DAILY NEWS

Stay Ahead, Stay Informed – Every Day

Advertisement
TECH & AI
Companies Are Getting Burned by Burning Tons of Tokens
jackminion 0



Just last month, the most important metric in Silicon Valley was tokens burned—the units of measurement for the computing power being used by AI models. CEOs were giving employees the Matthew McConaughey “those are rookie numbers, you gotta pump those numbers up” speech from The Wolf of Wall Street. Now, they’re asking their staff to pump the brakes. According to a report from the Wall Street Journal, corporate leaders have realized that it actually costs money to burn AI tokens, and doing almost exclusively for the sake of doing it with no other goal in mind is actually not a great business strategy. Good thing these guys get paid tens of millions of dollars a year to figure these things out. Earlier this week, Uber CEO Dara Khosrowshahi said that it was “getting harder to justify” the cost of AI initiatives within the company because the output was not keeping up with the token burn rate, while acknowledging that part of the reason that they went so ham on token burning in the first place is that it “can feel” like AI is free.

It’s not, as it turns out. An anonymous AI consultant told Axios that one of its clients accidentally spent half a billion dollars in a single month because it never bothered to put a usage limit on employee access to Anthropic’s Claude. That is… a lot. Like to the point of straining credulity. The Journal didn’t find anything quite that egregious, but did hear about a financial institution that saw its employees burn through hundreds of thousands of dollars worth of tokens per month, with employees using premium-tier models to ask basic questions and have inane back-and-forth conversations.

That was pretty much always going to be how this very dumb era of justifying AI expenditure was going to go. Corporations have already spent tons of money to embrace these systems, and they need to justify the spend. To do so, they encourage their employees to use it as much as possible. In turn, the employees do—even when there is no point in using AI for a task. Meta killed its token-burning leaderboard last month after it leaked, revealing that the top “Token Legend” managed to burn 281 billion tokens in a month—more than the amount of compute it would take to reproduce the entirety of Wikipedia 33 times over. Amazon joined in that rollback this week, according to the Financial Times, axing its scoreboard of employees who were using the company’s internal AI tools the most—a decision that reportedly was made after it became obvious that employees were giving AI agents pointless tasks just to hold their position on the leaderboard.

It’s clear the corporate world is willing to frivolously light money on fire in an effort to justify their existing cash burn pits. Turns out they do have limits, though. You can only use “tokens burned” as a metric at so many quarterly earnings calls before shareholders start to wonder what the price tag of all those tokens adds up to.



Source link

TECH & AI
The Booing Will Continue Until Commencement Speeches Improve
jackminion 0



As someone who built a large fortune over the last few decades, I couldn’t help but notice artificial intelligence is something young people now perceive as a source of adversity. Well, roomful of 22-year-olds, let me tell you something about adversity: I myself faced a great deal of adversity back when I was building my large fortune, and when I encountered adversity, I simply turned that adversity into opportunity, which I believe to be your duty vis-a-vis artificial intelligence, which by the way is something you are powerless to halt in any way, and which I, what with my large fortune, feel excited about. Good luck to you, class of 2026! I hope that encapsulates everything 2026 commencement speakers are tempted to say about AI. If you’ve been honored with the opportunity to address a graduating class on their big day, you’re welcome to just link them to that block of text I wrote instead of actually saying it, because the thing is, if you say it, you’re gonna get booed. It happened to real estate executive Gloria Caulfield when she spoke to the graduates of University of Central Florida, and then it happened to ex-Google CEO Eric Schmidt when he spoke to the graduates of University of Arizona.

But like contact tracers working backwards to find the source of a contagion, the internet just surfaced an early example of the AI-booing phenomenon, which was given the day after Caulfield’s speech, before it went viral. The speech was by guy with a $450 million net worth Scott Borchetta, a record executive who founded the Big Machine Label Group, and was one of Taylor Swift’s adversaries in the dispute over her masters a few years ago. Middle Tennessee State University just named its media college after Borchetta after he donated $15 million. He also gave this year’s commencement address:

It’s impossible to hate the whole speech. Borchetta says now is “arguably the most exciting and challenging time ever” for the media, which, fair enough, it’s exciting in a way. At one point he says, “there’s a lot more to this world than mad wealth and political power,” and how can one not nod along? Plus, he is, to be absolutely clear, not characterizing AI as an unalloyed good for the world. Toward the beginning of the speech, he spells this out. “Our biggest challenge today? Pretty easy guess: AI,” Borchetta says, and for the time being, the crowd is with him.

The problem is that Borchetta rhymes AI with a major challenge he and Big Machine faced: how to profit from streaming back when it first took over the music business. In his telling, he made it his job to “sound the alarm” to the record industry when Spotify was about to put the final nail in the coffin of the CD, and turned the “tool” of streaming to his advantage. He found, it seems, a way to secure some kind of profitable equilibrium for himself and his artists under the new system. And good for him. But while the story goes that Spotify did pull record labels out of their piracy-induced nosedive and return them to profitability, it’s also reputed to have done so at the cost of financial stability for the artists themselves. The New York Times wrote in 2021 that whereas Spotify’s stated goal was to help a million artists make a living, the reality is that the streaming model mostly funnels money to labels and already-rich artists, and that at the time, only 13,000 total artists worldwide out of seven million total artists on Spotify—about two tenths of a percent—were receiving $50,000 or more in royalties. But Borchetta is the hero of his own story, and that’s the version he’s telling the graduates. When the room realizes he’s rhyming that story with AI—essentially telling them to arm themselves with AI the way he wielded streaming, and that one must use one’s weapons to slit throats before one’s own throat gets slit—you can clearly hear some of them revolt. The booing and yelling are barely audible in the video, but Borchetta reacts like the world’s smuggest, surfer-accented sea captain, worried he might have to put down a very un-chill mutiny:

