Walmsley came out fighting, declaring herself a “change agent” and “business leader”. Forget her lack of scientific training. She’d spent “the last four years leading a transformational programme of change precisely to address some perennial underperformance”, including ramping up R&D spend “by 30 per cent”.

Contrary to impressions, GSK was, she said, on the cusp of a “step-change in growth”. And, giving herself nowhere to hide, she set out punchy targets: compound annual growth from 2021 to 2026 of at least 5 per cent in sales and 10 per cent in adjusted operating profits — all en route to £33 billion sales by 2031.

Her boldness could have easily backfired. And, yes, there is a caveat: GSK shares, up 2 per cent to £15.68 on the full-year results, still lag the £16.76 where she came in. But, on the ground, as the latest figures show, Walmsley is delivering.

Strip out the impact of Covid drugs and sales rose 14 per cent last year to £30.3 billion. At constant currencies, adjusted operating profits were up 12 per cent to £8.8 billion. To boot, she’s raised her targets. Now she’s shooting for more than 7 per cent annual sales growth and 11 per cent in profits out to 2026, while predicting that sales will top £38 billion by 2031. That’s even allowing for Dolutegravir, GSK’s market-leading HIV treatment, coming off-patent in 2028.

Why the extra confidence? First, from delivering ten straight quarters of growth. Second, thanks to a drug that demonstrates her “step-change” thesis. Back in 2021, Walmsley said she had 11 future blockbusters in development. One of them was Arexvy: a new jab for respiratory syncytial virus infection, a condition affecting tens of millions worldwide. Less than six months in existence, it’s delivered £1.24 billion of sales, taking two thirds of a new market, against a rival product from Pfizer.

Now, when Walmsley says GSK is “planning for at least 12 major launches from 2025”, each capable of £2 billion in peak sales, the market’s less likely to write it off as wishful thinking. Even Citi analyst Andrew Baum, a long-term GSK sceptic, has spotted grounds “to support a more positive thesis”.

So how come the shares have gone nowhere? Answer: US lawsuits over GSK’s historic heartburn drug Zantac. Shore Capital analyst Sean Conroy reckons that, in “a worst-case” scenario, “up to $30 billion for litigation is being reflected in the share price” — or a sum equating to 35 per cent of GSK’s £65 billion market value. But all this may soon prove overdone. A Delaware judge is due to rule this quarter whether GSK even has a case to answer over four fifths of claims. Could that be the moment Walmsley also delivers a step-change for the shares?

Tesla headache

Talking of Delaware judges, what about this one: Kathaleen McCormick, the beak who’s just nixed a £56 billion payday for Tesla’s Elon Musk?

She thinks it an “unfathomable sum”, even for a “paradigmatic ‘superstar CEO’ ” keen to use the loot to fund interplanetary travel. And, of course, she has a point. No one needs that sort of lucre, not least a bloke with a $210billion fortune. McCormick may be right, too, that Musk “enjoyed thick ties with the directors” setting his pay.

Yet a majority of Tesla investors approved 2018’s options package, figuring that Musk would probably be worth the money if he took the electric carmaker’s market value from $59 billion to $650 billion.In fact, he got it revved up beyond $1 trillion, at least for a bit before the shares came racing down again as Musk went around blowing $44 billion on Twitter/X, getting into rows over antisemitic tweets, telling advertisers to “go f*** yourself” and being accused of a penchant for ketamine. Yet even now Tesla has a market cap of $600 billion.

So what right has a judge to take his share award away, just because one shareholder, the former thrash metal drummer Richard Tornetta, didn’t like it? It does create a bit of a slippery slope.

Not only that: it gives the Tesla board an even bigger headache. Musk is already agitating for a 25 per cent voting stake, with the threat that unless he gets one, he’ll develop the AI gizmos on which Tesla’s value is based elsewhere. But if Musk can’t have his options, that’s an 8 per cent stake in Tesla gone, reducing his holding to 13 per cent. Does the board have a plan to make up the difference that won’t also get kiboshed by a Delaware court?

Musk is now polling his fan club on X over relocating the Tesla soap opera to Texas or Nevada. How long before he suggests Mars?

Nuclear non-event

So much for Rishi Sunak’s “nuclear roadmap”, calling for 24 gigawatts by 2050, or seven more Hinkley Point Cs. The French government is now demanding loan guarantees from the Brits before the state-backed EDF even finishes that £46 billion éléphant blanc.

Having let costs rip, the French also know it’s in their interests to string along the PM until the election in the hope of cutting a fresh financing deal with a Starmer government. So, one question: if Hinkley’s not sorted, how exactly can Sunak take his promised final investment decision on Sizewell C, the lookalike nuke also being built by EDF?

“It opened a million doors for me,” says Evans. “Without TikTok, being in the music industry would never have happened for me. It has given me so many opportunities. One video changed my life.”

Evans, 29, is a prime example of the enormous sway that the Chinese-owned social media app now has over the music industry. TikTok, with its almost two billion users scrolling through videos, has become the place for fledgling artists to showcase their talent and for established stars to push their songs to receptive audiences.

While Evans can toast his success, TikTok’s dominance over the music industry also raises serious questions. Despite its user base, the company pays a relative pittance in royalties to record labels and artists. Meanwhile, the very structure — and arguably the quality — of music made today is being torqued to suit the snappy virality of TikTok’s algorithm. Everything is changing at warp speed. Hits are getting shorter; nowhere is this more apparent than at this weekend’s Grammy awards, where one fifth of the nominated songs run for less than three minutes.

An industry fightback against TikTok is now under way and seems likely to be bitter. Evans, and many singers like him, are in limbo this weekend because his record label, Universal, is in dispute with the app over how much it should be paid for the use of its artists’ music.

