreme Court ruled that Deliveroo riders were not entitled to trade union rights such as collective bargaining.
Deliveroo exited Spain after it became the first European Union nation to give food delivery riders labour rights, requiring that they be recognised as employees instead of being considered self-employed freelancers.
DoorDash in February agreed to pay out nearly $17 million to drivers in New York state, who accused the company of swindling them out of tip money.
ajb/bcp/lth

pope

Wait for Vatican white smoke fires up social media

BY DAXIA ROJAS

  • "Everyone on social networks has an opinion and everyone wants to decode the news, look for clues, know which cardinal will become the pope.
  • Hype has been building on social media around the Catholic Church's secretive, centuries-old tradition of conclaves to elect a new pope, animating users from the White House on down.
  • "Everyone on social networks has an opinion and everyone wants to decode the news, look for clues, know which cardinal will become the pope.
Hype has been building on social media around the Catholic Church's secretive, centuries-old tradition of conclaves to elect a new pope, animating users from the White House on down.
US President Donald Trump on Saturday posted an apparently AI-generated image of himself wearing papal vestments and sitting on a throne, one finger directed to the heavens.
The striking picture was the most notorious among thousands that have bubbled up since the death of Pope Francis on April 21 and ahead of the cardinals' gathering from Wednesday.
More than 1.3 million tweets have been published on X about the conclave, according to monitoring platform Visibrain, while TikTok videos on the topic have been viewed over 363 million times on the network with unparallelled reach among the young.
Particularly passionate pope-watchers can fire up online game "Fantapapa" to pick their favourite cardinals and make predictions for the next pontiff in a style similar to sports betting.

Pomp and secrecy

The mystery, pomp and ritual around the conclave -- from the opulent Sistine Chapel surroundings to the ethereal black or white smoke signalling ballot results -- "lends itself to the narrative formats of social networks" said Refka Payssan, a researcher in information and communication sciences.
"A conclave means both gilt, protocol, ceremony, but also secrecy and mystery" cannily nurtured by the Vatican, agreed Stephanie Laporte, founder of digital strategy consultancy OTTA.
"Young people love to speculate" about outcomes, Laporte added.
"Everyone on social networks has an opinion and everyone wants to decode the news, look for clues, know which cardinal will become the pope. It's almost like an 'escape game'," she suggested.
Payssan noted that the conclave fires up the "curiosity of seeing history happen live", marking a rare event -- the first in 12 years -- with potential global consequences.
Even if not Catholic themselves, "young people are very conscious of the pope's influence on hundreds of millions, even billions of people, whether it's in his stance on contraception or the environment," Laporte said.

Digital turn

Conclave fever is also a reflection of the Vatican's successful turn to digital communications in recent years to build bonds with younger generations.
Created by Benedict XVI in 2012 but mostly used by Francis, the papal X account @pontifex reaches 50 million followers across its nine languages.
And Francis's own Instagram account had more than 10 million followers.
The Church has backed many cardinals' own ventures into the digital realm, with some becoming bona fide internet stars.
New York prelate Timothy Dolan has been publishing videos about the run-up to the conclave to his almost 300,000 X followers and 55,000 on Instagram -- without giving away any sensitive information.
Moderate Philippine cardinal Luis Antonio Tagle has made his mark online with karaoke videos, tallying 600,000 Facebook followers.
Selfie snapshots are in the mix, with Tokyo's archbishop Tarcisio Isao Kikuchi posting a photo with colleagues from the bus on the way to pray at Francis's grave.
Cardinals "are absolutely fascinating personalities who've taken their place in pop culture," firing public enthusiasm for the event, Laporte said.
That fascination has been stoked by pop culture blockbusters like Dan Brown's novel "Angels and Demons", adapted for film in 2009, or the acclaimed thriller "Conclave" released this year, based on a book by novelist Robert Harris.
dax/tgb/ach/tym 

