The producers of Blade Runner 2049 are suing Elon Musk over his robotaxi launch imagery. Too similar? CC-licensed photo by Rikard Auregård on Flickr.

You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

---

There’s another post coming this week at the Social Warming Substack on Friday at 0845 UK time. Free signup.

---

A selection of 10 links for you. POB, RTB. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. Observations and links welcome.

---

‘Operating in the Stone Age’: NHS staff’s daily struggle with outdated tech • Financial Times

Laura Hughes:

»

In the paediatric centre at one of London’s largest hospitals, doctors are confounded each day by a ward computer that is not connected to a printer.

The computer is used for managing the daily list of patients. Doctors can only access and update the list, using one shared account.

So twice a day, two doctors on the ward said one of them had to log in to this computer, update the patient list, send the list to themselves via NHS email, and then log in to another nearby computer to print it off for the team.

“I am at a top London hospital and yet at times I feel as though we are operating in the Stone Age,” said one paediatrician on the ward.

Tackling the frustrating delays caused by outdated technology is one of health secretary Wes Streeting and Prime Minister Sir Keir Starmer’s core missions, having vowed to shift the service “from an analogue to a digital NHS”.

The monumental task of moving the world’s largest publicly funded health service into the digital age is not lost on doctors working on the frontline of the NHS.

While many sectors of the economy have been “radically reshaped” by technology in recent years, a landmark report into the state of the health service in England last month concluded that the NHS stood “in the foothills of digital transformation”. 

«

There are multiple challenges to improve the NHS: outdated technology, insufficient management, insufficient data sharing. The public has been invited to offer suggestions. They are mad.
unique link to this extract

---

Inside the mind of an AI girlfriend (or boyfriend) • WIRED

Will Knight:

»

Last month, OpenAI unveiled an ambitious new language model capable of working through challenging problems with a simulated kind of step-by-step reasoning. OpenAI says the approach could be crucial for building more capable AI systems in the future.

In the meantime, perhaps a more modest version of this technology could help make AI girlfriends and boyfriends a bit more spontaneous and alluring.

That’s what Dippy, a startup that offers “uncensored” AI companions is betting. The company recently launched a feature that lets users see the reasoning behind their AI characters’ responses.

Dippy runs its own large language model, which is an open source offering fine-tuned using role-play data, which the company says makes it better at improvising when a user steers a conversation in a particular direction.

Akshat Jagga, Dippy’s CEO, says that adding an additional layer of simulated “thinking”—using what’s known as “chain-of-thought prompting”—can elicit more interesting and surprising responses, too. “A lot of people are using it,” Jagga says. “Usually, when you chat with an LLM, it sort of just gives you a knee-jerk reaction.”

Jagga adds that the new feature can reveal when one of its AI characters is being deceptive, for instance, which some users apparently enjoy as part of their role-play. “It’s interesting when you can actually read the character’s inner thoughts,” Jagga says. “We have this character that is sweet in the foreground, but manipulative in the background.”

I tried chatting with some of Dippy’s default characters, with the PG settings on because otherwise they are way too horny. The feature does add another dimension to the narrative, but the dialog still seems, to me, rather predictable, resembling something lifted from a bad romance novel or an overwrought piece of fan fiction.

«

*supremely bored voice* You don’t say. I wonder what sort of content it might have been trained on.

More concerning: what sort of people are going to use these things? What’s their relationship with humanity?
unique link to this extract

---

Inside the bungled bird flu response, where profits collide with public health • Vanity Fair

Katherine Eban:

»

on March 25, the USDA lab confirmed that dairy cows in Texas and Kansas had indeed been sickened by a form of bird influenza known as H5N1. Though versions of the so-called bird flu virus have circled the globe for almost two decades, spreading to species ranging from pelicans and polar bears to sea lions and skunks, the announcement stunned the scientific and agricultural communities. “Every honest virologist will tell you: We did not see this coming,” says Kimberly Dodd, dean of Michigan State University’s College of Veterinary Medicine.