Universal, which includes the likes of Taylor Swift and The Weeknd on its roster, accused TikTok of trying “to intimidate us into conceding to a bad deal that undervalues music and shortchanges artists and songwriters as well as their fans”. TikTok, which is owned by Beijing-based ByteDance, shot back and said Universal was “putting their own greed” above the interests of their artists.

The economics of TikTok are rather unusual. Unlike streaming services such as Spotify and the likes of Meta, the owner of Instagram and Facebook, TikTok does not pay labels — and, by extension, artists — for each use of a song on its platform. Instead, it has agreed deals with record companies under which, for a flat fee, its users can use as much music from artists as they like. Universal’s deal lapsed on January 31, and bosses of other labels are waiting to see who blinks first before deciding what they do next.

The current arrangement means that singers are not paid extra by TikTok when their songs go viral. For example, Kate Bush became a TikTok sensation in 2022 when her 1985 hit Running Up That Hill (A Deal With God) was used in the Netflix series Stranger Things. Although almost two million videos, viewed more than five billion times, using the song have since appeared on TikTok, she has not earned any extra money from the app.

TikTok paid a mere $220 million (£173 million) to the music industry for the use of songs in 2022, out of its $9.4 billion revenue. According to a Goldman Sachs report, this was less than Peloton, the exercise and media company, which paid $267 million in royalties, even though it only has seven million users.

Despite the relatively meagre sums on offer directly from TikTok, such is its viral potency that no aspiring musician can afford to ignore it. One music manager says that he has an offer from Universal for a young new artist but is leaning towards turning it down because it would be too difficult for her to get traction without being able to use the app. “I have to break this artist and if I can’t put her on TikTok, I can’t do it,” he says.

The app has even impacted artists and repertoire (A&R) teams, who are responsible for scouting and developing talent, at the biggest labels. Most of the large labels have dedicated teams to monitor TikTok and try to engineer a hit — often without success. “A&R meetings are just people sitting around looking at data points,” says one leading artist. “It is really, really bleak.”

The result of all this analysis is the changing shape and structure of songs themselves. Popular music is increasingly tailored to short social media attention spans. Many of the most popular songs on apps such as TikTok start with choruses or short introductions, while devices such as the middle eight and bridge — such as when the Beatles sing “Life is very short …” in We Can Work it Out — are waning.

This year is on course to have the shortest average song length in the top ten of the charts since 1966, according to a Sunday Times analysis, with the mean currently at 2 min 54 sec. At this weekend’s Grammys, about 20 per cent of nominated songs clock in at less than three minutes, according to research by The Washington Post — they include Midnight Rain by Taylor Swift and Nicki Minaj’s Barbie World.

Joe Bennett, a forensic musicologist and professor at Berklee College of Music in Boston, Massachusetts, says that songwriters are not consciously deciding to write shorter songs, but have stumbled upon it as a way to thrive.

“It’s more akin to a Darwinist ecosystem: the songs that become popular exhibit those characteristics,” he says. “Those are the songs we hear about and they are the songs that culturally propagate. Some songwriters might still be cheerfully writing longer songs, but we’re just not hearing them as much.”

The death of the middle eight, made famous by the likes of Bruce Springsteen’s Born to Run and Good Vibrations by the Beach Boys, is lamented by some. “It saddens me that pop is less harmonically interesting and the structures have become simpler with the near elimination of the middle eight,” says Tom Gray, singer of the indie band Gomez. “The medium has always dictated the art — if you don’t like the direction of the art, you do have to wonder about the medium.”

Music distribution technology has long affected what sort of songs artists produce. The original charts, in the 1930s, were based on sales of sheet music for pianos. In the 1950s, songs were mostly less than three minutes long because of the constraints imposed by vinyl technology.

The dawn of cassettes and, later, CDs gave artists in the 1970s the opportunity to be more experimental, with Donna Summer’s I Feel Love clocking in at 5 min 56 sec. The trend continued in the 1980s, and George Michael hit 6 min 30 sec when he released an extended mix of Careless Whisper in 1984. The trend for longer songs peaked in the mid-1990s, with Meat Loaf’s I’d Do Anything For Love (But I Won’t Do That) clocking in at 7 min 42 sec.

The rise of Napster and, later, Spotify caused a decline in song lengths that continues today, especially as artists only get royalty payments from streamers if listeners play songs for at least 30 seconds. The TikTok ascendancy has helped to drive lengths lower still: Tate McRae’s Greedy is in the Top 10 this year and is just 2 min 12 sec long.

TikTok virality is all very well when it works and can be translated into sustained success, but most music uploaded on the app goes unnoticed and most musicians in Britain still earn a pittance from their work. The Musicians’ Census, a survey of almost 6,000 artists in the UK last year, found that the average annual income of musicians is £20,700.

For the rest of us, the TikTokification of music makes it difficult to imagine mega-ballads such as Queen’s Bohemian Rhapsody or Guns N’ Roses’ Paradise City becoming definitive anthems today. When he teaches in Boston, Bennett says that protracted prog-rock is still alive among his students, but as more of a niche pursuit. Still, he remains optimistic. “One thing’s for sure: pop songs can’t keep getting shorter for ever,” he says. “This has got to bottom out at some point. When that happens, the only way is up.”

Only one in ten of the small and medium-sized businesses polled in Britain, Ireland, New Zealand, Australia and Canada regularly made use of artificial intelligence assistants such as ChatGPT and Gemini.

Of those that did, more said they worried about the quality of the responses and where the data that they entered was stored, when compared with the results of the same survey conducted by Peninsula, the HR and employment law specialist, last year.

Forty per cent of the businesses from the UK cited inaccuracies in the information provided as a key concern, up from 14 per cent last year. There were also increases in those limiting their use because of the risk of reputational damage and the risk of breaking laws, such as those governing data protection.

