Home
Member since 2017-07-15T03:50:57Z. Last seen 2025-06-14T19:39:11Z.
2755 blog posts. 128 comments.
America’s most consequential technology
SHIRA OVIDE
The Washington Post 12 Apr 2024 App is the internet’s healthiest economy and No. 1 for music, videos
(Spoiler alert: It’s Youtube.) You think you know Youtube. It’s where billions of people learn how to change a tire, follow a favorite yoga workout or catch footage of Monday’s solar eclipse.
But maybe you don’t know that Youtube is also the most popular way to hear music and one of the country’s largest cable TV providers. Youtube is the healthiest economy on the internet. And it has been rocket fuel for artificial intelligence.
I’m digging into Youtube’s identity because it’s essential to understand the influence of technologies in our lives. As popular as Youtube is, its power over the internet and us is somehow still underrated.
Let me try to persuade you that Youtube is the most consequential technology in America:
Youtube is No. 1 in video and music listening
Youtube might be best known for wasting a few minutes at a supermarket checkout or on the toilet. (Sorry.)
Youtube, however, is also America’s top living room streaming destination. Data from Nielsen consistently shows that Americans spend more time watching Youtube on TV sets than we do any streaming service including Netflix.
There isn’t reliable data counting the combined time we stream on TV, phones, computers and other devices, but Youtube would likely be tops on that measure, too.
Youtube TV, which is like cable TV but accessed over the internet, is also now one of the leading cable TV providers in the country.
In music, more people tune into songs on Youtube than we do on Spotify, the radio or any other audio service.
In a survey of people in several countries by Mark Mulligan of MIDIA Research, about twothirds of respondents watched music videos on Youtube. About 43 percent listened to music online another way and 31 percent have a streaming music subscription like those from Spotify and Apple Music.
P.S. The most widely used social app among American adults is Youtube by a mile, Pew Research Center says. The most widely used app among teens isn’t Tiktok. It’s Youtube again by a mile.
It’s the healthiest economy on the internet
If you post on Instagram, Facebook, Tiktok, Reddit or X, you are basically making those companies’ products for free. Youtube doesn’t work that way.
From each dollar that advertisers pay for commercials on many millions of Youtube videos, the person who made the video gets 55 cents. Google, which owns Youtube, keeps the rest.
Youtube has had this financial arrangement for close to 20 years. Still today, no other large app has such a consistent way for people to earn income from what they create and post online.
Youtube’s revolutionary payment system matters to you even if you never earn a dime from making a Youtube video.
A healthy internet economy, like a well-functioning United States economy, is one in which everyone believes they have a shot to thrive. That includes you as the viewer, the people making the information or entertainment you’re watching and the companies distributing the material.
Youtube is far from perfect on this score, but it may be the closest thing to the financially ideal online economy.
By the way, if you buy a subscription to Youtube Premium, which lets you watch videos without ads, Youtube hands over portions of your money to the video makers, in proportion to your viewing time.
If you watch a lot of videos from Mrbeast and Not Just Bikes, those Youtube channels will receive a large chunk of your subscription money. The relatively democratic system to pay the people making stuff is not how most music services like Spotify or Netflix work.
Even if you only listen to jazz music on Spotify, Taylor Swift will still get a bunch of your subscription money.
Youtube is essential, controversial AI data fuel
You know an app is important when it becomes a wrestling match for companies grabbing every morsel of data to “train” their AI.
According to the New York Times, CHATGPT owner Openai invented a way to suck up more than 1 million hours of Youtube videos and podcasts and turned the spoken words into fuel to coach its AI. Google has also transcribed Youtube videos to train its AI software, the Times reported.
What Openai did might violate Youtube’s terms of service and what Google did might violate the copyright of people who make Youtube videos, the Times reported.
Openai said it uses “numerous sources including publicly available data and partnerships for nonpublic data.” Google reiterated the Youtube CEO’S recent comment that the company’s AI is trained with some Youtube material, “in accordance with our agreements with Youtube creators.”
Youtube is definitely flawed. It’s been used to mislead and harass people and to spread propaganda. But for good or for ill, Youtube matters even more than you probably think.
Shared via PressReader
connecting people through news
Help wanted: Intelligent humans to tutor chatbotsSAN FRANCISCO BY YIWEN LU
The New York Times (International Edition) 12/4/2024 The work of teaching A.I. is getting more complex as the technology improves
After her second child was born, Chelsea Becker took an unpaid, yearlong leave from her full-time job as a flight attendant. After watching a video on TikTok, she found a side hustle: training artificial intelligence models for a website called Data Annotation Tech.
For a few hours every day, Ms. Becker, 33, who lives in Schwenksville, Pa., would sit at her laptop and interact with an A.I.-powered chatbot. For every hour of work, she was paid $20 to $40. From December to March, she made over $10,000.
The boom in A.I. technology has put a more sophisticated spin on a kind of gig work that doesn’t require leaving the house. The growth of large language models like the technology powering OpenAI’s ChatGPT has fueled the need for trainers like Ms. Becker, fluent English speakers who can produce quality writing.
It is not a secret that A.I. models learn from humans. For years, makers of A.I. systems like Google and OpenAI have relied on low-paid workers, typically contractors employed through other companies, to help computers visually identify subjects. (The New York Times has sued OpenAI and its partner, Microsoft, on claims of copyright infringement.) They might label vehicles and pedestrians for self-driving cars or identify images on photos used to train A.I. systems.
But as A.I. technology has become more sophisticated, so has the job of people who must painstakingly teach it. Yesterday’s photo tagger is today’s essay writer.
There are usually two types of work for these trainers: supervised learning, where the A.I. learns from human-generated writing, and reinforcement learning from human feedback, where the chatbot learns from how humans rate their responses.
Companies that specialize in data curation, including the San Franciscobased
start-ups Scale AI and Surge AI, hire contractors and sell their training data to bigger developers. Developers of A.I. models, such as the Toronto-based start-up Cohere, also recruit in-house data annotators.
It is difficult to estimate the total number of these gig workers, researchers said. But Scale AI, which hires contractors through its subsidiaries, Remotasks and Outlier, said it was common to see tens of thousands of people working on the platform at a given time.
But as with other types of gig work, the ease of flexible hours comes with its own challenges. Some workers said they never interacted with administrators behind the recruitment sites, and others had been cut off from the work with no explanation. Researchers have also raised concerns over a lack of standards, since workers typically don’t receive training on what are considered to be appropriate chatbot answers.
To become one of these contractors, workers have to pass an assessment, which includes questions like whether a social media post should be considered hateful, and why. Another one requires a more creative approach, asking contracting prospects to write a fictional short story about a green dancing octopus, set in Sam Bankman-Fried’s FTX offices on Nov. 8, 2022. (That was the day Binance, an FTX competitor, said it would buy Mr. Bankman-Fried’s company before later quickly backing out of the deal.)
Sometimes, companies look for subject matter experts. Scale AI has posted jobs for contract writers who hold master’s or doctoral degrees in Hindi and Japanese. Outlier has job listings that mention requirements like academic degrees in math, chemistry and physics.
“What really makes the A.I. useful to its users is the human layer of data, and that really needs to be done by smart humans and skilled humans and humans with a particular degree of expertise and a creative bent,” said Willow Primack, vice president of data operations at Scale AI. “We have been focusing on contractors, particularly within North America, as a result.”
Alynzia Fenske, a self-published fiction writer, had never interacted with an A.I. chatbot before hearing a lot from fellow writers who considered A.I. a threat.
So when she came across a video on TikTok about Data Annotation Tech, part of her motivation was just to learn as much about A.I. as she could and see for herself whether the fears surrounding A.I. were warranted.
“It’s giving me a whole different view of it now that I’ve been working with it,” said Ms. Fenske, 28, who lives in Oakley, Wis. “It is comforting knowing that there are human beings behind it.” Since February, she has been aiming for 15 hours of data annotation work every week so she can support herself while pursuing a writing career.
Ese Agboh, 28, a master’s student studying computer science at the University of Arkansas, was given the task of coding projects, which paid $40 to $45 an hour. She would ask the chatbot to design a motion sensor program that helps gymgoers count their repetitions, and then evaluate the computer codes written by the A.I. In another case, she would load a data set about grocery items to the program and ask the chatbot to design a monthly budget. Sometimes she would even evaluate other annotators’ codes, which experts said are used to ensure data quality.
She made $2,500. But her account was permanently suspended by the platform for violating its code of conduct. She did not receive an explanation, but she suspected that it was because she worked while in Nigeria, since the site wanted workers based in only certain countries.
That is the fundamental challenge of online gig work: It can disappear at any time. With no one available for help, frustrated contractors turned to social media, sharing their experiences on Reddit and TikTok. Jackie Mitchell, 26, gained a large following on TikTok because of her content on side hustles, including data annotation work.
“I get the appeal,” she said, referring to side hustles as an “unfortunate necessity” in this economy and “a hallmark of my generation and the generation above me.”
Public records show that Surge AI owns Data Annotation Tech. Neither the company nor its chief executive, Edwin Chen, responded to requests for comments.
It is common for companies to hire contractors through subsidiaries. They do so to protect the identity of their customers, and it helps them avoid bad press associated with working conditions for its low-paid contract workers, said James Muldoon, a University of Essex management professor whose research focuses on A.I. data work.
A majority of today’s data workers depend on wages from their gig work. Milagros Miceli, a sociologist and computer scientist researching labor conditions in data work, said that while “a lot of people are doing this for fun, because of the gamification that comes with it,” a bulk of the work is still “done by workers who actually really need the money and do this as a main income.”
Researchers are also concerned about the lack of safety standards in data labeling. Workers are sometimes asked to address sensitive issues like whether certain events or acts should be considered genocide or what gender should appear in an A.I.-generated image of a soccer team, but they are not trained on how to make that evaluation.
“It’s fundamentally not a good idea to outsource or crowdsource concerns about safety and ethics,” Professor Muldoon said. “You need to be guided by principles and values, and what your company actually decides as the right thing to do on a particular issue.”