“We plan for every agricultural health emergency, but all of our red teaming missed this” scenario: an agricultural outbreak that potentially imperils public health and leaves cows sick but mostly still standing, says David Stiefel, a former national security policy analyst for the USDA.

With continued spread amongst cows, or to another “mixing-vessel” species like pigs, the virus “could mix and match, then you get a whole new genetic constellation,” says Jürgen Richt, regents and university distinguished professor at Kansas State University. Experts are hesitant to speculate about what could happen if the virus were to begin more widely infecting humans, for fear of spreading panic, but the toll could, in the worst case, dwarf that of COVID-19. If the virus “infects a person infected with a human flu strain, and something comes out that is reassorted and adapted to humans? I don’t even want to imagine,” Richt says. “Not good.”

The Institute for Disease Modeling, a research institute within the Bill & Melinda Gates Foundation, has estimated that a global flu pandemic could kill close to 33 million people within six months.

At that existential moment back in March, when the virus was first detected in cows, veterinarians involved in the response had every expectation that a well-honed network of experts, led by USDA scientists, would immediately rev to life.

But it didn’t. “Nobody came,” says one veterinarian in a Western state. “When the diagnosis came in, the government stood still. They didn’t know what to do, so they did nothing.”

«

Very weary watching brief.
unique link to this extract

---

ByteDance intern fired for planting malicious code in AI models • Ars Technica

Ashley Belanger:

»

After rumours swirled that TikTok owner ByteDance had lost tens of millions after an intern sabotaged its AI models, ByteDance issued a statement this weekend hoping to silence all the social media chatter in China.

In a social media post translated and reviewed by Ars, ByteDance clarified “facts” about “interns destroying large model training” and confirmed that one intern was fired in August.

According to ByteDance, the intern had held a position in the company’s commercial technology team but was fired for committing “serious disciplinary violations.” Most notably, the intern allegedly “maliciously interfered with the model training tasks” for a ByteDance research project, ByteDance said.

None of the intern’s sabotage impacted ByteDance’s commercial projects or online businesses, ByteDance said, and none of ByteDance’s large models were affected.

Online rumors suggested that more than 8,000 graphical processing units were involved in the sabotage and that ByteDance lost “tens of millions of dollars” due to the intern’s interference, but these claims were “seriously exaggerated,” ByteDance said.

The tech company also accused the intern of adding misleading information to his social media profile, seemingly posturing that his work was connected to ByteDance’s AI Lab rather than its commercial technology team. In the statement, ByteDance confirmed that the intern’s university was notified of what happened, as were industry associations, presumably to prevent the intern from misleading others.

«

unique link to this extract

---

Tim Cook on Apple Intelligence, Vision Pro and more bets the company believes will pay off • WSJ

Ben Cohen:

»

Maybe the most surprising aspect of Vision Pro is how it makes you feel. You might not believe that strapping yourself into a piece of technology could be emotionally overwhelming. But when you experience an ultra-high-resolution spatial photo of your daughter at age 3, or watch an immersive video of a grandparent who’s since died, it’s no longer a headset. It’s a time machine. You put on this device from the future and find yourself reliving the past. You come back to the present and have tears in your eyes. 

“That really is why we did this product,” says Richard Howarth, vice president of industrial design. “It’s got the ability to do things that the other products can’t do.” 

There is no killer use case for the Vision Pro yet, so I asked Cook how he’s using it. At work, of course, when he wants several windows open for multitasking. But especially at home. “I’ve always viewed having to sit in a certain place in your living room as really constrained,” he says. He prefers to lie flat on the couch, project Ted Lasso and The Morning Show on the ceiling and stare into the Vision Pro. “It’s a lot more pleasant way to watch something than to sit like a statue in front of a TV,” he insists.

Jon M. Chu agrees. The director of Wicked grew up in Silicon Valley and bought a Vision Pro the first day it went on sale. From the second he put it on, he knew it would have a dramatic effect on his creative process. “Everyone here laughs at me because I’m so obsessed with it,” he says. Jobs once famously described computers as a bicycle for the mind. “I feel like Vision Pro is a rocket ship for the mind,” Chu says. “You don’t know where you’re headed, but you get to go someplace and figure it out with everybody.” 