internet

New Zealand PM proposes banning under-16s from social media

  • "This is about protecting our children.
  • New Zealand's prime minister on Tuesday proposed banning children under 16 from social media, stressing the need to protect them from the perils of big tech platforms.
  • "This is about protecting our children.
New Zealand's prime minister on Tuesday proposed banning children under 16 from social media, stressing the need to protect them from the perils of big tech platforms.
Regulators the world over are wrestling with how to keep children safe online, as social media is increasingly flooded with violent and disturbing content.
Prime Minister Christopher Luxon unveiled draft laws that would force social media companies to verify users were at least 16 years old, or face fines of up to NZ$2 million (US$1.2 million). 
The proposed ban was modelled on that of Australia, which sits at the forefront of global efforts to regulate social media. 
"This is about protecting our children. It's about making sure social media companies are playing their role in keeping our kids safe," Luxon said.
It was not clear when the legislation would be introduced to parliament, but Luxon said he was hopeful of garnering support across the chamber.
The laws were drafted by Luxon's centre-right National Party, the biggest member in New Zealand's three-way governing coalition. 
To be passed they would need the support of Luxon's two other coalition partners. 
"Parents are constantly telling us that they are really worried about the impact that social media is having on their children," Luxon said. 
"And they say they are really struggling to manage access to social media."
Australia passed landmark laws in November banning under-16s from social media -- one of the world's toughest crackdowns on popular sites such as Facebook, Instagram and X.
The move sparked a fierce backlash from big tech companies who variously described the laws as "rushed", "vague", and "problematic". 
sft/lec/rsc

tech

OpenAI abandons plan to become for-profit company

  • "OpenAI is not a normal company and never will be," Altman wrote in an email to staff posted on the company's website.
  • OpenAI CEO Sam Altman announced Monday that the company behind ChatGPT will continue to be run as a nonprofit, abandoning a contested plan to convert into a for-profit organization.
  • "OpenAI is not a normal company and never will be," Altman wrote in an email to staff posted on the company's website.
OpenAI CEO Sam Altman announced Monday that the company behind ChatGPT will continue to be run as a nonprofit, abandoning a contested plan to convert into a for-profit organization.
The structural issue had become a significant point of contention for the artificial intelligence (AI) pioneer, with major investors pushing for the change to better secure their returns.
AI safety advocates had expressed concerns about pursuing substantial profits from such powerful technology without the oversight of a nonprofit board of directors acting in society's interest rather than for shareholder profits.
"OpenAI is not a normal company and never will be," Altman wrote in an email to staff posted on the company's website.
"We made the decision for the nonprofit to stay in control after hearing from civic leaders and having discussions with the offices of the Attorneys General of California and Delaware," he added. 
OpenAI was founded as a nonprofit in 2015 and later created a "capped" for-profit entity allowing limited profit-making to attract investors, with cloud computing giant Microsoft becoming the largest early backer.
This arrangement nearly collapsed in 2023 when the board unexpectedly fired Altman. Staff revolted, leading to Altman's reinstatement while those responsible for his dismissal departed.
Alarmed by the instability, investors demanded OpenAI transition to a more traditional for-profit structure within two years.
Under its initial reform plan revealed last year, OpenAI would have become an outright for-profit public benefit corporation (PBC), reassuring investors considering the tens of billions of dollars necessary to fulfill the company's ambitions.
Any status change, however, requires approval from state governments in California and Delaware, where the company is headquartered and registered, respectively.
The plan faced strong criticism from AI safety activists and co-founder Elon Musk, who sued the company he left in 2018, claiming the proposal violated its founding philosophy.
In the revised plan, OpenAI's money-making arm will now be fully open to generate profits but, crucially, will remain under the nonprofit board's supervision.
"We believe this sets us up to continue to make rapid, safe progress and to put great AI in the hands of everyone," Altman said.