It has resulted in a fall of six percentage points in the number of businesses crediting generative AI assistants as having a transformational impact at work. Many more see it as a useful tool that needs to be used carefully and selectively. For the second year running, Canadian businesses were the most cautious, with only 23 per cent believing that generative AI would be transformative in the workplace.

Alan Price, Peninsula’s chief operations officer, said: “The legal risks around compliance, reputational damage and loss of intellectual property are believed to be of great concern for small businesses. While AI can certainly speed up processes, there is still a very long way to go before employers could rely on AI alone.”

Among the comments from companies taking part in the survey were those highlighting how generative AI services were being misused: “We currently receive many job applications with CVs written by AI — they are usually badly written, confused and often not representative of the applicant,” one business owner said.

A charity was concerned about where the data they entered might be stored. “[The] harvesting of sensitive information from emails and performance reports and its uncontrolled reuse is concerning for us as a charity whose reputation is based on confidentiality,” it said.

“So symbolic,” Sun says when we speak by video call. The astonishing problem was that the men, and they were all men, who made their money exploiting disasters in the real world were either too selfish or too busy — “There was that much time pressure,” she says — to clear up their own. Sun acknowledges her frustration that this woman had to be hired and that no one bothered to learn her name. This is the finance sector seen up close, at its crudest and most profound: “The rich not taking care of their own shit.”

On one level Sun’s memoir, Private Equity, does for finance what The Devil Wears Prada did for fashion. As it begins Sun is 29, the personal assistant to a founder of an investment firm in Manhattan. This man was “the first billionaire I would ever meet”. Sun herself was possibly also the most overqualified assistant he would ever meet: she had a degree in maths from MIT, went to the Wharton School of Business for an MBA, and had many years of experience as an investment manager at Fidelity Investments. She could perhaps have had his job in another life, but billionaires know they can get what they want. She was on nearly $300,000, plus bonuses, plus $5,000 department store vouchers he could scatter towards her like tips out of his loose change. “Remember,” he says to Sun, “money can solve nearly everything.”

Her memoir, then, has some of that democratic deliciousness, an exploitative boss exposed by his most intimate and trusted aide. Anyone who watched Meryl Streep playing a version of the Vogue editor Anna Wintour making imperious demands in The Devil Wears Prada will find some crossover here. Sun’s job was to make the impossible happen. She had to ghost-write inspirational speeches at the last minute; organise fairgrounds in Central Park for corporate “family days” that will remain empty because no analyst has the time to be with his family; distil self-help books for his endless “personal growth”; and prep her boss with information that should be private by utilising all legal means available — the latter a useful skill in a writer.

But the central insight runs deeper than that. Sun was a committed capitalist. Her parents fled to America from a communist China where “my dad was hungry much of his life”, she tells me. Her parents “ate bitterness”. As a result, when Sun was a child she wanted nothing but financial stability. She illustrated her first babyish attempt at a book with herself in a business suit: “CEO Sun”. She wrote a story in which money literally grew on trees. To her, like so many, finance was “an industry built on the American dream: you can be anyone as long as you can make the right calls. The industry thinks of itself as among the most meritocratic.”

However, her time spent deeply embedded with the private jet set challenged everything. This wasn’t democracy in operation, it was the opposite: power being concentrated among the shadowy super-rich, gobbling up large swathes of the companies that rule our lives and with less and less accountability. Private Equity is a thoughtful meditation on what happens when rampant self-interest is left unchecked.

● Why private equity is back — and which companies are on the hit list

She is now 39, married to a composer and living in New Jersey, living an artistic life completely alien to her decade on Wall Street. Who, I ask her, is the more ruthless, the writer or the investment banker?

“The honest answer is both,” she says, deep-voiced, articulate. “No characters in this book are made-up or composite. It is entirely true apart from some details changed to preserve confidentiality and privacy. I struggled a lot with the sense of betrayal … A writer has to be ruthless for the truth. An investment manager is ruthless for profits. In my world view, truth is more valuable than profits.”

The people Sun worked for were not obvious villains. They disdained the typical amoral “fat cat” bosses of the 20th century, immortalised in films such as Wall Street and books like Liar’s Poker. Instead, they believed themselves the 21st-century ideal of benign dictators, superior to everyone according to superhuman Nietzschean ideals: smarter, stronger, bettering humanity while buying a yacht. Among similar we can count the high-profile tech bros Elon Musk and Mark Zuckerberg. She says her boss, whom she anonymises as Boone, would never cheat on his wife — “so inefficient” — but that he, like her company, which she calls “Carbon”, was private to the point of paranoia.

I’m amazed it was possible to write this book at all, I say. Sun explains that after her 14 interviews to get the job, she signed some confidentiality agreements when she began work for Carbon.

“My publishers, and I also work with a lawyer, feel I am able to publish this book and tell my story of the working conditions.” She does not know the reaction of her former boss or the company, saying: “I’ve heard nothing from them.”

At first, the trappings of wealth were seductive. Sun writes with wit about her induction to the firm, a tour of the gym with its floor-to-ceiling windows onto Central Park, racks of fresh designer workout gear just waiting to be taken from the hanger, and the vast platters of food from the most expensive restaurants that are slopped into the bin at the end of the day. His food preferences are, in his briefing note, “protein — not heavy. Salad/soup. Clean. Not mayo. No caffeine/red wine. Anti-cancer.” She is given an employee handbook noting that “individuals should refrain from using vulgarities, obscenities, jokes, sarcasm”. Next she is awarded “the sable fur of the investor class, a Patagonia fleece jacket with another Carbon logo embroidered on the left arm”. She got Boone into meetings with other billionaires or, as she puts it, she became the “UPS guy whose packages were rich white men”.