But that rocket ship is an expensive ride. When the Vision Pro came out this year, mixed reality crashed into the reality that most consumers aren’t ready to shell out $3,500 for a cool toy. 

“Over time, everything gets better, and it too will have its course of getting better and better,” Cook says. “I think it’s just arguably a success today from an ecosystem-being-built-out point of view.” 

«

It’s not even close to having an ecosystem! It’s a flop in that regard. But perhaps the funniest part of the interview is this:

»

[I ask Cook: what is..] His wallpaper? A photo with his nephew in Grand Teton National Park. His most underrated app? Notes, where he types or dictates thoughts before he forgets them.

The best name of a group chat? He looked at me like I’d asked him to recommend the best Android phone.

“The best—name?” he said. “I don’t name them. Do you name yours? Interesting. I may take that on.”

The next time we meet, Cook proudly reports that he’s named the group chat with his college roommates: Roommates.

«

unique link to this extract

---

Apple’s AirPods Pro hearing health features are as good as they sound • The Verge

Chris Welch:

»

You’ll need a quiet space when taking Apple’s hearing test. Before getting started, your iPhone will do a quick analysis of ear tip fit and environmental noise to ensure you’re good to go. All of these hearing health features are calibrated for Apple’s stock silicone tips, so if you’re using aftermarket third-party tips (including foam), there’s no guarantee you’ll get the optimal experience. Once the test begins, you just tap the screen whenever you hear any of the three-beep tone sequences.

There are a few key things to know about Apple’s hearing test. For one, it’s designed so that you can’t predict or game it. The test can play any frequency at any time, so no two are the same. Apple tests your left ear first, and here’s something I wish I’d known going in: it’s completely normal to hear nothing at all for several seconds at a time. It was in those moments, when five, six, or even 10 seconds would pass without an obvious tone sequence, where I’d start feeling pretty anxious. 

My best advice is to avoid wondering if you should be hearing something at a given moment and instead just focus on the tones as they come. Some can be incredibly faint. There are visual cues that let you know the test is still moving along even during silence — the most obvious one being a large circle that animates onscreen throughout the process. (You’ll also notice a progress dial for each ear that fills as you take it.)

«

The Hearing Test is coming with iOS 18.1, which releases next week. Such fun to come as we discover how we’ve all lost hearing acuity – with graphs.
unique link to this extract

---

Sky Follower Bridge – Chrome Web Store

»

Instantly find and follow the same users from your Twitter follows on Bluesky.

«

Spotted via Ryan Broderick’s Garbage Day email. A Chrome extension, of course.
unique link to this extract

---

Can journalism survive? The media elite on its future • NY Post

Charlotte Klein in a biiiig piece about the media’s past, present and possible future:

»

To many, the most instructive failure of 2024 was The Messenger, an all-things-to-all-people news site run by ex Condé Nast and People executives with a “chief growth officer” formerly of Gawker Media. It planned on building out a 550-person newsroom and flooding the internet with viral scoops but instead burned through most of its $50 million in funding within a year and shut down. “I’m always suspicious when someone has a huge splashy launch saying they’re going to get up to 300 million page views in six months and reach a massive national audience,” says Betsy Reed, editor of The Guardian US. “I just feel like you can’t do that out of the gate. You need to have a much clearer grasp of who you’re reaching and why you’re going to be relevant.” What actually has succeeded this year are operations — many of them run through Substack — that have low overhead and a focused appeal. Some longtime media executives find this new world befuddling. “I’m surprised that people are okay with the subscription model, where they don’t have that many listeners or viewers but are making money, so they’re just good with it,” says one of them. “The Substack writers, people with Patreon podcasts. My generation was wired completely differently. We wanted to be read or listened to by as many people as possible. And now this new generation is like, I’m totally cool with having 9,000 die-hard fans.”