SoftBank sign-off

OpenAI's major investors will likely have a say in this proposal, with Japanese investment giant SoftBank having made the change to being a for-profit a condition for their massive $30 billion investment announced on March 31.
In an official document, SoftBank stated its total investment could be reduced to $20 billion if OpenAI does not restructure into a for-profit entity by year-end.
The substantial cash injections are needed to cover OpenAI's colossal computing requirements to build increasingly energy-intensive and complex AI models.
The company's original vision did not contemplate "the needs for hundreds of billions of dollars of compute to train models and serve users," Altman said.
SoftBank's contribution in March represented the majority of the $40 billion raised in a funding round that valued the ChatGPT maker at $300 billion, marking the largest capital-raising event ever for a startup.
The company, led by Altman, has become one of Silicon Valley's most successful startups, propelled to prominence in 2022 with the release of ChatGPT, its generative AI chatbot.
arp/aha

layoffs

Meta content moderator cuts over 2,000 jobs in Spain: union

BY VALENTIN BONTEMPS

  • Telus, which operates locally as Barcelona Digital Services, said during a Monday meeting that it had terminated the contracts "of all workers who were performing content moderation tasks" for Meta, affecting 2,059 people, union CCOO said in a statement.
  • Canadian tech firm Telus International, tasked by Meta to moderate content on its social media networks Facebook and Instagram, will slash over 2,000 jobs in Barcelona, a Spanish union said Monday.
  • Telus, which operates locally as Barcelona Digital Services, said during a Monday meeting that it had terminated the contracts "of all workers who were performing content moderation tasks" for Meta, affecting 2,059 people, union CCOO said in a statement.
Canadian tech firm Telus International, tasked by Meta to moderate content on its social media networks Facebook and Instagram, will slash over 2,000 jobs in Barcelona, a Spanish union said Monday.
The move comes as Mark Zuckerberg's company has cut its third-party fact-checking in the United States and overhauled its content moderation policies.
Telus, which operates locally as Barcelona Digital Services, said during a Monday meeting that it had terminated the contracts "of all workers who were performing content moderation tasks" for Meta, affecting 2,059 people, union CCOO said in a statement.
According to the union, the redundancy plan was decided after Meta cancelled its contract with Telus, which had provided content moderation for the tech giant since 2018.
CCOO added it had signed a preliminary agreement that will grant "the highest possible legal compensation" for the workers affected.
Contacted by AFP, a Telus spokesman refused to disclose how many jobs would be lost. 
"The priority remains to support the team members affected" by offering them "full assistance, including relocation opportunities for as many people as possible", the spokesman said.
UGT, another union that signed the deal, said the redundancies would be spread out during May, June, July and September.
"The moderation sector requires the professionalisation that these workers offered," it said in a statement.
Meta had not responded to a request for comment. In April, the California-based company said the end of its contract with Telus's Barcelona site would not mean a reduction in its content revision efforts.

'Politically biased'

Meta invested heavily and hired thousands of content moderators globally over the years to police sensitive content. It has also used third-party fact-checkers.
Roughly two weeks before President Donald Trump's January inauguration, Zuckerberg said his company would replace US-based fact-checkers with a system of community notes similar to what is used by X, owned by Trump ally Elon Musk.
Zuckerberg said "fact checkers have just been too politically biased and have destroyed more trust than they've created, especially in the US."
Meta also said it will stop proactively scanning for hate speech and other types of rule-breaking, reviewing such posts only in response to user reports.
The company's announcement echoed many of the complaints made by Republicans and Musk about fact-checking programmes that many conservatives see as censorship, a claim professional fact-checkers reject.
AFP is involved in a partnership with Meta providing fact-checking services in Asia-Pacific, Europe, the Middle East, Latin America and Africa.
vab/imm/ds/rl