Boone heavily invests in companies promoting 21st-century convenience tech. These offer the dream of more leisure through more efficiency. In fact, they do the opposite: more efficiency breeds more work, and tech enables the bleed of work into personal life that has accelerated modern stress and isolation for so many of us. Sun lived this up close. When she is hired Boone wonders aloud what she will do with her free time outside her working hours of nine to six. Yet this is the very appeal of her taking this demotion, to give her time to work on her writing.

“I believed in the possibility of good billionaires,” she writes. “Believing in him meant I didn’t have to change my world view.”

In reality, on her second day she is presented with a list of 96 of his demands on her time, which she is to give unstintingly. That morning she is called to a “sit” (“which means a meeting”) with Boone in which she is admonished that she did not instantly reply to an email he sent the night before. She works “11 hours a day” at the office and never leaves before Boone, unless she is injured.

And she does once get injured. While running on the treadmill, she realises an email has arrived from him. Wrestling between getting a few minutes of personal time and jumping to his tune, she attempts to look at her phone while running on the gym’s hamster wheel — a metaphor for 21st-century office life if ever I heard one. She flies off and her designer leggings don’t protect her knees from deep wounds. She gets an Uber home and an Uber back to the office the next morning. “No more personal goals,” she reminds herself.

“I might not be working all day Saturday or Sunday, but I might receive an email or a text that means I have to instantly change my plans for the day. It wasn’t just the number of hours I worked, it was the quality of the hours off.”

● Confessions of a banker: office porn, drug binges, £25,000 bar bills

In two years she lost touch with her friends, broke up with boyfriends and developed an eating disorder as she ordered more sweet snacks to cope with the stress. He offers her diet tips: when she does a juice cleanse for a week, he tells her to extend it for two. He owned you, didn’t he?

“Yes.”

She is not looking for sympathy; she was well paid and there by choice.

“But the inequities that I felt at work are very universal,” she says. “Symbolic, not just in finance, but all over the world.”

Our lives are ever more owned by our bosses, our privacy ever more invaded by work or tech.

It takes longer for her to develop cynicism about the “goodness” of Boone. “I believe he believed it,” she says. Financiers like to say “earn a lot now, give a lot later”. This way they can offset any guilt at their lavish lifestyles. In a particularly televisual moment, Sun has the job of transporting an obscene amount of cash via taxi, which gets held up and surrounded on Wall Street by anti-capitalist protests. While Sun is charged with addressing the philanthropic side of the business, she has an epiphany. Their work is a vicious circle. The business served to widen inequality, concentrating wealth. The effect was hardly offset by donations to charities and museums, with its favourable tax breaks.

“It was like I was frosting over the fundamental truth of what they were doing: actually widening inequality every day,” she says. “I began to question: if you’re making it worse when you’re making your billions, the philanthropy won’t be able to make it up on the back end.

“Also, their philanthropy was still a way for them to control the decision-making: they get to choose the world they want to bring about and the problems that their philanthropy wants to fix. I didn’t see any billionaires, when I was working there, giving their money towards solving major systemic issues that would eat away at their ability to make their billions. It was self-serving. It goes back to self-interest.”

● How to stay young like a billionaire

The public fascination with wealth, which she freely admits she shared, shows up in cultural life with the popularity of shows such as Succession or White Lotus.

“There’s a deep curiosity: who are these people? There’s so little information flowing out, when in fact these firms and these billionaires have a lot more influence and power than maybe people even realise,” she says. “There’s also a sense in which many people with extreme wealth often are able to get away with a lot, including criminal things. I think sometimes people look to those fictional ‘eat the rich’ satires like Succession for a sense of cosmic justice. There is something satisfying about seeing the rich get what they deserve.”

That rarely happens in real life. How does she feel about Boone now? Does she feel he exploited her? She pauses.

“I think he did exploit me, but never intentionally,” she says carefully. He saw “burnout” as a badge of honour, “never mind that his burnout enriches himself and that he gets to be a billionaire, whereas my burnout further enriches him.”

In the end she quit. She realised she too had forgotten the name of the woman whose only job is to flush away the men’s toxic waste, and nobody else in the firm knows.Private Equity by Carrie Sun (Bloomsbury £20). Buy from timesbookshop.co.uk Special discount available for Times+ members

It hit the shelves in America last week, sparking a downpour of (mostly fawning) reviews. Having tried it myself, one thing is inarguable: using the Vision Pro is a bewildering experience in every respect.

That starts with its presentation. After an iPhone camera is used to calibrate the size and shape of one’s face, the ski-goggle-style headset is fitted with a set of seals to keep light out and ensure optimal fit.

Moments later, an Apple store employee materialises in blue latex gloves and presents, with the solemnity of someone transporting the crown jewels, a sculpted tray that cradles the device. What follows is a half-hour demonstration where I am walked through how to use my eyes, hands and voice to control a device that projects apps, stunning immersive video and photographs onto the wearer’s field of view. The Vision Pro is a stunning technological achievement, of which more shortly.

But it also faces immense obstacles to Apple’s lofty aim of creating an entirely new industry of “spatial computing”. The stakes are very high. Meta has famously spent more than $46 billion (£36.5 billion) on building the metaverse, an immersive digital world to which access can be gained through its rival headset, the Meta Quest. More than a decade since the first version was released, however, it remains a niche product used mostly by gamers.

The most obvious barrier to Apple’s new gadget is the price. At $3,500, the Vision Pro is simply too expensive except for the biggest, and most well-heeled, among Apple’s super-fans. The Meta Quest, by comparison, costs $500.

A price tag of $3,500 demands a mind-blowing experience — a sense that a new technological frontier has been opened. On that score, Apple has done well. For one thing, the eye-tracking is spookily accurate. As I gaze at the apps floating in front of me, the one I focus on is highlighted. Pinch the air where that highlighted app is, and it opens.

