Earlier this century Volvo had an anti-theft device that relied on your heartbeat. Why do you think it died? CC-licensed photo by Lars P. 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.
It’s Friday, so there’s another post due at the Social Warming Substack at about 0845 UK time.
A selection of 9 links for you. Still pumping. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.
OpenAI halts Studio Ghibli-style images trend • EWeek
Fiona Jackson:
»
If you’ve been wondering why your social media feeds have been awash with Studio Ghibli-style images this week, OpenAI’s new image generator is the answer. On Tuesday, the company embedded the multimodal tool into GPT-4o, and users have been transforming their photos into vibrant, whimsical scenes reminiscent of the Japanese animation studio behind “Spirited Away” and “My Neighbor Totoro.”
OpenAI’s new image generator allows you to upload an existing image to ChatGPT and simply ask it to change the art style in the accompanying text prompt. Even OpenAI’s CEO Sam Altman Ghibli-fied his X profile picture and playfully lamented about how his decade of work in AI has boiled down to people editing their photos into these images.
However, the fun didn’t last long. The system card for GPT-4o’s native image generator states as of Thursday that OpenAI “added a refusal which triggers when a user attempts to generate an image in the style of a living artist.” OpenAI acknowledged that the fact its tool can emulate named artists’ styles “has raised important questions and concerns within the creative community.”
The same was written in DALL-E 3’s system card from 2023, so OpenAI is not making a wild U-turn on its position when it comes to AI art ownership and copyright. Nevertheless, there is an ongoing debate about whether AI companies violate copyright law by training their models on publicly available content.
Ironically, Studio Ghibli co-founder Hayao Miyazaki is vocally anti-AI. While not specifically addressing the copyright debate, he referred to AI-generated artwork as “creepy stuff” and that he “would never wish to incorporate this technology into (his) work at all” in 2016, as shown in NHK’s “Never-Ending Man” documentary.
«
Earlier on Thursday there had been some chatter that the Ghibli-style stuff was OK (given it would have had to have come from vacuuming up entire films) because Japan has different copyright laws which do allow AI training. No confirmation of that has emerged, though, and the “living artist” block has come down.
unique link to this extract
Facebook launches Friends tab: a new feed that ditches algorithmic junk • The Verge
Emma Roth:
»
Facebook is trying to go back to its roots with a new “Friends” tab that filters out the algorithmic recommendations that have taken over its main feed. The new tab is rolling out in the US and Canada, and will show your friends’ stories, reels, posts, and birthdays.
In a blog post announcing the feature, the company says, “Over the years, Facebook evolved to meet changing needs and created best-in-class experiences across Groups, Video, Marketplace and more, but the magic of friends has fallen away.”
Facebook previously housed friend requests and friend suggestions within the Friends tab, but now it’s been replaced by a dedicated friends-only feed. I’m not sure I’ll find many recent posts there, but it might be a reprieve from the sponsored content and random group posts that have drowned out all other content on my home feed.
«
This is interesting: essentially it’s accepting that all the AI slop which has taken over the algorithmic feed is a big turnoff for a lot of people, and so this redesign is trying to make that go away, or at least offer an option that makes that vanish. The first sign that slop is properly detrimental to user experience and, more importantly, to time spent on the app, which is what Meta really notices.
unique link to this extract
You need to use Signal’s nickname feature • 404 Media
Joseph Cox:
»
According to screenshots of the chats and the group chat’s members published by The Atlantic on Wednesday, the outlet’s editor Jeffrey Goldberg used the display name “JG” on Signal. He also said in the original article that he displayed as JG. Presumably National Security Adviser Michael Waltz, who accidentally added Goldberg, added the wrong JG. This is a big, big mistake obviously.
But there is a somewhat overlooked setting inside Signal that can ensure you don’t make the same mistake. It’s the nickname feature.
…You can add your own nickname to a Signal contact by clicking on the person’s profile picture in a chat with them then clicking “Nickname.” Signal says “Nicknames & notes are stored with Signal and end-to-end encrypted. They are only visible to you.” So, you can add a nickname to a Jason saying “co-founder,” or maybe “national security adviser,” and no one else is going to see it. Just you. When you’re trying to make a group chat, perhaps.