economy

Facing a broken economy, Ghana's tech-savvy teens turn to fraud

BY WINIFRED LARTEY

  • As dusk settles over Nima, Ghost logs off, pockets his burner phone, and steps into the fading light. 
  • In the dusty alleys of Nima, a shanty town in the heart of Ghana's capital, a 17-year-old called Ghost reclines on a faded plastic chair inside a dimly lit internet cafe.
  • As dusk settles over Nima, Ghost logs off, pockets his burner phone, and steps into the fading light. 
In the dusty alleys of Nima, a shanty town in the heart of Ghana's capital, a 17-year-old called Ghost reclines on a faded plastic chair inside a dimly lit internet cafe.
Outside, barefoot children chase a punctured football. Inside, Ghost's fingers dance across the keyboard, his eyes locked on WhatsApp as he engineers a phishing scam that could earn him thousands of cedis in just a few hours.
"I made GHC12,000 ($770) last month," Ghost told AFP, his voice low and calculated, describing an online store he set up on Instagram. 
"People bought phones and laptops. None of it existed."
Ghost, a pseudonym he gave AFP, is one of a growing number of Accra's teenagers turning to cybercrime to survive in a country mired in economic crisis, battling both youth unemployment and broken dreams. 
From mobile money fraud to investment scams, a murky digital underworld is sucking in minors, many working from their bedrooms or small kiosks operating in plain sight -- anywhere with a stable internet connection.

Too good to be true

Nima’s 441 neighbourhood is a tightly packed, working-class community of corrugated iron housing and open drains. Opportunities are scarce -- but mobile phones are not. 
Ghana's Cyber Security Authority (CSA) has raised the alarm over a surge in cyberfraud, with financial losses tallying $282,776 between January and March 2025, nearly doubling the $154,241 recorded during the same period in 2024. 
Officials warn that youth-led scams, powered by social media and peer pressure, are driving the spike.
Phishing scams, brand impersonation and fake online shops dominate the scene, with teenagers posing on Snapchat and TikTok as vendors with offers that are too good to be true.
Mercy Adumoah, 20, was one such victim. 
"I saw a page on Snapchat selling heels. I needed a pair for an event, so I paid without thinking twice," she recounted.
After they received the money, the sellers blocked her account.
Experts say these crimes have become systemic in a country with a battered economy that is still recovering from a 2023 debt default. Inflation shot above 54 percent in 2022 and has remained above 20 percent into 2025.
At a junior high school near Nima, teacher Mohammed Inusah has witnessed a transformation.
"Some of my students have iPhones more expensive than my salary," he told AFP. "They flaunt cash, buy designer clothes and sneakers."
"The parents are either unaware or too afraid to confront them."

'I know it's wrong'

A soft-spoken boy with a mop of dreadlocks who gave the name Tricky said he got his start in scamming by copying scripts from online forums in Nigeria. 
Later, his cousin taught him mobile money fraud -- how to pose as an agent from the local telecom company to get access to people's accounts.
Tricky claims his biggest hit since in his two years of scamming was $500 -- twice the monthly salary of a public health nurse or a teacher. 
"I bought clothes, helped my mum pay rent," he said. "I know it's wrong, but tell me, what else can I do?"
The CSA has also flagged a dramatic rise in online investment fraud. Between January and August 2024 alone, 149 cases were recorded with losses nearing $128,534.
"I fell for one," admits 18-year-old "Bronzy", who went from victim to perpetrator. "A guy scammed me using a fake forex trading site. So I learnt the game."
He now runs a group on Telegram promising 20 percent weekly returns.
"People invest and I disappear," he said.
Abubakar Issaka, president of the Cyber Security Experts Association of Ghana, said the situation is only getting worse.
"The regulations exist... but enforcement is weak. The number of professionals is not growing fast enough to match the fraud cases," he said. 
Tracing perpetrators "is a challenge due to poor data integration" between telecoms operators and the national ID database, he added.
In some cases, the phone numbers used in frauds "belong to people who died years ago. Fraudsters are steps ahead."

No end in sight

Victims suffer anxiety and financial ruin. Scammers are not untouched. Ghost admits he's often afraid. 
"Sometimes, I can't sleep. I wonder if the police will knock. But when I see my friends living large, I feel like I must keep going."
As dusk settles over Nima, Ghost logs off, pockets his burner phone, and steps into the fading light. 
He didn't make any money that day. His world is one of quick, fickle wins and quiet fear. But with Ghana's economy showing no signs of rapid recovery, many more are likely to follow him into the shadows.
strs/nro/cw