The most jaw-dropping experiences, however, are of the “immersive” videos made especially for the Apple Vision Pro. Before I click (pinch?) on it, the Apple employee reads a script that sounds a bit like a pre-flight safety briefing. I quickly understand why, as I am suddenly placed on a vertiginous cliff next to a mountain climber. I am then engulfed by turquoise water, surrounded by great white sharks, after which I am floating in a hot air balloon. Looking down inspires a drop in the stomach; a pang of nausea strikes. The sensory experience of being suddenly and viscerally somewhere else, as I sit on a stool in an Apple store, is unsettling.

James Cameron, the director of the film Avatar, told Vanity Fair magazine that the Vision Pro was “revolutionary” and would transform the entertainment landscape. Others have been more circumspect, with both YouTube and Netflix pointedly declining to make their services available on it. Ted Sarandos, Netflix’s co-chief executive, explained last month: “The device is so sub-scale that it’s not really particularly relevant to most of our members.”

When the demo is over, I take off the headset — and the most immediate sensation is, oddly enough, relief, not unlike when you get off a rollercoaster and have a new-found affection for solid ground.

And this is something that Cook & co have to address. The Vision Pro is an expensive joy ride. It is not clear what problem Apple is solving, and what will inspire hundreds of millions of people to buy the device.

Whether Apple can pull it off will hinge on its ability to persuade software developers to make apps for the Vision Pro. These are the same developers that for years have groused about Apple’s treatment of them — about the 30 per cent commission it rakes in on in-app purchases and has become the centrepiece of multiple antitrust lawsuits around the world.

The iPhone example, however, is instructive. It was not until Apple opened the App Store, a year after it launched the iPhone, that the device transformed from just another phone into the taxi-hailing, food-ordering, social media hub to rule them all. It now offers nearly two million apps that together comprise their own micro-economy.

Apple launched the headset with 600 apps — not a lot, but more than the 570 or so that the Meta Quest App Store offers, despite a headstart of several years.

By 2030, Gene Munster of the Minnesota-based Deepwater Asset Management predicted, the Vision Pro will account for 15 per cent of Apple’s global revenue — but he acknowledged that it has a very long way to go: “I have this sinking feeling that I just spent $3,800 [$300 tax added to the retail price] on a device that is going to have pretty limited utility over the next one to three years.”

Indeed, the in-store demo encapsulates the challenge — devoting half an hour to teaching people how to use the device is a labour-intensive sales strategy. Another problem is the weight of the Vision Pro: 20 ounces. After that half-hour, I was very happy to get the technology, which left a bright pink crease across my forehead, far away from my face.

The biggest question, however, is whether this is the tech the world wants — or needs. Its release came in the same week in which the chief executives of the top social media companies were hauled before Congress to account for the role their technology plays in a terrifying catalogue of online harms to young people, from bullying to loneliness.

The Vision Pro is genuinely impressive — the product of more than 5,000 patents, billions of dollars of investment and seven years of toil. And yet, to put it on is to put a screen between yourself and the world. Apple unwittingly highlighted this reality in part of the demo, which revolves around a three-dimensional video of a child’s birthday.

The picture is so crisp, the spatial audio so adept at matching sounds and voices to the scene, that it feels like you are actually there — that you can reach out and touch it. And yet, you can’t. The result is a strange sense of melancholy, of unbridgeable distance. You’re a ghost.

No such sensations occur to Cook, who rhapsodised about the day he spent at an Apple store last week watching consumers doing their demos with the device. “People had tears in their eyes,” he wrote. “Our mission is to enrich people’s lives, and I could feel that happening in real time. What a day!”

The US Food and Drug Administration gave Neuralink permission last year to conduct its first human trials. The company hopes its technology will allow paralysed patients to control external computers using neural signals. Such technology could prove life-changing for sufferers of conditions such as motor neurone disease.

Announcing the first human transplant, Musk said the patient was “recovering well”, adding: “Initial results show promising neuron spike detection.”

Neuralink is perhaps the best-known company working in the field of brain-computer interface, or BCI. Similar devices to the Neuralink’s chip have allowed disabled patients to type by thinking about keyboard inputs and to walk again despite being paralysed from the waist down.

Famously, in 2016 a patient was able to fist-bump Barack Obama with a robotic hand that he controlled with a brain chip.

Neuralink, however, not only wants to restore function but to enhance human capabilities. “We want to surpass able-bodied human performance with our technology,” the company has previously said.

The start-up, which is based in Fremont, California, said that during the study a robot would place the implants’ “ultra-fine” threads that help transmit signals in participants’ brains.

Musk, the 52-year-old entrepreneur whose other companies include Tesla, has suggested the implant could be upgraded, similar to how mobile phones receive software updates. “I’m pretty sure you would not want the iPhone 1 stuck in your head if the iPhone 14 is available,” he said at an event in 2022.

Neuralink has attracted controversy. In 2022 it was reported that employees had complained about rushed animal testing that allegedly resulted in needless death and suffering. Reuters said the company had killed about 1,500 animals in experiments from 2018, including more than 280 sheep, pigs and monkeys.

This month, it also emerged that Neuralink was fined by the US Department of Transportation for violating rules regarding the movement of hazardous materials.

“Progress is good. The patient seems to have made a full recovery, with no ill effects that we are aware of,” Musk said during a broadcast on X, the social media platform formerly known as Twitter, which he also owns.

He added that the first volunteer, who has not been named, was now “able to move a mouse around the screen by just thinking” after the device was embedded into their grey matter last month.