«
Simple advice. One could probably charge a very high consultancy fee for doing this for, say, a government struggling with catastrophic information leakage.
unique link to this extract
Volvo’s heartbeat-sensing anti-carjacking technology was Y2K safety at its weirdest • Hagerty Media
Benjamin Hunting:
»
Every once in a while, an automaker debuts a feature that is truly unique, standing apart from everything else on the market. In some cases, these novel technologies serve as the thin end of a wedge that opens up a whole new world of vehicle development, with rivals quickly co-opting and branding their own version of the innovation.
In others, you end up standing in a parking lot in the year 2006, watching “Ray the Burglar” be apprehended by a security system that’s been specifically designed to detect his heartbeat as he hides in the backseat, like some digital combination of Sherlock Holmes and Robocop.
Such was the media introduction given to the most intriguing capability of Volvo’s Personal Car Communicator, a thin slab of plastic that initially served as a super-fob for the flagship S80 sedan. Billed as a bulwark against the threat of kidnappers, carjackers, and bloodthirsty murderers who were apparently lurking around every corner of a Volvo owner’s action-packed existence, the heartbeat sensor was an entirely singular high-tech achievement that no other car company ever came close to implementing in any of their vehicles.
What prompted Volvo, a brand known for sensibility and safety above all else, to infer that its customers were living a lifestyle that required this level of electronic overwatch to keep them from becoming just another statistic on a police blotter? And why didn’t the heartbeat sensor catch on long-term in the industry—or even with Volvo’s modern-day lineup? It turns out that the answers to those questions are tied directly to the times that produced this utterly unusual, and unforgettable, feature.
«
It’s a wild ride of a story. Odd how this technology hasn’t continued, eh. (Might there be people out there still using it?)
unique link to this extract
Six unsettling thoughts Google’s former CEO has about artificial intelligence • NPR
Steve Inskeep:
»
Eric Schmidt, the former Google CEO, is thinking about artificial intelligence – how it interacts with humans, and how it may reshape democracy. Or replace it.
Schmidt coauthored a new book, Genesis, with former Microsoft executive Craig Mundie and the late Henry Kissinger, who died in 2023 about a year before the book’s publication.
Kissinger, Schmidt says, had been thinking about the nature of reality “since before we were born,” and used some of his final years exploring how technology might warp our understanding of reality.
Genesis includes a story from history: the Spanish conquistadors who invaded present-day Mexico in 1519. The ruling Aztecs seem to have mistaken the newcomers for gods. Their emperor first met them, took their advice, and then became their hostage before the conquistadors simply took over.
That’s the unsettling start to a chapter that asks if AI might conquer us.
«
Those six thoughts:
• AI will be available to almost anyone
• AI can be a tool for demagogues
• People are interacting with tech they don’t fully grasp
• Tech leaders may not grasp the implications either
• People might allow themselves to be governed by AI
• The recent presidential inauguration showed a concentration of [tech] power.
Plenty of expansion in the article.
unique link to this extract
Turkish female coders leave tech industry over sexism • Rest of World
Kaya Genç:
»
Turkish women who become software engineers generally expect a secure career that can vault them to a better position in society. Instead, they’ve found they are a mistreated and undervalued minority. In Turkey’s tech industry, less than a quarter of workers are women, according to various surveys — lower than the global average of 28%. Female IT engineers have complained about male-dominated workplaces, with 70% saying it negatively affected their careers, according to a 2018 report by consulting firm Deloitte Turkey and the Information Technology Industrialists Association.
At the same time, the government is pushing for women to prioritize their families over work. Turkish President Recep Tayyip Erdoğan has said that women should have at least three children. In 2021, Turkey withdrew from the Istanbul Convention, which prevents and fights violence against women.
Rather than face the turbulent social conditions, many young IT professionals, particularly women, are choosing to leave the country, according to Füsun Sarp Nebil, a tech entrepreneur and CEO of the technology journalism website Turk Internet.
“The increasing pressure in recent years, the obstructions on social life, like the cancellation of concerts, the lack of freedom of expression, the environment of violence … disturbs young people and especially women,” Nebil told Rest of World.
“Salaries are at the bottom of the list of reasons for leaving Turkey,” she said.