她的第二个孩子出生后，切尔西·贝克尔（Chelsea Becker）从她作为空姐的全职工作中休了一年的无薪假。在 TikTok 上观看了一段视频后，她找到了一个副业：为一个名为 Data Annotation Tech 的网站训练人工智能模型。 住在宾夕法尼亚州施文克斯维尔的33岁的贝克尔每天有几个小时坐在笔记本电脑前，与人工智能聊天机器人互动。每工作一小时，她就能得到20到40美元的报酬。从 12 月到 3 月，她赚了超过 10,000 美元。 人工智能技术的蓬勃发展为一种不需要离开家的零工工作带来了更复杂的旋转。大型语言模型的发展，如为 OpenAI 的 ChatGPT 提供动力的技术，推动了对像贝克尔女士这样的培训师的需求，他们能说流利的英语，可以写出高质量的文章。 人工智能模型向人类学习已经不是什么秘密了。多年来，谷歌和OpenAI等人工智能系统的制造商一直依靠低薪工人（通常是通过其他公司雇用的承包商）来帮助计算机直观地识别受试者。（《纽约时报》以侵犯版权为由起诉OpenAI及其合作伙伴Microsoft。他们可能会为自动驾驶汽车标记车辆和行人，或者识别用于训练人工智能系统的照片上的图像。
但是，随着人工智能技术变得越来越复杂，必须煞费苦心地教授它的人的工作也越来越复杂。昨天的照片标记者是今天的文章作者。 这些培训师通常有两种类型的工作：监督学习，人工智能从人类生成的写作中学习，以及从人类反馈中学习强化学习，聊天机器人从人类如何评价他们的回答中学习。 专门从事数据管理的公司，包括总部位于旧金山的公司 初创企业 Scale AI 和 Surge AI，雇用承包商并将其训练数据出售给更大的开发人员。人工智能模型的开发者，如多伦多的初创公司Cohere，也招募了内部数据注释员。 研究人员说，很难估计这些零工工人的总数。但通过其子公司Remotasks和Outlier雇用承包商的Scale AI表示，在特定时间看到成千上万的人在平台上工作是很常见的。 但与其他类型的零工工作一样，灵活工作时间的便利性也有其自身的挑战。一些工人说，他们从未与招聘网站背后的管理人员互动，其他人则被切断了工作，没有任何解释。研究人员还对缺乏标准表示担忧，因为工人通常没有接受过关于什么是适当的聊天机器人答案的培训。 要成为这些承包商之一，工人必须通过评估，其中包括社交媒体帖子是否应被视为仇恨以及为什么。另一个需要更具创造性的方法，要求签约潜在客户写一个虚构的短篇小说，讲述一只绿色跳舞的章鱼，故事发生在 2022 年 11 月 8 日 Sam Bankman-Fried 的 FTX 办公室。（就在这一天，FTX 的竞争对手币安表示将收购班克曼-弗里德的公司，但后来迅速退出了这笔交易。 有时，公司会寻找主题专家。Scale AI 已经为拥有印地语和日语硕士或博士学位的合同作家发布了工作。Outlier 的职位列表提到了数学、化学和物理学位等要求。 “真正让人工智能对用户有用的是人类的数据层，这确实需要由聪明的人类和熟练的人类以及具有特定程度的专业知识和创造性倾向的人来完成，”Scale AI数据运营副总裁Willow Primack说。“因此，我们一直专注于承包商，尤其是在北美地区。” 阿林齐亚·芬斯克（Alynzia Fenske）是一位自我出版的小说作家，她从未与人工智能聊天机器人互动过，然后从其他作家那里听到了很多将人工智能视为威胁的声音。 因此，当她在 TikTok 上看到一个关于数据注释技术的视频时，她的部分动机只是尽可能多地了解人工智能，并亲眼看看围绕人工智能的恐惧是否合理。 “现在我一直在和它打交道，这让我对它有了完全不同的看法，”住在威斯康星州奥克利的28岁的芬斯克说，“知道背后有人，我感到很欣慰。自 2 月以来，她的目标是每周进行 15 小时的数据注释工作，这样她就可以在追求写作事业的同时养活自己。 28岁的Ese Agboh是阿肯色大学（University of Arkansas）计算机科学专业的硕士生，他被赋予了编码项目的任务，每小时支付40至45美元。她会要求聊天机器人设计一个运动传感器程序，帮助健身者计算他们的重复次数，然后评估人工智能编写的计算机代码。在另一种情况下，她会将有关杂货的数据集加载到程序中，并要求聊天机器人设计每月预算。有时，她甚至会评估其他注释者的代码，专家说这些代码是用来确保数据质量的。 她赚了 2,500 美元。但她的账户因违反其行为准则而被平台永久暂停。她没有得到解释，但她怀疑这是因为她在尼日利亚工作，因为该网站只需要某些国家的工人。 这是在线零工工作的根本挑战：它随时可能消失。由于没有人可以提供帮助，沮丧的承包商转向社交媒体，在 Reddit 和 TikTok 上分享他们的经验。26 岁的杰基·米切尔 （Jackie Mitchell） 在 TikTok 上获得了大量追随者，因为她的副业内容包括数据注释工作。 “我明白这种吸引力，”她说，并称副业是这种经济中“不幸的必需品”，也是“我这一代人和我之上的一代人的标志”。 公开记录显示，Surge AI 拥有 Data Annotation Tech。该公司及其首席执行官埃德温·陈（Edwin Chen）均未回应置评请求。 公司通过子公司雇用承包商是很常见的。他们这样做是为了保护客户的身份，并帮助他们避免与低薪合同工的工作条件相关的负面新闻，埃塞克斯大学（University of Essex）管理学教授詹姆斯·马尔登（James Muldoon）说，他的研究重点是人工智能数据工作。 如今，大多数数据工作者都依赖零工工作的工资。研究数据工作劳动条件的社会学家和计算机科学家米拉格罗斯·米塞利（Milagros Miceli）表示，虽然“很多人这样做是为了好玩，因为随之而来的游戏化”，但大部分工作仍然是“由真正需要钱并将其作为主要收入的工人完成的。 研究人员还担心数据标签缺乏安全标准。工作人员有时会被要求解决敏感问题，例如某些事件或行为是否应该被视为种族灭绝，或者人工智能生成的足球队图像中应该出现什么性别，但他们没有接受过如何进行评估的培训。 “从根本上说，将对安全和道德的担忧外包或众包不是一个好主意，”Muldoon教授说。“你需要以原则和价值观为指导，以及你的公司在特定问题上实际决定的正确做法。”
通過分享 PressReader
通過新聞連接世界
A.I. JOBS IN THE GIG ECONOMY
New York Times (International Edition) 12/4/2024
Pulling all-nighters to assemble PowerPoint presentations. Punching numbers into Excel spreadsheets. Finessing the language on esoteric financial documents that may never be read by another soul.
Such grunt work has long been a rite of passage in investment banking, an industry at the top of the corporate pyramid that lures thousands of young people every year with the promise of prestige and pay.
Until now. Generative artificial intelligence — the technology upending many industries with its ability to produce and crunch new data — has landed on Wall Street. And investment banks, long inured to cultural change, are rapidly turning into Exhibit A on how the new technology could not only supplement but supplant entire ranks of workers.
The jobs most immediately at risk are those performed by analysts at the bottom rung of the investment banking business, who put in endless hours to learn the building blocks of corporate finance, including the intricacies of mergers, public offerings and bond deals. Now, A.I. can do much of that work speedily and with considerably less whining.
“The structure of these jobs has remained largely unchanged at least for a decade,” said Julia Dhar, head of BCG’s Behavioral Science Lab and a consultant to major banks experimenting with A.I. The inevitable question, as she put it, is “do you need fewer analysts?”
Some of Wall Street’s major banks are asking the same question, as they test A.I. tools that can largely replace their armies of analysts by performing in seconds the work that now takes hours, or a whole weekend. The software, being deployed inside banks under code names such as “Socrates,” is likely not only to change the arc of a Wall Street career, but also to essentially nullify the need to hire thousands of new college graduates.
Top executives at Goldman Sachs, Morgan Stanley and other banks are debating how deep they can cut their incoming analyst classes, according to several people involved in the ongoing discussions. Some inside those banks and others have suggested they could cut back on their hiring of junior investment banking analysts by as much as two-thirds, and slash the pay of those they do hire, on the grounds that the jobs won’t be as taxing as before.
“The easy idea,” said Christoph Rabenseifner, Deutsche Bank’s chief strategy officer for technology, data and innovation, “is you just replace juniors with an A.I. tool,” although he added that human involvement will remain necessary.
Representatives for Goldman, Morgan Stanley, Deutsche Bank and others said it was too early to comment on specific job changes. But the consulting giant Accenture estimated that A.I. could replace or supplement nearly threequarters of bank employees’ working hours across the industry.
Goldman is “experimenting with the technology,” said Nick Carcaterra, a bank spokesman. “In the near term, we anticipate no changes to our incoming analyst classes.”
This week, JPMorgan Chase’s chief executive, Jamie Dimon, wrote in his annual shareholder letter that A.I. “may reduce certain job categories or roles,” and labeled the technology top among the most important issues facing the United States’ largest bank. Mr. Dimon compared the consequences to those of “the printing press, the steam engine, electricity, computing and the internet, among others.”
Investment banking is a hierarchical industry, and banks typically hire young talent through two-year analyst contracts. Tens of thousands of 20-somethings (both from undergraduate and M.B.A. programs) apply for some 200 spots in each major bank’s program. Pay starts at more than $100,000, not including year-end bonuses.
If they persevere, they move up the ranks to associate, then director and managing director; a handful end up running divisions. Although grueling, the life of a senior banker can be glamorous, involving traveling around the globe to pitch clients and working on big-money corporate merger deals. Many who get through the two-year analyst program have gone on to become business titans — the billionaires Michael Bloomberg and Stephen Schwarzman began their careers in investment banking — but a majority will leave before or after their two years are up, bank representatives said.