The Congress-focused media company Punchbowl News, which was founded by Politico veterans Jake Sherman and Anna Palmer in 2021, is well read inside the Beltway. “It’s so small and it’s so particular, and yet it seems like it has impact,” says Carolyn Ryan of the Times. “I don’t even know how many reporters they have — it feels like just a handful — but they really seem to have a sense of mission and what value they bring.” That value is priced at $350 a year — a lot for a general reader, but, as with Politico Pro, such subscriptions are often treated as a business expense by anyone with a need to be in the know.

«

Plenty more to read. But certainly right now the focus feels like it’s around niche subscriptions.
unique link to this extract

---

EVs are just going to win • Noahpinion

Noah Smith:

»

EVs are still winning. But they haven’t won yet; only 4% of the global passenger car fleet, 23% of the bus fleet, and less than 1% of delivery trucks are electrified.

But at this point I think the writing is on the wall. The phenomenon of a superior technology displacing an older, inferior technology is not uncommon, and it generally looks like the EV transition is looking now. When a new technology passes a 5% adoption rate, it almost never turns out to be inferior to what came before; with EVs, that threshold has now been reached in dozens of countries.

In fact, we don’t have to rely on trend-based forecasting to understand why EVs are just going to win. There are a number of fundamental factors that make EVs simply better than combustion vehicles. The longer time goes on, the more these inherent advantages will make themselves felt in the market.

The first of these is price. Currently, EVs often require government subsidies in order to be price-competitive with combustion cars. But batteries are getting cheaper and cheaper as we get better and better at building them. The cheaper batteries get, the smaller the subsidies required to get people to switch to EVs. Goldman Sachs reports that this crucial tipping point will be reached in about two years

…Once batteries cross that tipping point, the EV revolution will take on its own momentum. It will simply be cheaper to buy an EV than a combustion car. People will gravitate toward the cheaper option, especially if it comes with other advantages. And in this case it does.

EVs’ second advantage is convenience. Most EV owners will almost never have to fill their cars up at a station. This is because they will charge their cars at night, in their own home garages or driveway.

«

Not so sure about the latter. Lots of people live in flats in cities in the US same as Europe.
unique link to this extract

---

‘Blade Runner 2049’ producers sue Elon Musk over ‘Robotaxi’ imagery • The New York Times

Brooks Barnes:

»

The Hollywood company behind “Blade Runner 2049” sued Elon Musk for copyright infringement on Monday, accusing him of illegally using imagery from that film to promote Tesla’s new “robotaxi.”

Alcon Entertainment, a movie and television company backed by the FedEx founder Frederick W. Smith, filed the lawsuit in US District Court in Los Angeles. The complaint also names Tesla and Warner Bros. Discovery as defendants, saying that Alcon had denied a request by Mr. Musk and the companies to use imagery from “Blade Runner 2049” as part of an Oct. 10 marketing event on the Warner lot.

“He did it anyway,” the suit says.

Mr. Musk’s live-streamed presentation — a grand unveiling of a car that Tesla says will be able to drive itself — did not use exact “Blade Runner 2049” images, according to the complaint. Rather, the event showcased “AI-created images mirroring scenes from ‘Blade Runner 2049,’ including one featuring a Ryan Gosling look-alike,” Alcon said.

The lawsuit called the use of artificial intelligence tools to create near-identical images “a bad-faith and intentionally malicious gambit” to make the event “more attractive to a global audience and to misappropriate the ‘Blade Runner 2049’ brand to help sell Teslas.”

«

So he illegally used imagery from the film except it wasn’t imagery from the film? The comparison made in the story is with OpenAI using a soundalike to Scarlett Johansson after she refused permission to use her voice. I suppose the case will depend on precisely how closely the imagery matches: for a voice, there aren’t many dimensions that can vary, but with video?
unique link to this extract

---

• Why do social networks drive us a little mad? • Why does angry content seem to dominate what we see? • How much of a role do algorithms play in affecting what we see and do online? • What can we do about it? • Did Facebook have any inkling of what was coming in Myanmar in 2016? Read Social Warming, my latest book, and find answers – and more.

---

Errata, corrigenda and ai no corrida: none notified