Being able to move a mouse via a brain implant is not, by itself, a groundbreaking step. In the early 2000s, American scientists showed that monkeys implanted with neural interfaces could control robotic limbs with their thoughts.

Since then, humans fitted with similar systems have played video games, manoeuvred robots, sent emails and made purchases on Amazon, all without lifting a finger, using computers that were deciphering the electrical activity of their brain.

In another pioneering experiment, researchers at the University of California San Francisco fitted an implant that gave a paralysed man a voice for the first time in more than 15 years.

Musk has suggested that, eventually, he would like to go much further. He has described his ultimate aim as the creation of a mass-market “general population device” that would directly connect users’ minds with powerful computers to achieve “symbiosis with artificial intelligence”.

For now, though, Neuralink appears to be focused on people who have lost the ability to move their arms and legs. “Initial users will be those who have lost the use of their limbs,” Musk wrote on X last month. “Imagine if Stephen Hawking could communicate faster than a speed typist or auctioneer. That is the goal.”

Experts regard Neuralink as a serious player in a field that is advancing quickly. While other sophisticated brain implants have involved patients having wires trailing from their heads, Musk’s is designed to pick up a large amount of information and beam it out wirelessly.

“The wireless technology is a big step,” said Andrew Jackson, professor of neural interfaces at Newcastle University, who is not involved with Neuralink.

“This is a technology that could be able to send high volumes of data recordings from many, possibly hundreds, possibly thousands of brain cells to control external devices and that would be a genuine technological advance.”

Details are scarce, but the first trial participant seems to have been recruited for what the company has called the PRIME study.

The Neuralink website says that it will involve testing a brain-computer interface (BCI) — a coin-sized device that sits in a hole drilled in the skull and which would be invisible to others. It will also test a surgical robot, specially built to thread a set of very fine, flexible electrodes into the brain itself while minimising damage to the surrounding tissue.

“The device is designed to interpret a person’s neural activity, so they can operate a computer or smartphone by simply intending to move — no wires or physical movement are required,” the Neuralink website explains.

It tells potential volunteers: “If you have limited or no ability to use both hands due to cervical spinal cord injury or amyotrophic lateral sclerosis (ALS) you may qualify.”

It says that the implant will monitor activity in the motor cortex, a part of the brain that runs across the top of the head and directs movement. It is a region that is relatively accessible and AI systems have already been able to interpret its activity to help people who have been paralysed to stand and take steps again, albeit with the help of walkers.

“If you invent a new telescope or something like that, it makes sense to first point that telescope at something that you already know what it looks like,” Jackson said. “So you can kind of validate that your telescope is working in the same way as previous ones.”

The company’s N1 implant has 64 flexible threads that can be placed into the brain. These contain a total of 1,024 electrodes. By contrast, the most widely tested implant at the moment, known as the Utah array, has only 100 stiff needle-like electrodes, each about a millimetre long. Neuralink is betting that its device will provide a far more detailed picture of what neurons are firing and when, and that this will make the technology more useful.

Experts have said that it will take time to know how well it works. One question will involve whether scar tissue builds up around the electrodes, blurring the signals they provide.

However, they also said that merely embarking on a first human trial is significant. Anne Vanhoestenberghe, professor of active implantable medical devices at King’s College London, said: “Whilst there are many companies working on exciting products, there are only a few other companies who have implanted their devices in humans, so Neuralink has joined a rather small group.”

She added: “We know Elon Musk is very adept at generating publicity for his company [but] true success in my mind should be evaluated in the long-term, by how stable the interface is over time, and how much it benefits the participant.”

How does the Neuralink N1 implant work?

Neuralink’s N1 device is about 2cm wide and 8mm deep. It sits in a hole drilled in the skull and 64 very fine threads, each a few centimetres long, dangle from it. It contains a small battery that can be charged wirelessly

A robot arm built by Neuralink is used to place the threads into the brain; in the first trial they seem to have been inserted into the motor cortex region that runs along the top of the head. Each thread contains 16 electrodes, which monitor the activity of clusters of neurons.

This data is beamed wireless to a computer. We already know, from previous research, that when a person thinks about, say, moving their arm, an AI can decipher the bursts of electrical activity in the motor cortex to predict their intention. It can then move a cursor on a screen accordingly.

The folksy American executive, 59, was inspired by the eye-catching announcement that the streaming service had struck a $5 billion, ten-year deal with World Wrestling Entertainment to broadcast its most popular weekly live show WWE Raw.

However, despite the razzmatazz surrounding WWE, the deal was eclipsed by the streamer’s numbers. Over the final three months of 2023, the group added more than 13 million subscribers, its best performance in three years, to take its total base to 260 million ­accounts.

Revenues over the quarter grew 12 per cent to $8.8 billion with net income of $938 million. The results fixed its position at the top of the streaming tree as it stretched its lead over its closest rival Disney+, which topped 150 million subscribers over the three months to the end of September but lost $387 million.

Netflix’s results prompted one US analyst to proclaim total victory for The Crown broadcaster. “It is becoming increasingly clear that Netflix has won the streaming wars,” Jessica Reif ­Ehrlich, at Bank of America, said.

Worryingly for its rivals, Netflix’s appetite for dominance shows no signs of slowing. It sees significant room for growth after estimating that it accounted for only 5 per cent of the $600 billion market across pay TV, film, games and branded advertising, with a share of viewing under 10 per cent in every country.

Advertising will be an important driver of growth, going hand-in-hand with its crackdown on password ­sharing by its users, tactfully described as the introduction of “paid sharing”.

It is to phase out its £7.99 per month basic package, which does not include ads, with subscribers forced to either adopt the £4.99 advertising tier or the costlier standard subscription, at £10.99 in the UK.

