Always Be Curious #245: TSMC's AI boost, nuclear microreactors, and insectoid microrobots
This week in ABC: TSMC's has a commanding industry lead and plans to invest this year, nuclear power is back, and micro robots inspired by insects are around the corner
Sup Curious Clan! ✌️ This week, TSMC reported its Q4 2024 earnings and handily beat expectations as demand for leading-edge chips for AI applications continues to surge. The Taiwanese juggernaut’s revenue from High Performance Computing grew 19% in Q4 and surprisingly, Mobile also grew with a double digit 17%. The revenue out of their advanced 3 nanometer process grew a mindblowing 44% in comparison to Q3, and it now makes up over a quarter of TSMC’s total revenue. And to think that this technology wasn’t even there just one-and-a-half years ago. Isn’t that crazy? TSMC’s outlook for 2025 is a revenue growth of roughly 25%. Its investment plans are good news specifically for chip equipment makers: the Taiwanese giant intends to grow their capital expenditure to around $38 to $42 billion, representing a solid boost of 40% over last year.
Let’s. GOOOOOOO! 🚀
Have a good week, stay safe and sound,
👨💻The round-up in sci-tech💡
🚀 Roar of New Glenn’s engines silences skeptics of Bezos’ Blue Origin (The New York Times) 🎁
The successful flight to orbit of the Amazon founder’s powerful rocket suggests it could grow into a credible competitor with Elon Musk’s SpaceX.
⚛️ Two nuclear microreactors reach milestone (IEEE Spectrum)
Westinghouse and Radiant are developing some of the world’s smallest reactors
🗣️ Meta AI creates speech-to-speech translator that works in dozens of languages (Nature)
Machine-learning system can process words spoken in 101 languages, spitting out voice-synthesized translations in 36 target languages.
✨ Engineering a new dark matter detector (IEEE Spectrum)
Machine learning could boost quantum sensor sensitivity to dark matter.
🎓 Elon Musk says all human data for AI training ‘exhausted’ (The Guardian)
Tech boss suggests move to self-learning synthetic data, though some warn this could cause ‘model collapse’.
🎧 The Scientist vs. the Machine (Good On Paper)
Amid handwringing about AI’s effect on jobs, creativity, trust, and the environment, a new study shows the technology’s profound impact on scientific productivity. Aidan Toner-Rodgers, a Ph.D. candidate at MIT, recounts his research that shows the benefits and drawbacks of using AI to discover new scientific materials.
🤓This week in chips⚠
📈 TSMC fourth-quarter results top expectations, net profit surges 57% on robust AI chip demand (CNBC)
TSMC is king in an AI world.
🔥 TSMC says 1.6nm chips will be in volume production in 2026 (The register)
More innovation is coming.
😎 100x defect tolerance: how Cerebras solved the yield problem (Cerebras)
“Conventional wisdom in semiconductor manufacturing has long held that bigger chips mean worse yields. Yet at Cerebras, we’ve successfully built and commercialized a chip 50x larger than the largest computer chips – and achieved comparable yields. This seeming paradox is one of our most frequently asked questions: how do we achieve a usable yield with a wafer-scale processor?”
🇨🇳 China makes inroads in DRAM chips in challenge to Samsung and Micron (Nikkei)
New players seek to break global monopoly in volatile memory market
💰 Tech supplier Arm plans to hike prices, has considered developing its own chips (Reuters)
New players seek to break global monopoly in volatile memory market
🛑 White House unveils new curbs on exporting NVIDIA AI chips (Bloomberg) 🔐
The White House unveiled sweeping new limits on the sale of advanced AI chips by Nvidia Corp. and its peers, leaving the Trump administration to decide how and whether to implement curbs that have encountered fierce industry opposition.
🗯️ NVIDIA statement on the Biden administration’s misguided 'AI diffusion' rule (NVIDIA)
“For decades, leadership in computing and software ecosystems has been a cornerstone of American strength and influence worldwide. That global progress is now in jeopardy. The Biden Administration now seeks to restrict access to mainstream computing applications with its unprecedented and misguided “AI Diffusion” rule, which threatens to derail innovation and economic growth worldwide.”
🛑 Netherlands tightens export controls on sanctioned semiconductor equipment (Tom’s Hardware)
“The Netherlands has once again tightened export controls to ensure that ASML will have to apply for export licenses with the Dutch government instead of the United States. These new rules will come into effect on April 1st and is seemingly The Hague’s response to the updated list of sanctioned equipment.”
🛑 Dutch government excludes most ASML sales to China from 'dual use' export data (Reuters)
“The Dutch government excludes billions of euros of sales by technology company ASML to China from disclosures on sensitive goods exports, it has told Reuters, following a policy decision that has not been previously reported. The move is of interest because disclosure of exports of "dual use" goods with potential military applications had previously been routine in the Netherlands, and experts who rely on public data to understand states' military capabilities, including parliament, may no longer have the full picture.”
📈By the numbers📉
🔥 TSMC report Q4 and full-year 2024 financial earnings (TSMC)
“Our business in the fourth quarter was supported by strong demand for our industry-leading 3nm and 5nm technologies,” said Wendell Huang, Senior VP and Chief Financial Officer of TSMC. “Moving into first quarter 2025, we expect our business to be impacted by smartphone seasonality, partially offset by continued growth in AI-related demand.”
🔥🔥🔥 The formation of a semiconductor supercycle (Claus Aasholm)
TSMCs Stellar Result and What It Means to the Semiconductor Industry
❤️For the love of tech❤️
📸 Camera inspired by insect vision hits 1920 frame/s (eeNews Europe)
A high speed camera developed by KAIST in South Korea can capture 1920 frame per second with high quality and is just 1mm thick.
🐝 This fast and agile robotic insect could someday aid in mechanical pollination (MIT)
New insect-scale microrobots can fly more than 100 times longer than previous versions. The new bots, also significantly faster and more agile, could someday be used to pollinate fruits and vegetables.
Always Be Curious is the personal newsletter of Sander Hofman, Senior Creative Content Strategist at ASML. Opinions expressed in this curated newsletter are my own and do not necessarily reflect those of my employer.