There are jokes among junior bankers that the most common tasks of the job involve dragging icons from one side of a document to another, only to be asked to replace the icon over and again.
“One hundred percent drudgery and boring,” said Gabriel Stengel, a former banking analyst who left the industry two years ago. Val Srinivas, a senior researcher for banking at Deloitte, said a lot of the work involved “gathering material, poring through it and putting it through a different format.”
Gregory Larkin, another former banking analyst, said the new technology would start “a civil war” inside Wall Street’s biggest firms by tilting the balance of power to technologists who program A.I. tools, as opposed to the bankers who use them — to say nothing of technology giants like Microsoft and Google, which license much of the A.I. technology to banks for hefty fees.
“A.I. will enable us to do tasks that take 10 hours in 10 seconds,” said Jay Horine, co-head of investment banking at JPMorgan, describing analyst jobs. “My hope and belief is it will allow the job to be more interesting.”
A.I.’s impact on finance is simply one facet of how the technology will reshape the workplace for all. Artificial intelligence systems, which include large language models and question-and-answer bots like ChatGPT, can quickly synthesize information and automate tasks. Virtually all industries are beginning to grapple with it to some degree.
Deutsche Bank is uploading reams of financial data into proprietary A.I. tools that can instantaneously answer questions about publicly traded companies and create summary documents on complementary financial moves that might benefit a client.
Mr. Horine said he could use A.I. to identify clients that might be ripe for a bond offering, the sort of bread-and-butter transaction for which investment bankers charge clients millions of dollars.
Goldman Sachs has assigned 1,000 developers to test A.I., including software that can turn what it terms “corpus” information — or enormous amounts of text and data collected from thousands of sources — into page presentations that mimic the bank’s typeface, logo, styles and charts. One firm executive privately called it a “Kitty Hawk moment,” or one that would change the course of the firm’s future.
That isn’t limited to investment banking; BNY Mellon’s chief executive said on a recent earnings call that his research analysts could now wake up two hours later than usual, because A.I. can read overnight economic data and create a written draft of analysis to work from.
Morgan Stanley’s head of technology, Michael Pizzi, told employees in a January private meeting, a video of which was viewed by The New York Times, that he would “get A.I. into every area of what we do,” including wealth management, where the bank employs thousands of people to determine the proper mix of investments for well-off savers.
Many of those tools are still in the testing phase, and will need to be run past regulators before they can be deployed at scale on live work. Bank of America’s chief executive said last year that the technology was already enabling the firm to hire less.
Among Goldman Sachs’s sprawling A.I. efforts is a tool under development that can transfigure a lengthy PowerPoint document into a formal “S-1,” the legalese-packed document for initial public offerings required for all listed companies.
The software takes less than a second to complete the job.
通過分享 PressReader
通過新聞連接世界
HOW EXERCISE BOOSTS BRAIN
New York Times (International Edition) 12/4/2024
Growing up in the Netherlands, Henriette van Praag had always been active, playing sports and riding her bike to school every day. Then, in the late-1990s, while working as a staff scientist at the Salk Institute for Biological Studies in San Diego, she discovered that exercise can spur the growth of new brain cells in mature mice. After that, her approach to exercise changed.
“I started to take it more seriously,” said Dr. van Praag, now an associate professor of biomedical science at Florida Atlantic University. Today, that involves doing CrossFit and running five or six miles several days a week.
Whether exercise can cause new neurons to grow in adult humans — a feat previously thought impossible, and a tantalizing prospect to treat neurodegenerative diseases — is still up for debate.
But even if it’s not possible, physical activity is excellent for your brain, improving mood and cognition through “a plethora” of cellular changes, Dr. van Praag said.
WHAT ARE SOME OF THE BENEFITS?
Exercise offers short-term boosts in cognition. Studies show that immediately after a bout of physical activity, people perform better on tests of working memory and other executive functions. This may be in part because movement increases the release of neurotransmitters in the brain, most notably epinephrine and norepinephrine.
“These kinds of molecules are needed for paying attention to information,” said Marc Roig, an associate professor in the School of Physical and Occupational Therapy at McGill University. Attention is essential for working memory and executive functioning, he added.
The neurotransmitters dopamine and serotonin are also released with exercise, which is thought to be a main reason people often feel so good after going for a run or a long bike ride. The brain benefits really start to emerge, though, when we work out consistently over time. Studies show that people who work out several times a week have higher cognitive test scores, on average, than people who are more sedentary. Other research has found that a person’s cognition tends to improve after participating in a new aerobic exercise program for several months.
Dr. Roig added the caveat that the effects on cognition aren’t huge, and not everyone improves to the same degree. “You cannot acquire a super memory just because you exercised,” he said.
Physical activity also benefits mood. People who work out regularly report having better mental health than people who are sedentary. And exercise programs can be effective at treating people’s depression, leading some psychiatrists and therapists to prescribe physical activity.
The Centers for Disease Control and Prevention’s recommendation of 150 minutes of moderate aerobic activity or 75 minutes of vigorous aerobic activity per week is a good benchmark.
Perhaps most remarkable, exercise offers protection against neurodegenerative diseases. “Physical activity is one of the health behaviors that’s shown to be the most beneficial for cognitive function and reducing risk of Alzheimer’s and dementia,” said Michelle Voss, an associate professor of psychological and brain sciences at the University of Iowa.
HOW DOES EXERCISE DO ALL THAT?
It starts with the muscles. When we work out, they release molecules that travel through the blood up to the brain. Some, like a hormone called irisin, have “neuroprotective” qualities and have been shown to be linked to the cognitive health benefits of exercise, said Christiane Wrann, an associate professor of medicine at Massachusetts General Hospital and Harvard Medical School who studies irisin. (Dr. Wrann is also a consultant for a pharmaceutical company, Aevum Therapeutics, hoping to harness irisin’s effects into a drug.)
Good blood flow is essential to obtain the benefits of physical activity. And conveniently, exercise improves circulation and stimulates the growth of new blood vessels in the brain.
“It’s not just that there’s increased blood flow,” Dr. Voss said. “It’s that there’s a greater chance, then, for signaling molecules that are coming from the muscle to get delivered to the brain.”
Once these signals are in the brain, other chemicals are released locally. The star of the show is a hormone called brain-derived neurotrophic factor, or B.D.N.F., that is essential for neuron health and creating new connections — called synapses — between neurons.
“It’s like a fertilizer for brain cells to recover from damage,” Dr. Voss said. “And also for synapses on nerve cells to connect with each other and sustain those connections.”
A greater number of blood vessels and connections between neurons can actually increase the size of different brain areas. This effect is especially noticeable in older adults because it can offset the loss of brain volume that happens with age. The hippocampus, an area important for memory and mood, is particularly affected. “We know that it shrinks with age,” Dr. Roig said. “And we know that if we exercise regularly, we can prevent this decline.”
Exercise’s effect on the hippocampus may be one way it helps protect against Alzheimer’s disease, which is associated with significant changes to that part of the brain. The same goes for depression; the hippocampus is smaller in people who are depressed, and effective treatments for depression increase the size of the region.
WHAT KIND OF EXERCISE IS BEST?
The experts emphasized that any exercise is good, and the type of activity doesn’t seem to matter, though most of the research has involved aerobic exercise. But, they added, higher-intensity workouts do appear to confer a bigger benefit for the brain.
Improving your overall cardiovascular fitness level also appears to be key. “The more you can improve your cardiorespiratory fitness, the better the benefits are,” Dr. Wrann said.
Like Dr. van Praag, Dr. Voss has incorporated her research into her life, making a concerted effort to engage in higher-intensity exercise. For example, on busy days when she can’t fit in a full workout, she’ll seek out hills to bike up on her commute to work. “Even if it’s a little,” she said, “it’s still better than nothing.”
通過分享 PressReader
通過新聞連接世界
26-year-old works 20 minutes a day, brings in $462,000 a year from side hustle: I’m ‘doing less’ and ‘making more’ than ever
Francisco Rivera's side hustle selling candles on Etsy helped him quit his tutoring job last year. This story is part of CNBC Make It’s Six-Figure Side Hustle series, where people with lucrative side hustles break down the routines and habits they’ve used to make money on top of their full-time jobs. Got a story to tell? Let us know! Email us at AskMakeIt@cnbc.com.
Francisco Rivera doesn’t even like candles — but he brings in six figures per year selling them on Etsy.
In February 2023, Rivera was living in Orlando, Florida and working part-time for online tutoring company Outschool. Demand dropped when after-school activities resumed post-Covid, so he started looking for more income elsewhere.
He found a YouTube video about print-on-demand side hustles, where sellers create designs for products like T-shirts or mugs. They list their designs on marketplaces like Etsy or Amazon, and when a customer places an order, a manufacturer prints the design onto the product and ships it out.
For his product, Rivera chose neutral-colored organic candles with “witty” labels, he says. He creates his designs on Canva, lists them on Etsy and uses a service called Printify to connect with manufacturers.
DON’T MISS: The ultimate guide to earning passive income online
His Etsy shop brought in approximately $462,000 in sales last year, according to documents reviewed by CNBC Make It — enough for him to quit his tutoring job in December 2023. (Rivera says he’d prefer not to name his shop, to prevent potential copycats.)
About 30% to 50% of each sale is profit, Rivera estimates. His expenses include Etsy fees, nearly $55,000 last year, and money he spends on marketing and Printify’s services.
Often, he works only 20 minutes per day, he says. Some days, he works extra: up to two hours, researching trends and designing new candle labels. With the rest of his time, he’s pursuing a music career, he adds.
“I’m making more than I ever have, doing less than I ever have,” says Rivera, 26.
Here, Rivera discusses the side hustle advice he thinks actually works, the biggest downside to his print-on-demand gig and why — despite his distaste for them — he chose to sell candles.