It is seeking to offer brands real scale by expanding the 23 million monthly active users who are on the ­ad-funded plan, which are yet to make a material impact to its bottom line. Encouragingly, 40 per cent of sign-ups opt for the advertising-supported plan in the 12 countries where it is available.

Greg Peters, the co-chief executive, indicated that it was back to “business as usual” after the password crackdown as he ­indicated that further price rises were on the horizon. “We largely put price ­increases on hold as we rolled out paid sharing,” he said. “Now that we’re through that, we’re able to resume our standard approach.”

Tom Harrington, head of television for Enders Analysis, said people were only prepared to pay for a maximum of three video subscriptions — with Netflix, Disney+ and Amazon Prime Video emerging as the dominant players. Newer entrants such as Paramount+, NBCUniversal’s Peacock and Warner Bros Discovery’s Max are expected to struggle in the battle for subscribers. “If you’re watching one service, then your sense of the value for money that the others provide is lowering,” he said.

He pointed to the market-leading 45 minutes per day users spend with Netflix as driving its unexpected success in converting “parasitic” users who have been password sharing for up to a decade.

“The outperformance of this initiative is fundamentally a testament to the product itself,” he said. “If you have been getting something for free for a long period of time then it is difficult to persuade people to pay but Netflix ­appears to have converted many non-paying users.”

This leaves less room for rivals, who are starting to question how far they can compete. They are selling increasing numbers of the shows they produce back to Netflix, aborting the hardline stance of “warehousing” content to help build their own services.

Disney has struck a wide-ranging deal with Netflix for titles including Lost, This Is Us and How I Met Your Mother, while the legal drama Suits enjoyed a renaissance after being sold to its rival by NBCUniversal. HBO’s Sex and The City is expected to follow, rather than remaining solely on Max.

“Everyone has started to licence their content again because they have to make money,” Harrington said. “They have all this content that people aren’t watching and they can sell it to someone else. The market wanted content to be kept exclusively but everything has changed and people need to be a bit more realistic about spend.”

Paolo Pescatore, analyst for PP Foresight, said that Netflix remained in “pole position” because it was unencumbered by the legacy businesses that dogged the TV networks. Neither did it have to consider the wider concerns of selling other products which hamper the focus on video for its tech rivals Amazon and ­Apple. “Netflix is definitely the king among all streamers,” he said. “It has a strong catalogue, broad ­offering and is moving towards being a one-stop shop for all video services.”

Pescatore echoed Netflix’s expectation that there will be further industry consolidation as consumers reject paying for a host of smaller services, potentially leaving Netflix, Disney+ and Amazon standing alone as the major players.

“Amazon has an inferior catalogue but becomes a key service for households because of the value of the overall subscription and you cannot ignore the strength of Disney’s IP and franchises,” he said.

Netflix is unlikely to acquire its rivals, however, as executives underlined its preferred strategy to “build and not buy”. Despite Netflix’s WWE wrestling partnership, it is not expected to pave the way for more sports deals.

WWE was branded a unique live entertainment proposition that boasts a healthy shelf-life, perhaps matched ­only by the Ultimate Fighting Championship (UFC), which promotes mixed martial arts.

Insiders disputed the suggestion that the war was won, pointing to a future in which a trio of deep-pocketed rivals slug it out for subscribers while simultaneously ensuring that the streaming space remains healthily competitive. At heart they consider the quality of its content to remain paramount.

Emma Cooper, founder of Empress Films, who was responsible for the Netflix documentaries Who Killed Jill Dando? and The Disappearance of Madeleine McCann, said Netflix had not lost sight of what had made it the market leader. “There is a high benchmark of creative excellence,” she said. “They know what their audience wants and deliver it really well by giving filmmakers the freedom to tell stories in the way that they want.”

Patrick Holland, chief executive of Banijay UK, which is responsible for Black Mirror, added that Netflix knows its audience better than anyone. “It has been in the game longer, its data is richer and it has a nuanced set of commissioning criteria around the type of genres and programming tones that work.”

It is also better protected from the impact of the strikes by US actors and writers than its rivals due to its more international slate of shows. With a $17 billion war chest to invest this year, it may not yet have the industry in a chokehold but some of its smaller adversaries might be feeling pinned down.

The FTSE 100 pharmaceuticals company expressed its confidence in datopotamab deruxtecan, or Dato-DXd, when it presented overall survival data for the treatment at a world lung cancer congress in San Diego, but analysts promptly questioned the strength of the possible benefits.

That, in turn, weighed on AstraZeneca’s share price, which was down by 508p, or 4 per cent, at £122.04 in afternoon trading.

Unveiling results from a phase III trial of Dato-DXd in treating a type of non-small cell lung cancer, compared with standard chemotherapy, the Cambridge-based company said that results for the study, in patients who had undergone at least one previous therapy, showed a “clinically meaningful trend toward improving overall survival”.

Susan Galbraith, executive vice-president of oncology research and development at AstraZeneca, said the results built on previously reported progression-free survival data and “underscore our confidence in the important role datopotamab deruxtecan can play across segments and settings of non-small cell lung cancer”.

However, analysts at Bernstein, the broker, said that in the “sub-population” that forms the basis of a regulatory application for the drug to the US Food and Drug Administration, its “superiority, a 2.3-month overall survival benefit over chemotherapy, wasn’t statistically significant” and that therefore “the ‘play of chance’ could not be excluded from these results”.

Dato-DXd, which is being developed jointly with Daiichi Sankyo, of Japan, is among the key drugs in AstraZeneca’s pipeline as it looks to continue the successful launch of a series of new treatments, particularly in oncology, that have transformed it into the most valuable company on the London Stock Exchange.

At a capital markets event in May at its new research and development centre in Cambridge, AstraZeneca said it expected more than 40 late-stage clinical trial results by the end of next year, with about $20 billion of potential revenue in 2030 from so-called readouts and launches this year and next. These included Dato-DXd for lung and breast cancer.

