December 3 2024
Musk's pay; Daylight Saving Time; Job switch frustration; OpenAI's big aspirations; AI power demand; U.S. tightens Chinese AI chip rules; Cartels recruit chemistry students; Viral TikTok stick figure; "new phone who dis?"
FLASH…Musk’s Pay Package Is Rejected Again…“Department of Government Efficiency” considers ending daylight saving time…
1 - More Americans Eye Job Switch, but Job Market Woes Persist
2 - OpenAI Bets on New AI Products and Apple Partnership to Expand Reach
3 - AI Revolution Will Require Equivalent of Country of Japan Energy Demand by 2026
4 - U.S. Struggles to Block China’s AI Chip Advances Despite New Restrictions
5 - Mexican Cartels Recruit Chemistry Students to Strengthen Fentanyl Production
6 - Viral TikTok Stick Figure Takes 10 Months and 700 Miles to Create
December 3, 1992: First SMS text message is sent
See the new Ad Astra Podcast! Released on Apple and Spotify around 10a CST.
FLASH…Musk’s Multibillion-Dollar Pay Package Is Rejected Again by Judge…“Department of Government Efficiency” considers ending daylight saving time…
1 - More Americans Eye Job Switch, but Job Market Woes Persist
More Americans are looking to switch jobs than at any point in the past decade. In a cooling job market, that’s a lot easier said than done. White-collar hiring continues to slow, but workers’ restlessness to find new work is intensifying, new Gallup data show. More than half of 20,000 U.S. workers surveyed in November said they were watching for or actively seeking a new job. That’s the largest share since 2015, eclipsing the so-called Great Resignation of 2021 and 2022, when millions of people quit jobs for better ones. The result? Job satisfaction has fallen to its lowest level in recent years as employees feel more stuck—and frustrated—where they are, according to Gallup, whose quarterly surveys are widely viewed as a bellwether of workplace sentiment. Smaller raises and fewer promotions are spurring some of the discontent, workers say. So are cost-cutting moves and stepped-up requirements to be working in offices more often.
2 - OpenAI Bets on New AI Products and Apple Partnership to Expand Reach
OpenAI is betting on a suite of new AI products, building its own data centres and a crucial partnership with Apple to supercharge its next phase of growth, as it targets reaching 1 billion users over the coming year. The San Francisco-based group, whose popular ChatGPT chatbot has rocketed to 250mn weekly active users since its launch two years ago, plans to expand further through launching so-called AI “agents”, its own AI-powered search engine and ChatGPT’s integration with Apple devices. “[In 2025] we will be coming into our own, as a research lab serving millions . . . hoping it can be billions of consumers around the world,” Sarah Friar, the company’s chief financial officer, told the Financial Times. The goal comes as the nine-year-old start-up recasts itself as global technology giant and prepares for what founder and chief executive Sam Altman describes as the “Intelligence Age.”
3 - AI Revolution Will Require Equivalent of Country of Japan Energy Demand by 2026
Fifteen years after his death, David Blackwell’s name will be on every tech nerd’s lips in 2025. Nvidia, a semiconductor giant, has named its latest superchip after the mathematician and game theorist, who was the first African-American inducted into the National Academy of Sciences. The Blackwell chip’s debut, with full-scale production starting in early 2025, has provoked breathless anticipation. It will form the backbone of the first data centres purpose-built for the era of generative artificial intelligence (ai). Nvidia’s boss, Jensen Huang, calls them “ai factories”. But it will also draw attention to the bottlenecks—from the making of chips to the construction of data centres—created by demand for ai-related computational power. Blackwell is a graphics processing unit (gpu), the kind of ai workhorse chip that has turned Nvidia into a $3trn giant. The superchip is integral to a new Nvidia platform that will prompt some of the biggest ai firms, including the cloud-service “hyperscalers”—Amazon, Microsoft and Google—to build new server farms for generative-ai computation on a huge scale, with unprecedented energy requirements. Demand is booming. Baron Fung of Dell’Oro, a research firm, estimates that sales of servers for generative-ai tasks will rise by more than three-quarters in 2025 compared with 2024, reaching $147bn, mostly thanks to hyperscalers’ spending on Blackwell gpus. As a result, energy demand is also surging. Lucas Beran, also of Dell’Oro, says data centres used to have power requirements of 100-200 megawatts (mw), but big ones now require 300-500mw. The International Energy Agency (iea), a forecaster, estimates that over the next two years, global power consumption from data centres could more than double from its 2022 level, reaching 1,000 terawatt hours by 2026—equivalent to the electricity consumption of Japan. Expanding the grid is hard. Ensuring that electricity is carbon-free is harder still. There are plans to restart a nuclear power station at Three Mile Island in Pennsylvania to power several Microsoft data centres, and Google has ordered nuclear reactors from Kairos Power, a startup.
4 - U.S. Struggles to Block China’s AI Chip Advances Despite New Restrictions
The U.S. introduced its latest restrictions on transferring advanced chips to China, but the delay in cutting off chips useful in artificial intelligence showed how the Biden administration has struggled to stall Beijing’s advances. The rules, the fourth attempt in three years by U.S. policymakers to curb China’s access to cutting-edge semiconductor technology, limited the sale of memory chips that power AI applications and narrowed the suite of chip-making tools available to China. The Commerce Department also announced the addition of 140 Chinese companies and other entities to its trade blacklist. Industry analysts said a monthslong pause between the drafting of the rules and their release Monday allowed Chinese entities to stock up on semiconductors and machines they knew were likely to be restricted.
5 - Mexican Cartels Recruit Chemistry Students to Strengthen Fentanyl Production
In their quest to build fentanyl empires, Mexican criminal groups are turning to an unusual talent pool: not hit men or corrupt police officers, but chemistry students studying at Mexican universities. People who make fentanyl in cartel labs, who are called cooks, told The New York Times that they needed workers with advanced knowledge of chemistry to help make the drug stronger and “get more people hooked,” as one cook put it. The cartels also have a more ambitious goal: to synthesize the chemical compounds, known as precursors, that are essential to making fentanyl, freeing them from having to import those raw materials from China. If they succeed, U.S. officials say, it would represent a terrifying new phase in the fentanyl crisis, in which Mexican cartels have more control than ever over one of the deadliest drugs in recent history. “It would make us the kings of Mexico,” said one chemistry student who has been cooking fentanyl for six months.
6 - Viral TikTok Stick Figure Takes 10 Months and 700 Miles to Create
The video is only 27 seconds long, but it took Duncan McCabe 10 months — and about 700 miles — to make it. McCabe, an avid runner and animation enthusiast, set out on a journey in January to create a now-viral TikTok of a dancing stick figure. He made it using the popular GPS-tracking app Strava, which connects runners, cyclists, hikers and walkers and allows them to record their routes. Using Strava’s map function, McCabe recorded 120 runs — and when he strung each of the maps together, it revealed a hat-wearing stick figure dancing to the song "Purple Hat” across the streets of Toronto.