“AI is rewriting production as we sit here (booing). I know it. Deal with it. Like I said, it’s a tool. (Angry jeering) Hey, like I said, you can, you can hear me now, or you can pay me later. Hey, then do something about it, ‘kay? It’s a tool. Make it work for you.” Naturally, he also compares AI to a genie, and asserts, “It ain’t goin’ back in the bottle.” The line, “you can hear me now, or you can pay me later,” is accompanied by a clenched-teeth grin, and Borchetta comes across like someone on a throne atop of a mountain of corpses, taunting the onlookers. His “do something about it,” sounds like a genuine challenge. On his own terms, Borchetta is not wrong. Of course the economic system we have disproportionally benefits the ruthless, and it certainly doesn’t bend to complaints that it’s unjust. To the victor go the spoils, and who can dispute that Borchetta is a victor?

A recent report from the New York Fed exposes the underlying logic of Borchetta’s speech for all its brutality. Borchetta is a CEO, and the plurality of CEOs self-report that they’re declining to hire young people in favor of older workers. Survey data also suggest that CEOs with the future on their minds are imagining small staffs. Meanwhile, 90% say they’re deploying AI in their companies in some way. So, again, you can’t fault Borchetta’s honesty. But when did it become good practice among commencement speakers to tell the newest workers entering the economy they partly control, essentially, your piece of the pie is getting smaller; I have it; so you’d better come and take it from me? Who wants to throw their hat in the air after hearing that?



Source link

WORLD
The Wall Street Journal reveals Apple’s profit-making techniques by using graded chips to push low-priced products to grab the market | International | Central News Agency CNA
jackminion 0



Please agree to our privacy policy to enable news listening. (Central News Agency, San Francisco, comprehensive foreign news report on the 18th) The Wall Street Journal reported that while most electronic device manufacturers are being hit hard by rising costs, Apple in the United States has been able to launch low-priced new products such as MacBook Neo and iPhone 17e to grab the market, relying on the reuse of defective chips that have been used for many years. The report pointed out that Apple (Apple) has long been admired by consumers for its high-end price products, but its new entry-level notebook computer MacBook Neo is priced at only US$599 (approximately NT$18,900), and preliminary sales data show that this product is very popular. Although the MacBook Neo is equipped with Apple’s A18 Pro chip, which is the same chip as Apple’s high-end smartphone iPhone 16 Pro launched in 2024, it has been reduced from 6 cores to 5 cores. The Wall Street Journal said that this shows that Apple can retain some A18 Pro chips with a defective core for future use, because the defective core can be deactivated, and then such chips can be used in other lower-priced devices and still work very well. In fact, Apple is using the chip industry’s approach of finding ways to squeeze profits from less efficient processors, by binning its supply into “good”, “better” and “best” products, similar to how it sells eggs, gasoline, diamonds or hotel rooms. Analysts pointed out that this industry strategy, which has been in place for decades, was originally just to make the best use of chips, but now it has become the basis of Apple’s product line design strategy, allowing it to be accurately segmented to achieve effects that smaller rivals cannot match. In addition to using the flexibility of self-developed chips to launch lower-priced iPhones and Macs, Apple also leverages its strong supply chain advantages to reuse chips to engage in price wars and attract new users. The MacBook Neo, for example, is cheap enough to poach consumers who might otherwise buy a Google Chromebook or other desktop computer (PC). The iPhone 17e is also equipped with a “graded” chip and is cheap enough to attract users who originally used phones equipped with Google’s Android operating system. Market research companies Counterpoint and International Data Corporation (IDC) both said that due to the sharp increase in the price of memory and storage devices, it has become unprofitable for Apple’s competitors to sell lower-end products. On the other hand, Apple has an advantage in comparison and can seize the market by launching lower-priced products. Tim Culpan, a supply chain analyst who has written an article analyzing MacBook Neo chip orders, said: “If you can continue to use things that are not up to the highest specifications, in addition to saving time, money and reducing waste, you can also sell them to many customers that you might not otherwise be able to reach.” As long as Apple adds new users to its device products, it may add buyers for its services such as cloud storage iCloud and the App Store, and this type of business is also more profitable. The Wall Street Journal analyzed nearly 200 pages of Apple filings and found that since 2021, six versions of Apple’s A-series chips have been used in lower-priced devices with one less core count, while full-core versions were initially installed in higher-priced iPhones. However, people familiar with Apple’s supply chain revealed that the MacBook Neo is now so popular that even the remaining chips that were originally reused are running out, forcing Apple to place an order for the new A18 Pro in the near future. However, because Apple has only one supplier of its most advanced chips, TSMC, and TSMC is struggling to cope with the crazy demand for artificial intelligence (AI) chips. Therefore, as Ming-Chi Kuo, an analyst at Tianfeng International Securities in Hong Kong, said: “Apple no longer enjoys the (supply) flexibility it had in the past, and pressure is beginning to appear.” (Compiled by Zhang Zhengqian) ) 1150518 Support the Central News Agency’s choice to stand with the facts. Every donation you make is a small amount of sponsorship to download the Central News Agency’s “First-hand News” APP to get the latest news in real time. The text, pictures, and audio and video on this website may not be reproduced, publicly broadcast, or publicly transmitted and used without authorization.



Source link