«
Turkey withdrawing from the Istanbul Convention is quite the irony.
unique link to this extract
California now has 48% more EV chargers than gasoline nozzles in the state • Governor of California
California Governor’s office:
»
Governor Gavin Newsom has announced that California has reached 178,549 public and shared private electric vehicle (EV) chargers installed statewide.
In a significant milestone for the state, California now has 48% more public and shared private EV chargers than the number of gasoline nozzles. The California Energy Commission (CEC) estimates there are about 120,000 gas nozzles in the state, compared to 178,000 public and shared private chargers.
Governor Gavin Newsom said: “As the federal government works to make it harder for you to charge your electric car, California is doing the opposite. We now have nearly 50% more chargers than gas nozzles in the state, meaning you have more options than ever to charge your vehicle.
“We’re embracing our clean car future and providing consumers more choices – no matter what ‘big government’ mandates come out of Washington.”
Out of the more than 178,000 EV chargers installed in the state, over 162,000 are Level 2 chargers, and nearly 17,000 are fast chargers. In addition to the public network, the CEC estimates that more than 700,000 Level 2 chargers are installed statewide in single-family homes.
“The California EV driver experience is getting better by the day,” said CEC Chair David Hochschild. “The state continues to invest in EV infrastructure, with particular emphasis in hard-to-reach areas, making these vehicles an easy choice for new car buyers.”
California dominates in zero-emission vehicle (ZEV) infrastructure efforts, dedicating billions to support clean transportation goals. With more EVs on the road every day, consumers are responding to the state’s efforts to build a bigger, better, and more reliable charging network.
«
California has about 13.2m vehicles, of which at December 2023 – the last data update – a total of 1.8m are electric. Probably moved on a fair bit since then.
unique link to this extract
Carbon credit auditors suspended in sham rice-farming offsets • Climate Change News
Matteo Civillini:
»
Carbon credit registry Verra has suspended activities by four auditors related to carbon credit projects they vetted in China which claimed bogus emission reductions.
In an unprecedented move, TÜV Nord, China Classification Society Certification Company, China Quality Certification Center and CTI Certification will be prevented from auditing agriculture and forestry offsetting schemes on Verra’s registry. For German certification giant TÜV Nord, the measures will only apply to its operations in China. It is the first time Verra has taken such measures.
The auditors certified the activities of 37 programmes that aimed to slash planet-heating methane gas releases from rice fields across China, resulting in the generation of millions of carbon offsets. But Verra revoked the projects in August 2024 after a 17-month review found a string of integrity failures that the auditors had failed to identify.
Before this week’s suspension, Climate Home previously reported on ten of these projects closely linked to energy company Shell and revealed evidence raising serious doubts over whether any emission-cutting activities had been carried out on the ground at all.
Nearly two million worthless carbon credits produced by the projects – and partly used to offset emissions from Shell’s gas business – still need to be compensated.
«
So not only are the carbon credit schemes often a scam, the auditors do a poor job.
unique link to this extract
Paralysed man stands again after receiving “reprogrammed” stem cells • Nature
Smriti Mallapaty:
»
A paralysed man can stand on his own after receiving an injection of neural stem cells to treat his spinal cord injury. The Japanese man was one of four individuals in a first-of-its-kind trial that used reprogrammed stem cells to treat people who are paralysed.
Another man can now move his arms and legs following the treatment, but the two others did not show substantial improvements. The trial was run by Hideyuki Okano, a stem-cell scientist at Keio University in Tokyo, and his colleagues.
The results, which were announced at a press conference on 21 March and have not yet been peer reviewed, suggest that the treatment is safe, say researchers.
“That’s a great positive outcome. It’s very exciting for the field,” says James St John, a translational neuroscientist at Griffith University in the Gold Coast, Australia.
Previous trials using other types of stem cell have also demonstrated that the therapy is safe, but have so far shown mixed results. “Nothing’s really worked so far,” says St John.
Larger trials will be needed to establish whether the improvements observed in the two individuals in the current study were a result of the treatment. It’s possible the patients experienced a natural recovery, says St John.
In 2019, roughly 0.9 million people globally experienced a spinal cord injury, and some 20 million people were living with the condition1.
«
Wait for the peer review! But being able to use arms and legs is quite a peer review in itself. Perhaps there’s a breakthrough just coming here.
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
The Overspill: when there's more that I want to say 