CNBC Make It: Do you think your side hustle is replicable?
Rivera: Absolutely. The beauty of the [print-on-demand] model is it’s so low-risk. It’s $0.20 to list something on Etsy. I borrow someone else’s Canva account, but the Pro version costs $120.
I don’t think I’m special — I just work hard. There’s value in time and value in flexibility. I would take a pay cut if it still allowed me to do what I’m doing [outside of my Etsy shop].
There are so many people I know who are interested in this, but just can’t start. I always say: If you have a 9-to-5, you’re putting in work and you already are consistent. You just have to channel that consistency toward something else.
A lot of print-on-demand businesses sell items like T-shirts or mugs. Why did you choose candles?
I’m not super passionate about selling candles. I’m actually allergic to them.
But at the time, candles were a newer category in print-on-demand. After scouring YouTube and Printify’s product catalog, I liked the idea of coming up with witty phrases to put on a product, and I noticed a lot of people were already selling apparel and mugs.
It felt like there was more opportunity with candles. They make great gifts, a lot of people buy candles on Etsy and people who had funny candle shops typically went viral within a year.
What’s the biggest downside to your side hustle?
The biggest downside of this side hustle is copycats — people who use the exact same phrases or very similar designs. They see bestselling candles, replicate them and then skip [to the top of search results]. I have to file copyright infringement, and it’s a mess.
Etsy is very secretive about its algorithm [for search results]. You can revise things, like the images on your listing or use different words on your product descriptions.
I don’t find a huge amount of success changing those things, so I would rather focus on pushing out new candles.
Can you share a piece of side hustle advice that you think is overrated?
A lot of people recommend sales analytic tools that show you demand versus competition [customer searches for your product versus similar products].
I’m not fully convinced they help. Instead, I’d tell people it’s important to not be married to your creativity. If there’s a design in your shop you really like, but it’s not really getting the results you want, cut it.
Research what other products are selling well. Put your own spin on it.
Want to make extra money outside of your day job? Sign up for CNBC’s new online course How to Earn Passive Income Online to learn about common passive income streams, tips to get started and real-life success stories. Register today and save 50% with discount code EARLYBIRD.
Plus, sign up for CNBC Make It’s newsletter to get tips and tricks for success at work, with money and in life.
CEO says AI needs energy efficiency
Wall Street Journal 10 Apr 2024
TOKYO—Chip-design company Arm made its name by devising ways to minimize smartphones’ power consumption and extend battery life. Now, the company’s head says the same push for energy efficiency is needed in artificial-intelligence applications.
Rene Haas, chief executive of Arm, spoke ahead of an announcement Tuesday by the U.S. and Japan about a $110 million program to fund AI research at universities in the two countries. U.K.-based Arm and its parent, Tokyo-based SoftBank Group, are together offering $25 million in funding for the program.
AI models such as OpenAI’s ChatGPT “are just insatiable in terms of their thirst” for electricity, Haas said in an interview. “The more information they gather, the smarter they are, but the more information they gather to get smarter, the more power it takes.” Without greater efficiency, “by the end of the decade, AI data centers could consume as much as 20% to 25% of U.S. power requirements. Today that’s probably 4% or less,” he said. “That’s hardly very sustainable.”
The power issue has drawn growing attention from technology executives in recent months and helped drive up the stock prices of companies that own and operate electricpower plants.
In a January report, the International Energy Agency said a request to ChatGPT requires 2.9 watt-hours of electricity on average—equivalent to turning on a 60-watt lightbulb for just under three minutes. That is nearly 10 times as much as the average Google search. The agency said power demand by the AI industry is expected to grow by at least 10 times between 2023 and 2026.
“It’s going to be difficult to accelerate the breakthroughs that we need if the power requirements for these large data centers for people to do research on keeps going up and up and up,” Haas said.
He expressed hope that the U.S.-Japan research partnership would work on solutions to the power issue. Arm’s funding is going to a collaboration between Carnegie Mellon University in Pittsburgh and Keio University in Japan.
The partnership is a project of the U.S. ambassador to Japan, Rahm Emanuel, who last year organized a $150 million U.S.-Japan research program in quantum computing backed by IBM and Google. Elsewhere in the AI partnership, Amazon.com and Nvidia are set to spend $25 million each to support AI research at the University of Washington and Japan’s University of Tsukuba.
Emanuel said it was important for the two allies to work together in the face of competing AI and quantum-computing research in China. He said the U.S. and Japan would harness the efforts of industry and academia together, contrasting that with Chinese leader Xi Jinping’s moves in recent years to rein in the country’s tech giants.
“One of the downsides of what Xi has done is squash entrepreneurship,” Emanuel said. “Our model is going to be more successful.”
Liu Pengyu, spokesman at the Chinese Embassy in Washington, called the remarks irresponsible and said the Chinese government attaches importance to entrepreneurship and innovation and has taken steps to shore up the private sector. He said technology cooperation between countries should contribute to stability and open trade “instead of targeting or undercutting the interests of a third party.”
Shared via PressReader
connecting people through news
Future Toilets, Mirrors Give Medical Feedback
SARAH PAYNTER
The Wall Street Journal 10 Apr 2024
It’s the year 2034. You wake up and head to the bathroom. First, an alert from your toilet—you’re dehydrated—better remember to take the water bottle to work. The mirror over your sink advises you to apply your prescription cream. Next, step into the shower—it glows with infrared light, designed to soothe inflammation.
Tech-driven bathrooms could be common in many high-end U.S. homes in the next decade, with postpandemic wellness concerns helping drive the trend, designers and tech executives said. Smart bathroom tech is already used in top hospitals, hotels and some ultraluxury homes in China, said Thomas Serval, co-founder and chief executive of Baracoda, a French company that specializes in internet-connected health technology.
Advanced technology for bathrooms could alleviate burdens on the healthcare system and help aging populations stay longer in their homes, said Steve Scheer, president of Brondell, a San Francisco-based home-technology company.
Yet privacy is also a concern: Do people really want their time in the bathroom monitored and analyzed?
Here’s a look at the tech that could be coming to your bathroom.
Flush with health
In the coming years, homeowners will be looking for toilets with more features, like urinalysis and self-cleaning capabilities, said Bill Darcy, global president and CEO of the National Kitchen & Bath Association. Homeowners are increasingly giving priority to health and wellness in home design since the pandemic, he said. Health applications are among the innovations companies in Japan, Silicon Valley and France are working on.
Toilets in the near future will be able to detect a variety of conditions, such as urinary tract infections and kidney issues, using built-in chemicalactivated tests, said Vik Kashyap, CEO of San Francisco-based Toi Labs, a maker of smart-toilet technology.
In some senior-living facilities, toilet seats with downward-pointing cameras already use artificial intelligence from Toi Labs to monitor urine and stools for potential signs of illness and disease, Kashyap said. Toi Labs leases its smart toilet seats to senior-living facilities, charging between $45 and $65 a month per seat, a cost that includes daily reports.
The French company Withings sells pods, at around $500 apiece, that sit inside toilet bowls and use reactions in chemical cartridges to detect vitamin levels, glucose levels and ovulation timing, said product manager Inna Ndaw. Results are sent to an app.
Other inventions are in the works. Toi Labs plans to incorporate electronic-nose, or e-nose, technology to detect smells that could warn of disease. The company is also developing toilet seats that use
different types of light against the skin to determine users’ body temperature, heart rate and blood oxygenation.
When it comes to privacy concerns, Toi Labs said that its sensors don’t record body parts and that data collected from its technology is separated from identifiable names.
Makers of smart toilets are building in sterilizing ultraviolet lights and cleaning agents, said Terri Almendares of Farrey’s Lighting, Bath, Kitchen and Hardware in Miami. Ultraviolet light damages bacteria and viruses, preventing them from reproducing.
Self-cleaning toilets typically start at around $500 but can cost much more. At about $8,600, Kohler’s Numi 2.0 toilet has sanitizing functions, heated seats and personalized bidet settings.
Demand for devices with these capabilities is limited, but Bill Strang, an executive at Toto USA, a maker of selfcleaning toilets in Georgia, reports a surge in U.S. interest in the past five years.
E-mirror, e-mirror
Look into your bathroom mirror, and you might one day see a face other than your own—your doctor’s.
Tech companies imagine mirrors with screen-like interfaces
that could connect users to healthcare professionals when the devices flag health concerns, said Serval of Baracoda, which owns CareOS, a maker of software for smart mirrors. Smart mirrors with cameras could also help you respond to some of these needs, he said: Built-in artificial intelligence can make personalized skin-care recommendations, and augmentedreality-capable mirrors can guide users on “face yoga” or meditation techniques.
CareOS safeguards security by keeping all data local, Serval said. Instead of uploading the data to the internet, it is stored on devices in the bathroom.
NuraLogix, a Canadian health-products company, recently unveiled a tabletop mirror with a camera equipped to detect facial blood flow to reveal your blood pressure and predict risk of heart attack, stroke and hypertension, said Lindsay Brennan, a spokeswoman for the company. Results are displayed on the mirror.
High-tech home spa
Among trends for the future, homeowners will want their bathrooms to be relaxing and safe, supporting health, well-being and longevity, said Alexandra Yacavone, a design studio manager at Kohler, the Wisconsin-based manufacturing company known for its toilets, bathtubs and other fixtures.
Companies are already experimenting with tech to create a spa feeling at home. Earlier this year, Kohler came out with a bathtub called “Stillness Infinity Experience” that creates soothing waterfall sounds as bathwater overflows into a moat. It comes with a fog machine with aromatherapy capability and built-in custom colored lighting meant to relax the user.
Showers with sauna-like steam and infrared capabilities are already on the market, starting at around $3,000. Research suggests that infrared heat, with its red glow, has benefits including relaxation and stress relief, and it is sometimes promoted as a way to detoxify your body.
Voice-activated shower controls are already available, but new methods for finetuning the water pressure are in development.