Dato-DXd was acquired through a deal with Daiichi Sankyo for an upfront $1 billion four years ago. Shares in Daiichi Sankyo fell by almost 9 per cent in Tokyo after news of the latest results emerged.

Investors and analysts have been tracking the development of the drug, which analysts at Stifel said had “made noise in the past because results were not as positive and easy-to-read as expected”. Indeed, shares in AstraZeneca previously weakened in July last year amid disappointment at results showing the benefits for slowing the progression of lung cancer. Investors have reduced their commercial expectations for the medicine.

Sir Pascal Soriot, 65, the chief executive of who has overseen the transformation of AstraZeneca since 2012, said in October that he hoped people would “stop their myopic focus on Dato-DXd and look at the rest of the portfolio”.

Stifel said: “There is an unequivocal benefit with Dato-Dxd for some patients and at a reasonable price in terms of toxicity, but the exact target population and how it should be addressed are complex topics.”

In February, AstraZeneca said that an application to the FDA in America had been accepted for Dato-DXd after the earlier progression-free survival data and that a decision on regulatory approval was expected in the fourth quarter of this year.

Progression-free survival refers to the length of time during and after treatment that a patient lives with the disease but it does not get worse. Analysts at Jefferies said that “given that one of the trials’ co-primary end-points, progression-free survival, has been met, the probability of approval remains high”.

More than a million people are given a diagnosis of advanced non-small cell lung cancer each year. Although therapies have improved outcomes, the disease eventually progresses for most patients who receive chemotherapy, which for decades has been the last treatment available, despite “limited effectiveness and known side-effects”.

Last week AstraZeneca confirmed that a “small number” of its employees in China, an important market for the company and its cancer drugs, were under investigation after Bloomberg reported that five present and former staff had been detained by police over alleged illegal activities.

Some users are taking to social media to explain why the $3,500 headset is going back to the store, citing headaches, heaviness, loneliness and even burst blood vessels. Apple allows customers to return products within 14 days — a window that closes on Friday.

The Vision Pro has been promoted as an entertainment and productivity device, where users can immerse themselves in films, videos and photos and create a virtual office before their eyes.

Many have praised its technical quality and the way it is controlled using the eyes and hands.

However, some have complained about its weight and comfort. “Can’t wait to return the Vision Pro, probably the most mind-blowing piece of tech I’ve ever tried. Can’t deal with these headaches after 10 minutes of use though,” said Rjey Tech, a YouTube reviewer from Toronto.

A similar sentiment came from a member of Reddit’s Vision Pro forum called nindesk, who posted: “Absolutely loved it, but it was too heavy, causing neck pain.” The headset weighs up to 680g, depending on the configuration.

Parker Ortolani, a product manager at The Verge website, posted on Threads: “I have developed a large bloody blotch on my right eye. Thanks Vision Pro!”

He later said: “What a bummer of a day. Can’t believe it, but I’ve returned the Vision Pro. Just too uncomfortable to wear and it’s a strain on my eyes. It’s clearly the future. It works like magic. But the physical trade-offs are just not worth it for me right now. I’ll be back for the next one, assuming they fix these comfort issues.”

John Grant, a VR expert who works for the US Defence Intelligence Agency, wrote on LinkedIn: “I just returned my personal Apple Vision Pro headset. It is an amazing device, but generally for personal use at its price point isn’t for me.”

UrbanExtant posted on Reddit that despite the “wow factor” he and his wife found it isolating. “I had this dreadful sense of loneliness and guilt that I’d ignored my entire family all night.”

Apple is predicted to sell 600,000 headsets this year, according to Dan Ives, an analyst for Wedbush Securities, and there are no hard figures on the number of returns, making the extent of the backlash difficult to gauge.

There were also many positive reviews online: “Loving my Apple Vision Pro so far. It’s clear to me that this will be a major new paradigm of computing. I’m all in,” one person posted on Reddit.

Leander Kahney, editor of Cult of Mac, noted: “I don’t remember early adopters returning new iPhones, AirPods or even recent Apple silicon machines. The phenomenon seems restricted to Apple Vision Pro.”

He added that an “unscientific” poll his website was conducting showed only 55 per cent of respondents planned to keep the device, which could suggest “a massive number of returns”.

However, Mike Wuerthele, managing editor of AppleInsider, countered: “So far, based on the data that I have, there’s not a noticeable difference at the two-week cut-off than any other product. Yes, folks are returning, sure. There’s always some.”

• Apple Vision Pro gets rave reviews — but Tim Cook won’t predict future

Leo Gebbie, a connected devices analyst at CCS Insight, said: “It’s no real surprise to hear reports of discomfort from people who have bought a Vision Pro. Our research has consistently found that the weight of these devices, and users feeling sick when using VR, are pain points which put people off using the technology. This is a problem facing the VR industry overall, not just Apple.

“Apple may have exacerbated this problem by its insistence on a premium design philosophy. The Vision Pro uses materials like aluminium and curved glass to achieve a beautiful aesthetic, but these add on weight.

“There’s a chance that these people are actually suffering from buyers’ remorse. $3,500 is an eye-watering amount to spend on an experimental new product, and perhaps the buyers have found that the Vision Pro isn’t doing enough to justify its lofty price tag.”

One person adding to the criticism is Mark Zuckerberg, who sought to promote his rival device, the Meta Quest 3 (£480, 515g), by reviewing the Vision Pro. Unsurprisingly he concluded: “I’m pretty surprised that Quest is so much better for the vast majority of things that people use these headsets for with that price differential.”

The Vision Pro is on sale in the US only and is expected to be released worldwide later this year.

Apple was approached for comment.