A smart floor
After your shower, step onto your smart bath mat. These mats—essentially, advanced scales with pressure sensors inside cushioned, water-absorbent fabric—can measure your weight and body composition. They also detect posture and balance to predict risk of neurodegenerative disease by using thousands of sensors to assess how a user stands and walks, sending the results to an app. BBalance, a maker of smart bath mats that is owned by Baracoda, expects to roll out its mats in the U.S. by 2025, Serval said.
Smart floor tiles could also use pressure sensors to detect balance, gait and posture. Pressure sensors were embedded underneath ceramic tile floors at a nursing home for a 2020 study conducted at the University of Texas at Arlington. It found that sensors were able to extract meaningful data by measuring the width, length and speed of users’ steps, which could be used to predict fall risk and spot warning signs for illness.
Sinking up
In the future when you wash your hands and brush your teeth, you might get some feedback.
Tech companies would like to be able to analyze saliva to show the presence of strep or Covid-19 germs going down the drain, Serval said.
Though few scientists are currently developing sink technologies, public-health experts already analyze wastewater to track outbreaks of Covid and other diseases.
Using a similar technology, sinks could detect the presence of bacteria and viruses, said Alexandria Boehm, a professor of civil and environmental engineering who runs an academic program at Stanford University that analyzes municipal wastewater for infectious diseases. But she noted that the presence of toothpaste and other personal-care products could make it harder to get accurate results.
Other researchers are working to develop toothbrushes that could detect Covid or cancer cells in saliva, or predict risk of disease.
Shared via PressReader
connecting people through news
Stephen King - The rage in ‘Carrie’ feels more relevant than ever
Jayatissa AMANDA JAYATISSA is the author of three novels, most recently “Island Witch.”
The New York Times (International Edition) 10 Apr 2024
In “On Writing,” Stephen King’s nonfiction account of his career, he talks about a girl he calls Dodie Franklin. She attended his high school and, he recalls, was often bullied for wearing the same clothes every day. In their sophomore year, on the first day back after Christmas vacation, she came to school wearing newly fashionable clothes with a trendy hairstyle — but the bullying and teasing never stopped. “Her peers had no intention of letting her out of the box they’d put her in,” Mr. King writes. “She was punished for even trying to break free.”
The realization that nothing could change Ms. Franklin’s social standing, coupled with a few more unfortunate examples of young women he knew, helped inform a story about a bullied girl with telekinetic powers who is pushed to her limits and who wreaks brutal revenge on her classmates and, eventually, her abusive mother. “Carrie,” Mr. King’s first published novel, was released 50 years ago, in 1974.
There have been many iterations of “Carrie” since. Horror enthusiasts will recall the classic film directed by Brian De Palma and released in 1976; there have been several remakes, most recently one in 2013 starring Chloë Grace Moretz. There was an ill-fated stage adaptation, “Carrie: The Musical,” which the TV show “Riverdale” once paid homage to. Many things have changed in the half-century since Mr. King’s novel was published, yet Carrie White remains a strikingly relevant and highly relatable figure. She raged her way to a place in pop culture’s pantheon. But why?
I first read “Carrie” as a nerdy, horror-enthused 14-year-old growing up in Sri Lanka. At the library of the Christian school I attended, Mr. King’s books were extremely hard to come by, so when I saw a copy at a friend’s house, I was quick to borrow it. I vividly remember being drawn to Carrie’s wideeyed gaze on the cover, blood trailing from her forehead and dripping down her chin. “Nobody was really surprised when it happened,” it reads in the opening pages. “Not really, not at the subconscious level where savage things grow.” I was hooked. What did Mr. King mean by “savage things”? I didn’t realize then that I would spend so much of my adult life thinking about this very question.
I’ve reached for “Carrie” many times since, and my relationship with the story has continued to shift and evolve. Like most teenagers, I suppose, I initially reacted to Carrie’s story with pure horror; I was mortified by the way she was teased, repulsed by the pig’s blood that gets dumped over her at prom and fascinated by the death and destruction she wrought in retaliation. In my 20s, when I revisited the novel, the horror I felt at her tale turned to something closer to sympathy. By that point, I’d moved from Colombo to California to Britain and then back to my hometown in Sri Lanka and had chalked up enough life lessons to understand Carrie’s suffering in a different way.
Now, as a woman in my 30s, I no longer see Carrie as simply a victim to be pitied. I’ve learned to relish her rage. Her anger has inspired much of my own fiction writing and, more important, has taught me that anger, when channeled, can be an asset. This truly hit home for me in July 2022, when I joined thousands of protesters in Colombo marching against corruption and the economic mismanagement of the country’s leaders. Years of feeling powerless finally erupted. We were all angry, of course, but we used our rage as fuel.
In the past year, women in the United States have had many reasons to figuratively burn down auditoriums and destroy towns. The war on women is still very much alive, as Roe v. Wade was overturned, in vitro fertilization procedures were endangered in Alabama and pregnant women are still not allowed to divorce their husbands in Missouri.
These days I see Carries everywhere. At the end of 2023, Gypsy Rose Blanchard — who had been convicted of second-degree murder in connection with the death of her mother, Dee Dee Blanchard, after years of being subjected to abuse and Munchausen syndrome by proxy — was released from prison. There are many interesting parallels between Gypsy Rose Blanchard and Carrie, the most glaring being the obvious torment of each by her mother.
What struck me as most interesting (read: most depressing) was the public response Ms. Blanchard received after her release. While some hailed her as a folk hero, many labeled her a killer, much like Carrie, for fighting back against her tormentor — not just in a court of law but also in the court of social media. TikTok was rife with hot takes, arguing there was something sinister about Ms. Blanchard or claiming that her husband was actually her brother. Rather than being viewed as a young woman trying to navigate her way through an absolutely horrendous situation, she was criticized for participating in a television series. Many people seemed content when she was the victim, but it infuriated them when she tried to take a stand for herself. What was true of Mr. King’s account of Ms. Franklin proved true for Ms. Blanchard, too: “She was punished for even trying to break free.”
Look at the way that Meghan Markle has been treated — criticized as an attention seeker for speaking out in a society that constantly blames women for staying silent. Or take Britney Spears. We all cried “Free Britney” and lamented the ignorance of our ways when we learned of how those in her life had treated her terribly, yet all it takes is a social media post of her dancing in clothes deemed by some to be too provocative to leave viewers shaking their heads at her again, saying she has gone off the deep end. God forbid women choose to fight back by simply expressing themselves in a way that defies convention.
I believe we still too often look at women who fight back against their oppressors and see them as villains rather than assigning responsibility for their situations to the people who tormented them. Carrie has always been the antidote to that predicament: She forces us to confront our feelings about what happens when women instill some of the same fear in others that they are too often forced to deal with themselves. Carrie’s plight still speaks to feelings in women of rage, helplessness and a desire for justice or, failing that, retribution. None of that has gone away in 50 years.
Beyond being a supremely well-told story, Mr. King’s novel still connects on the same “savage” subconscious level he mentioned at the start of the book. “Carrie” was packaged and marketed as horror, but what is it about the character of Carrie that’s truly horrifying? Is it the revenge that’s exacted by a bullied girl? Or is it the actions of those who stood around and allowed her to be tormented? The question at the heart of the story is: Who is the real monster? Fifty years later, we’ve come to understand that it’s not Carrie but the world that made her.
Shared via PressReader
connecting people through news
TECH GIANTS CUT CORNERS FOR A.I.
The New York Times (International Edition) 10 Apr 2024
In late 2021, OpenAI faced a supply problem.
The artificial intelligence lab had exhausted every reservoir of reputable English-language text on the internet as it developed its latest A.I. system. It needed more data to train the next version of its technology — lots more.
So OpenAI researchers created a speech recognition tool called Whisper. It could transcribe the audio from YouTube videos, yielding new conversational text that would make an A.I. system smarter.
Some OpenAI employees discussed how such a move might go against YouTube’s rules, three people with knowledge of the conversations said. YouTube, which is owned by Google, prohibits use of its videos for applications that are “independent” of the video platform.
Ultimately, an OpenAI team transcribed more than one million hours of YouTube videos, the people said. The team included Greg Brockman, OpenAI’s president, who personally helped collect the videos, two of the people said. The texts were then fed into a system called GPT-4, which was widely considered one of the world’s most powerful A.I. models and was the basis of the latest version of the ChatGPT chatbot.
The race to lead A.I. has become a desperate hunt for the digital data needed to advance the technology. To obtain that data, tech companies including OpenAI, Google and Meta have cut corners, ignored corporate policies and debated bending the law, according to an examination by The New York Times.
At Meta, which owns Facebook and Instagram, managers, lawyers and engineers last year discussed buying the publishing house Simon & Schuster to procure long works, according to recordings of internal meetings obtained by The Times. They also conferred on gathering copyrighted data from across the internet, even if that meant facing lawsuits. Negotiating licenses with publishers, artists, musicians and the news industry would take too long, they said.
Like OpenAI, Google transcribed YouTube videos to harvest text for its A.I. models, five people with knowledge of the company’s practices said. That potentially violated the copyrights to the videos, which belong to their creators.
Last year, Google also broadened its terms of service. One motivation for the change, according to members of the company’s privacy team and an internal message viewed by the Times, was to allow Google to be able to tap publicly available Google Docs, restaurant reviews on Google Maps and other online material for more of its A.I. products.
The companies’ actions illustrate how online information — news stories, fictional works, message board posts, Wikipedia articles, computer programs, photos, podcasts and movie clips — has increasingly become the lifeblood of the booming A.I. industry. Creating innovative systems depends on having enough data to teach the technologies to instantly produce text, images, sounds and videos that resemble what a human creates.
The volume of data is crucial. Leading chatbot systems have learned from pools of digital text spanning as many as three trillion words, or roughly twice the number of words stored in Oxford University’s Bodleian Library, which has collected manuscripts since 1602. The most prized data, A.I. researchers said, is high-quality information, such as published books and articles, which have been carefully written and edited by professionals.
For years, the internet — with sites like Wikipedia and Reddit — was a seemingly endless source of data. But as A.I. advanced, tech companies sought more repositories. Google and Meta, which have billions of users who produce search queries and social media posts every day, were largely limited by privacy laws and their own policies from drawing on much of that content for A.I.
Their situation is urgent. Tech companies could run through the high-quality data on the internet as soon as 2026, according to Epoch, a research institute. The companies are using the data faster than it is being produced.
Tech companies are so hungry for new data that some are developing “synthetic” information. This is not organic data created by humans, but text, images and code that A.I. models produce — in other words, the systems learn from what they themselves generate.
OpenAI said each of its A.I. models “has a unique data set that we curate to help their understanding of the world and remain globally competitive in research.” Google said that its A.I. models “are trained on some YouTube content,” which was allowed under agreements with YouTube creators, and that the company did not use data from office apps outside of an experimental program. Meta said it had “made aggressive investments” to integrate A.I. into its services and had billions of publicly shared images and videos from Instagram and Facebook for training its models.
For creators, the growing use of their works by A.I. companies has prompted lawsuits over copyright and licensing. The Times sued OpenAI and Microsoft last year for using copyrighted news articles without permission to train A.I. chatbots. OpenAI and Microsoft have said using the articles was “fair use,” or allowed under copyright law, because they transformed the works for a different purpose.
In May, Sam Altman, the chief executive of OpenAI, acknowledged that A.I. companies would use up all viable data on the internet.
“That will run out,” he said in a speech at a tech conference.
Mr. Altman had seen the phenomenon up close. At OpenAI, researchers had gathered data for years, cleaned it and fed it into a vast pool of text to train the company’s language models. They had mined the computer code repository GitHub, vacuumed up databases of chess moves and drawn on data describing high school tests and homework assignments from the website Quizlet.
By late 2021, those supplies were depleted, said eight people with knowledge of the company, who were not authorized to speak publicly.
OpenAI was desperate for more data to develop its next-generation A.I. model, GPT-4. So employees discussed transcribing podcasts, audiobooks and YouTube videos, the people said. They talked about creating data from scratch with A.I. systems. They also considered buying start-ups that had collected large amounts of digital data.
OpenAI eventually made Whisper, the speech recognition tool, to transcribe YouTube videos and podcasts, six people said. But YouTube prohibits people from not only using its videos for “independent” applications, but also accessing its videos by “any automated means (such as robots, botnets or scrapers).”
OpenAI employees knew they were wading into a legal gray area, the people said, but believed that training A.I. with the videos was fair use. Mr. Brockman, OpenAI’s president, was listed in a research paper as a creator of Whisper. He personally helped gather YouTube videos and fed them into the technology, two people said.
Mr. Brockman referred requests for comment to OpenAI, which said it uses “numerous sources” of data.
Last year, OpenAI released GPT-4, which drew on the more than one million hours of YouTube videos that Whisper had transcribed. Mr. Brockman led the team that developed GPT-4.
Some Google employees were aware that OpenAI had harvested YouTube videos for data, two people with knowledge of the companies said. But they didn’t stop OpenAI because Google had also used transcripts of YouTube videos to train its A.I. models, the people said. That practice may have violated the copyrights of YouTube creators. So if Google made a fuss about OpenAI, there might be a public outcry against its own methods, the people said.
Matt Bryant, a Google spokesman, said the company had no knowledge of OpenAI’s practices and prohibited “unauthorized scraping or downloading of YouTube content.” Google takes action when it has a clear legal or technical basis to do so, he said.
In late 2022, after OpenAI released ChatGPT and set off an industrywide race to catch up, Google researchers and engineers discussed tapping other user data. Billions of words sat in people’s Google Docs and other free Google apps. But the company’s privacy restrictions limited how they could use the data, three people with knowledge of Google’s practices said.
In June, Google’s legal department asked the privacy team to draft language to broaden what the company could use consumer data for, according to two members of the privacy team and an internal message viewed by The Times.
The privacy team wrote new terms so Google could tap the data for its “A.I. models and build products and features like Google Translate, Bard and Cloud AI capabilities,” which was a wider collection of A.I. technologies.
Mr. Bryant said that the privacy policy changes had been made for clarity and that Google did not use information from Google Docs or related apps to train language models “without explicit permission” from users.
“We did not start training on additional types of data based on this language change,” he said.
Mark Zuckerberg, Meta’s chief executive, had invested in A.I. for years — but suddenly found himself behind when OpenAI released ChatGPT in 2022. He immediately pushed to match and exceed ChatGPT, calling executives and engineers at all hours of the night to push them to develop a rival chatbot, said three current and former employees, who were not authorized to discuss confidential conversations.
But by early last year, Meta had hit the same hurdle as its rivals: not enough data.
Ahmad Al-Dahle, Meta’s vice president of generative A.I., told executives that his team had used almost every available English-language book, essay, poem and news article on the internet to develop a model, according to recordings of internal meetings, which were shared by an employee.
Meta could not match ChatGPT unless it got more data, Mr. Al-Dahle told colleagues. In March and April 2023, some of the company’s business development leaders, engineers and lawyers met nearly daily to tackle the problem.
Some debated paying $10 a book for the full licensing rights to new titles. They discussed buying Simon & Schuster, which publishes authors such as Stephen King, according to the recordings.
They also talked about how they had summarized books, essays and other works from the internet without permission and discussed sucking up more, even if that meant facing lawsuits. One lawyer warned of “ethical” concerns around taking intellectual property from artists but was met with silence, according to the recordings.
Shared via PressReader
connecting people through news
FIVE EASY WAYS TO START BAKING
New York Times (International Edition) 10 Apr 2024
The poet Emily Dickinson was an avid baker, and on the back of a recipe card for coconut cake, she wrote these opening lines:
The Things that never can come back, are several — Childhood — some forms of Hope — the Dead —
But while her gifts as a poet are clear here, she expresses the opposite of what baking can do. A birthday cake brings back the joy of childhood, maybe even raises hope, and baking beloved recipes from the deceased resurrects memories of them.
Even if the goal isn’t reclaiming what’s lost, the simple act of baking can conjure unexpected delight. When you’re preparing a meal and starting with salmon and potatoes, you end up with cooked salmon and potatoes. But when you’re baking, you start with a slew of powders, golden butter and an egg, and you end up with crackly-edged, chewy blondies.
If you’re a beginner in the kitchen, baking is an ideal entry point. Unlike cooking, there’s no pressure to make food that’s meant to sustain, no urgency from step to step. You can go at your pace, and the process can even feel relaxing. These five foolproof recipes are the best place to start: They require only a handful of tools and ingredients — and no experience. They welcome other seasonings and flavors, and guarantee something tasty. Make them all to learn the basics of baking, or try any one that appeals to you. Not only are they easy, but they also offer the satisfaction of dessert and the wonder at having made it yourself.
BAKING SUPPLIES FOR BEGINNERS
To make all five of these recipes, you need:
• A sheet pan
• An ovenproof skillet
• An 8-inch (20-centimeter) square cake pan
• Two bowls
• A whisk
• A silicone spatula
• A set of dry measuring cups
• A set of dry measuring spoons
START WITH A HANDFUL OF TOOLS
You don’t have to invest in an expensive electric mixer, stacks of pans or a kitchen scale. Danielle Sepsy, the chef and owner of the Hungry Gnome Bakery, remembers a childhood of standing on a stool next to her grandmother Rosemarie Marullo, who scooped flour with a coffee cup. Ms. Sepsy doesn’t recommend trying this at home and uses a scale at her bakery, but said, “If you have trusty measuring cups on hand, you’re OK.”
In fact, you can use dry measuring cups for liquids even though it’s a little tricky to not spill with them. (Liquid measuring cups don’t work for measuring dry ingredients, though.)
SHOP FOR BASIC INGREDIENTS
Baked goods can incorporate everything from chiles to miso, but their foundation requires only a small group of essentials. Baking can be traced back to ancient civilizations, but the sweets here come from the style that spread from Europe to America and the world, built on flour, baking powder, baking soda, sugar, salt, butter or oil and, usually, eggs. Often, other dairy products or flavorings like vanilla extract and cocoa powder are used, and sometimes not all of the basics are necessary.
BASK IN THE PRECISION
Baking can be perceived as stressful because it requires following a recipe, but that’s what can make it feel calming. “I actually have a lot of anxiety, and I started baking because it gave me a sense of control,” Ms. Sepsy said. “If you follow the recipe exactly, it’ll result in exactly what you want.” Even though she now weighs ingredients to run her professional kitchen, she uses regular measuring cups at home because it makes baking “more stress-free and fun.”
For simple baked goods like these, it’s fine if you end up adding a little too much flour or don’t beat the eggs enough. As long as you’re mixing and baking sugar, fat and starches as described in the instructions, you’re going to end up with something delicious.
GO AHEAD AND TWEAK RECIPES
Some cooks don’t like baking because there doesn’t seem to be room for improvisation without risking a failure in the oven. Don’t mess with the base formula for the batter or dough, but do customize seasonings that don’t affect baking chemistry. Stir in spices like cardamom or ras el hanout and add your choice of toppings or mix-ins. When making fruit desserts, use what’s in season (and on sale).
Shared via PressReader
connecting people through news
How the French hope the Olympics will help reshape the capital
Economist (UK) 06 Apr 2024
ON THE SITE of a former piano factory in the northern Paris suburb of Saint-Denis, a 40-storey tower is being converted into a gleaming luxury hotel with a rooftop bar. A short walk away, beside the river Seine, builders are finishing off a vast new “eco-neighbourhood” of flats, lined with saplings and lamp posts made from recycled scaffolding. These will briefly lodge 10,500 athletes during the Paris Olympic games, which take place from July 26th-August 11th. Across the railway tracks, on land that formerly housed a gasworks, workers are completing a brand-new aquatics centre, under a gently curved timber frame of French and Finnish pine.
These developments are part of an attempt by French urban planners to use the Olympics to revive Seine-Saint-Denis, a banlieue (suburb) that hugs the north and eastern edges of Paris. During the games many sporting events will take place in the historic city centre, including beach volleyball under the Eiffel Tower. But some of the most prestigious, such as athletics, will be held at the Stade de France in Seine-Saint-Denis. More than this, the Olympics is part of a big rethink of greater Paris, and its transport system, which could in time radically change the capital’s geography.
As in many of Europe’s old cities, the historic centre of Paris, with its tree-lined avenues and cycle lanes, is fringed by poverty, high-rise blocks and former industrial land. Paris, delineated by its forbidding périphérique, an eight-lane ring-road, is particularly cut off. In the capital’s cobbled centre, urban planners enthuse about the “15-minute city”, in which work, cafés and cinemas are but a short walk away. In the banlieues, it often takes longer than that just to reach the station—if there is one.
Fully 1.7m people of different tongues and faiths crowd into Seine-Saint-Denis, where tower blocks were built in the 1960s and 1970s partly to house industrial workers recruited in North Africa, and their families. Its poverty rate is 28%, nearly twice that in Paris. Almost a quarter of families are headed by a single parent. At the age of ten, says Quentin Gesell of the Métropole du Grand Paris, an administrative body, half the children in Seine-Saint-Denis today do not know how to swim.
The various greater Paris authorities now hope that the Olympics can help turn this area’s fortunes, and reputation, around. The sheer scale of investment, much of which will not be complete until the end of the decade, is eye-watering. Under a plan first hatched in 2007, tunnels are still being dug to create a giant loop of 200km of new driverless Metro lines and 68 new stations, known as the Grand Paris Express. When finished, it will double the existing Metro network. Crucially, instead of carrying people from the banlieues only in and out of the centre, it will also link the
outskirts to each other and the city’s airports, all at a cost of some €42bn ($45bn).
The idea, says Marie Barsacq, in charge of Olympics legacy-planning, is that the games leave behind facilities and transport, as well as creating jobs and boosting skills. Seine-Saint-Denis, which today has the fewest swimming pools per head in France, will be bequeathed not just the new aquatics centre but over a dozen new or renovated public pools. The staff needed to prepare and serve 13m meals in the athletes’ village are all being hired locally.
Karim Bouamrane, the Socialist mayor of Saint-Ouen, sees the Olympics as an accelerator for reviving the area. Projects already under way have been given a boost; sporting facilities a facelift. New transport links will make Saint-Ouen one of the best-connected hubs in the Paris banlieue. Elon Musk’s Tesla, an electric-car firm, has picked Saint-Ouen for its new French headquarters. Tony Parker, a French-American former professional basketball player, is opening a new sports academy there. “Before, when you used to come to Saint-Ouen it was because you had no choice,” says Mr Bouamrane. “Now it’s a choice.”
The lesson from other European cities that have tried similar projects, including London’s Stratford or Hamburg’s Hafen-City, is that better infrastructure can help revive neighbourhoods and lure private developers. But it takes time, and huge upfront public investment. Post-covid, developments that rely on office space are fragile. Extravagant vanity projects can leave a pile of debt and little else. France certainly loves its grands projets. But Paris is focusing events mostly on existing facilities, in order to avoid the fate of host cities such as Athens, where the stadiums at Hellinikon were left abandoned for years afterwards.
For places like Saint-Ouen, just beyond the ring-road, the chances of such investment paying off are probably greater than for those on the outer periphery. Yet the challenge of bridging the gulf between Paris and its banlieues is still daunting. Advanced sales of flats at the athletes’ village have been slow. Many Parisians seldom, if ever, venture to the outskirts. Locals feel their address, and the département code 93, stigmatises them. “We really struggle with the 93,” says a woman outside a boulangerie in Saint-Ouen; “As soon as you mention 93, it’s no good.” Mr Bouamrane notes that “etymologically the word banlieue means the ‘place of the banished’. It’s the place where we don’t like you.”
Across from the nearly finished Metro station at Saint-Denis Pleyel, a teenage girl on a street bench thinks that not much will change: “There will still be drugs, alcohol, traffickers, dealers and all that.” A young man hanging out with friends by the Turkish kebab shop laughs at the idea that they would use the swimming pool afterwards: “We’re not children!” Yet others in his group display an unexpected optimism. “There will be a before and after for Saint-Denis,” declares one: “After the games, Saint-Denis will be super-beautiful.”
Shared via PressReader
connecting people through news
Why doesn’t graduate stay ?
Economist (UK) 06 Apr 2024 Official statistics reveal why some places have so few graduates
BRITONS TEND to believe that jobs should come to them, not the other way round. A poll in 2021 showed that a majority wanted the government to focus on ensuring that everyone could find a good job in their local area. “For many, if you want to get on you need to get out,” explained the government the following year. It promised “to change that for good”.
But how exactly do people move around? In particular, why do some places end up denuded of young graduates, and therefore less attractive to employers, while others are stuffed with them? Last month the Office for National Statistics released data that track people through school and university and into jobs. This allows you to see, in slightly creepy detail, how the young migrate.
Take Richmond, a lovely small town in North Yorkshire represented in Parliament by the prime minister, Rishi Sunak. It has a castle, a tradition of hand-knitting props for its annual pantomime, and some successful schools. Of the roughly 430 teenagers who did GCSEs there between 2008 and 2011, fully 205 ended up graduating from university (see chart on next page).
Only one-third were living in Richmond by their late 20s. Paul Harrison of
the Original Richmond Business and Tourism Association (“original” to distinguish the town from some jumped-up Richmonds, including one in London) suggests several reasons. The town offers few hightech jobs; homes are expensive; public transport to job hotspots like Darlington is not great. Other rural towns have similar flaws, and similar out-migrations.
Mr Harrison says that Richmond has started to think about how to retain more young graduates. Building lots more housing is unlikely to be a favoured solution: one plan to build 32 homes on a field next to the town was rejected last year. More than 700 people signed a paper petition against the development—not bad, given Richmond’s population of about 8,000.
South Shields, a bigger, poorer post-industrial town not far from Newcastle and Sunderland, has a different pattern. It churns out fewer graduates—only a quarter of young people obtain degrees, compared with almost half in Richmond. But more than two-thirds of them end up living in South Shields. One likely reason is the range of universities in and around Tyneside. Whereas just 2% of young graduates from Richmond study nearby, 38% from South Shields do. They never break the link to home.
The two towns take different routes to a similar outcome. Once you include the (rather few) young people who move to Richmond and South Shields from somewhere else, both end up with only 22 young graduates for every 100 people who took GCSEs. Richmond probably has the edge when it comes to attracting middle-aged and old graduates, including knitters.
In this light, the success of comfortable suburbs in south-east England becomes a little clearer. Bushey, on the outskirts of London, produces graduates as reliably as Richmond and holds on to them like South Shields. It also draws more young graduates from elsewhere than either northern town. Bushey ends up with 47 graduates for every 100 GCSE-takers. It is a happy hunting-ground for recruiters.
Shared via PressReader
connecting people through news
Stormy weather
Economist (UK) 06 Apr 2024 How the Bank of England has managed four years of economic shocks—and how it might do better in the next four
How Britain’s central bank has managed four years of economic shocks—and how it might do better in the next four,
SEVEN DAYS after Andrew Bailey became governor of the Bank of England in March 2020, Britain’s first covid-19 lockdown began. Storms have kept coming. The economy has had to endure two more supply shocks (the completion of Brexit and a leap in energy prices after Russia invaded Ukraine); a demand boost (postpandemic fiscal stimulus); and a bondmarket blowup (after Liz Truss’s reckless mini-budget in September 2022). Any central banker would want calmer weather.
Halfway through Mr Bailey’s eight-year term, how has the bank fared? Judged by its main job, maintaining price stability, direly. The annual inflation rate has exceeded the bank’s target, 2%, for almost three years, peaking at 11.1% in late 2022 before falling to 3.4% in February. The core rate— excluding food and energy—is at 4.5%.
Still, most forecasters expect the bank to be nearing its target again within a few months. And central bankers are not allpowerful. Their tools—in essence, spurring or reining in demand by setting shortterm interest rates and intervening in bond markets—can be little use in the face of big supply shocks. A fairer yardstick is inflation expectations: has the bank persuaded firms, households and markets that inflation won’t stay high for long? Expectations of persistently high inflation can be selffulfilling, as firms and workers bid up prices and wages in an attempt to stay ahead.
Here, the record looks better. Britons’ long-term inflation outlook has barely budged (see chart 1 on next page). But the bank’s credibility, polished over a quartercentury of independence, is showing a few dents. Prices of inflation-linked bonds imply that markets expect inflation to be 0.3 percentage points higher in the long term than they did when Mr Bailey took charge. Surveys say that 39% of Britons expect inflation to exceed 3% in five years, against 31% in early 2020. The number of Google searches for “inflation” has trebled in the past four years. When inflation surges again, the bank cannot afford to dither.
Even allowing for its blunt tools, might the bank have done better? It was taken by surprise not once but twice. The first surprise was that inflation emerged at all. The trouble started in 2021, as the pandemic ebbed and the need for stimulus wore off. The bank delayed raising rates until that December (see chart 2), earlier than many central banks but still too late.
At the time the need to tighten was not wholly clear-cut. Spending on goods had boomed in the pandemic, straining supply chains. But it was hard to be sure that that was truly a sign of broad overheating rather than of isolated stress. The labour market looked slack—a government “furlough” scheme was supporting around 1m jobs— but proved not to be. When the scheme ended, unemployment kept falling.
Even so, the bank was too sanguine. A “Taylor rule”—a rule of thumb devised by John Taylor, an American economist—suggests that rates should have been rising by mid-2021. Other countries, including Brazil, Chile and Norway, did lift rates sooner.
Worse, the bank continued quantitative easing (a policy of buying bonds begun when rates were at rock bottom after the financial crisis of 2007-09) deep into 2021, even as markets had stabilised and bond yields sat at their lowest for centuries. (The bank has lost an estimated £50bn-130bn, or $63bn-164bn, on its bond purchases, covered by the Treasury.) And the sheer size of the government’s pandemic spending, amounting to 10% of GDP in 2020-21, should have caused concern sooner, especially given the disruptions to supply.
The second surprise was that inflation stayed high. By 2022 post-pandemic inflation had been amplified by Russia’s invasion of Ukraine. Natural-gas prices went parabolic, hitting ten times the levels of the 2010s. The bank consistently predicted that inflation would soon fall (see chart 3). Instead, it spread from energy, food and other goods to services, wages and housing, from which it is harder to root out.
Even as gas prices plummeted in 2023 after a warm winter in Europe and a slowdown in China, core inflation kept rising. Last summer the bank asked Ben Bernanke, an ex-chairman of the Federal Reserve and a Nobel economics laureate, to review its forecasting practices. His findings are due on April 12th.
Macroeconomic forecasting is notoriously tricky. Data are sparse and inexact; the structure of economies shifts over time. A key variable in forecasting inflation is the size of “second-round” effects: how far will price rises be self-perpetuating? Such effects fed a decade of brutal inflation after the oil shocks of the 1970s. But from the 1990s they faded. Workers did not chase inflation-reinforcing pay increases. Weaker trade unions, migration and globalisation probably played a part; so did credible, independent central banks.
The bank was wrong-footed when that post-1990s pattern did not persist. Secondround effects were back. Academics suspect that recent price shocks were severe enough to make history a poor guide to how firms and workers might respond.
The bank will doubtless revise its models accordingly. But good central banking ought to try to take account of uncertain forecasts. Mr Bernanke is expected to recommend scenario-oriented forecasting. Developing and publishing a scenario incorporating 1970s-style second-round effects, for instance, would have been valuable. Another good idea, advocated by Catherine Mann (who is on the bank’s ratesetting committee) and Isabel Schnabel (of the European Central Bank) is to lean towards tighter policy when rising inflation is not well understood (eg, when it has defied forecasts).
Many central banks have struggled with forecasting. But the Bank of England’s lapses in communication have stood out. Sometimes it has been tin-eared (if economically accurate). Mr Bailey called pay rises “unsustainable”. Huw Pill, the bank’s chief economist, said Britons should “accept that…we’re all worse off” from the energy shock. On occasion, it has embarrassed itself. Rumours trickled out about Mr Bailey’s penchant for falling asleep in meetings. Uncharitable colleagues nicknamed him “Rock-a-bye Bailey”.
Communications mishaps about policy are more serious. Especially during crises, central bankers must speak clearly to financial markets, which set the long-term borrowing costs facing households and businesses. Policymakers’ priorities will then be transmitted quickly across markets. But the bank approached rate rises slowly and piecemeal.
Mr Bailey did not, for example, give a speech like that of Jerome Powell, who heads the Fed today, in August 2022 at Jackson Hole, Wyoming. Mr Powell hammered home the idea that he would tighten policy as much as needed to tame inflation. Mr Bailey sometimes tried unhelpfully to talk markets out of pricing in interestrate rises. In March 2023 he questioned whether it was wise to expect rates to go as high as 4.75%, as markets then did. They ultimately reached 5.25%.
With inflation high, by 2022 public confidence in the bank had sunk uncomfortably low. But the bank found unlikely saviours in the 49-day prime minister, Ms Truss, and her 38-day chancellor, Kwasi Kwarteng. Their spendthrift fiscal plan caused an implosion in bond prices that the bank handled deftly with an injection of liquidity into the gilt market.
That earned the bank political capital, which will be especially valuable as a general election, due by next January, draws near. When moves in interest rates could, in principle, help one party or the other, the bank must be above suspicion.
If, as looks highly likely, the Labour Party wins the election, what might that mean for the bank? Much less than after Labour’s most recent return, in 1997. Within days the new chancellor, Gordon Brown, surprised markets by declaring the bank independent. Rachel Reeves, the shadow chancellor today, who once worked at the bank, has more modest intentions. She promises to uphold its independence, keep the 2% inflation target and refocus financial regulation on climate risk.
Some Conservatives may be less friendly, blaming the bank for both the inflation and the stagnation likely to contribute to their defeat. Out of office, they will pose no direct danger, but if the economy remains lacklustre they may nevertheless fuel distrust of the bank.
So the bank will have to get better at politics. The best way to do that will be to get better at its main job, fighting inflation. The past four years have battered its credibility, though it remains mostly intact; and in a more volatile, inflation-prone economic environment, it will be called upon to police price rises more than to stave off recession. Mr Bailey may hope that the next four years will be less stormy. Don’t expect the weather to oblige.
Shared via PressReader
connecting people through newsL
China’s risky reboot
Economist (UK) 06 Apr 2024 Xi Jinping wants to escape economic stagnation using a high-tech revolution. His plans will backfire
IT IS CHINA’S gravest economic test since the most far-reaching of Deng Xiaoping’s reforms began in the 1990s. Last year the country achieved growth of 5%, but the pillars of its decades-long miracle are wobbling. Its famously industrious workforce is shrinking, history’s wildest property boom has turned to bust and the global system of free trade that China used to get richer is disintegrating. As our reporting explains, President Xi Jinping’s response is to double down on an audacious plan to remake China’s economy. Blending technoutopianism, central planning and an obsession with security, this sets out China’s ambition to dominate the industries of tomorrow. But its contradictions mean it will disappoint China’s people and anger the rest of the world.
Compared with 12 months ago, let alone the go-go years, the mood in China is dour. Although industrial production perked up in March, consumers are depressed, deflation lurks and many entrepreneurs are disillusioned. Behind the angst lie deeper fears about China’s vulnerabilities. It is forecast to lose 20% of its workforce by 2050. A crisis in the property industry, which drives a fifth of GDP, will take years to fix. It will hurt cash-strapped local governments that relied on land sales for revenues and flourishing real estate for growth. Relations with America are steadier, as a phone call between Mr Xi and President Joe Biden this week attested. But they remain fragile. Chinese officials are convinced that America will restrict more Chinese imports and penalise more Chinese firms, whoever wins the White House in November.
China’s response is a strategy built around what officials call “new productive forces” (see Finance & economics section). This eschews the conventional path of a big consumer stimulus to reflate the economy (that’s the kind of ruse the decadent West resorts to). Instead Mr Xi wants state power to accelerate advanced manufacturing industries, which will in turn create high-productivity jobs, make China self-sufficient and secure it against American aggression. China will leapfrog steel and skyscrapers to a golden era of mass production of electric cars, batteries, biomanufacturing and the drone-based “low-altitude economy”.
The scope of this plan is breathtaking. We estimate annual investment in “new productive forces” has reached $1.6trn—a fifth of all investment and double what it was five years ago in nominal terms. This is equivalent to 43% of all business investment in America in 2023. Factory capacity in some industries could rise by over 75% by 2030. Some of this will be made by world-class firms keen to create value, but much will be prompted by subsidies and implicit or explicit state direction. Foreign companies are welcome, even though many have been burned in China before. Mr Xi’s ultimate aim is to invert the balance of power in the global economy. Not only will China escape dependence on Western technology, but it will control much of the key intellectual property in new industries and charge rents accordingly (see Business section). Multinationals will come to China to learn, not teach. However, Mr Xi’s plan is fundamentally misguided. One flaw is that it neglects consumers. Although their spending dwarfs property and the new productive forces, it accounts for just 37% of GDP, much lower than global norms. To restore confidence amid the property slump and thereby boost consumer spending requires stimulus. To induce consumers to save less requires better social security and health care, and reforms that open up public services to all urban migrants. Mr Xi’s reluctance to embrace this reflects his austere mindset. He detests the idea of bailing out speculative property firms or giving handouts to citizens. Young people should be less pampered and willing to “eat bitterness”, he said last year.
Another flaw is that weak domestic demand means some new production will have to be exported. The world has, regrettably, moved on from the free-trading 2000s—partly because of China’s own mercantilism. America will surely block advanced imports from China, or those made by Chinese firms elsewhere. Europe is in a panic about fleets of Chinese vehicles wiping out its carmakers. Chinese officials say they can redirect exports to the global south. But if emerging countries’ industrial development is undermined by a new “China shock”, they, too, will grow wary (see Chaguan). China accounts for 31% of global manufacturing. In a protectionist age, how much higher can that figure go?
The last flaw is Mr Xi’s unrealistic view of entrepreneurs, the dynamos of the past 30 years. Investment in politically favoured industries is soaring, but the underlying mechanism of capitalist risk-taking has been damaged. Many bosses complain of Mr Xi’s unpredictable rule-making and fear purges or even arrest. Relative stockmarket valuations are at a 25-year low; foreign firms are wary; there are signs of capital flight and tycoons emigrating. Unless entrepreneurs are unshackled, innovation will suffer and resources will be wasted.
China could become like Japan in the 1990s, trapped by deflation and a property crash. Worse, its lopsided growth model could wreck international trade. If so, that could ratchet geopolitical tensions even higher. America and its allies should not cheer that scenario. If China was stagnating and discontented, it could be even more bellicose than if it were thriving.
Old reductive forces
If these flaws are obvious, why doesn’t China change course? One reason is that Mr Xi is not listening. For much of the past 30 years, China has been open to outside views on economic reform. Its technocrats studied global best practice and welcomed vigorous technical debates. Under Mr Xi’s centralising rule, economic experts have been marginalised and the feedback leaders used to receive has turned into flattery. The other reason Mr Xi charges on is that national security now takes precedence over prosperity. China must be prepared for the struggle ahead with America, even if there is a price to pay. It is a profound change from the 1990s and its ill-effects will be felt in China and around the world. ■
Shared via PressReader
connecting people through news