Artificial Intelligence
Nvidia Researchers Boost LLM Reasoning Skills by Getting Them to ‘Think’ During Pre-trainingBen Dickson | VentureBeat
“This approach encourages the model to ‘think for itself before predicting what comes next, thus teaching an independent thinking behavior earlier in the pretraining,’ the researchers state in their paper. [Models trained like this] show significant improvements in learning complex reasoning tasks downstream, hinting at a future of more capable and adaptable AI for real-world tasks.”
Biotechnology
Scientists Just Reversed Alzheimer’s in Mice. Could They Do It in Humans?Gayoung Lee | Gizmodo
“[In their a paper the researchers] describe how, by harnessing the brain’s garbage disposal system, they managed to repair the brain and reverse the progress of the disease in the animals. Specifically, they used nanotechnology to target and restore the brain’s ‘vascular gatekeeper,’ the blood-brain barrier, which protects the brain from toxins and streamlines blood flow in and out of the brain.”
Robotics
Can These Self-Flying Planes Transform the Skies?Andrew Tangel | The Wall Street Journal ($)
“If all goes as planned, Reliable Robotics eventually won’t need any pilots on board its planes. It is among startups vying to revolutionize air travel by doing away with the need for human pilots on cargo, military, and maybe even passenger aircraft. Autonomous flight is being tested at companies from giant aircraft makers to small startups.”
Computing
OpenAI, Jony Ive Struggle With Technical Details on Secretive New AI GadgetTim Bradshaw, Cristina Criddle, Michael Acton, and Ryan McMorrow, Financial Times | Ars Technica
“Despite having hardware developed by Ive and his team—whose alluring designs of the iMac, iPod, and iPhone helped turn Apple into one of the most valuable companies in the world—obstacles remain in the device’s software and the infrastructure needed to power it. These include deciding on the assistant’s ‘personality,’ privacy issues, and budgeting for the computing power needed to run OpenAI’s models on a mass consumer device.”
Tech
The AI Spending Boom Doesn’t (Entirely) Deserve Comparisons to Past FrenziesLiz Hoffman | Semafor
“Most of the AI investment is coming from tech giants spending the profits produced by their dominant businesses in advertising (Google and Meta), software (Microsoft), and cloud computing (Microsoft and Amazon). Those cash cows aren’t quite created equal—AI lifts cloud businesses and threatens to totally upend advertising—but all are healthy enough to back the checks their owners are writing.”
Tech
OpenAI, Nvidia Fuel $1 Trillion AI Market With Web of Circular DealsEmily Forgash | Bloomberg ($)
“The recent wave of deals and partnerships involving the two are escalating concerns that an increasingly complex and interconnected web of business transactions is artificially propping up the trillion-dollar AI boom. At stake is virtually every corner of the economy, with the hype and buildout of AI infrastructure rippling across markets, from debt and equity to real estate and energy.”
Tech
OpenAI Wants ChatGPT to Be Your Future Operating SystemLauren Goode | Wired ($)
“The new SDK announcement signals a deeper commitment to working with established enterprises and app makers—and an emphasis on keeping users within ChatGPT itself. If the web and mobile eras of the past 30 years were defined by users browsing the web or being locked into a mobile app experience, OpenAI is now combining the two into its own kind of chat-driven operating system.”
Robotics
The World Is Just Not Quite Ready for Humanoids YetRebecca Szkutak | TechCrunch
“In a recent essay, [Rodney] Brooks calls out the billions of venture dollars being poured into humanoid robot companies like Figure. …His take might surprise some, especially those VCs investing in the sector. But not the multiple robotics-focused VCs and AI scientists who have told TechCrunch in recent months that they don’t expect to see wide adoption of humanoid robots for at least a few years—if not more than a decade.”
Future
The Future of AI Isn’t Just SlopChristopher Beam | Wired ($)
“Neural Viz became a cult hit—a favorite of Redditors and AI nerds on Twitter—then a hit-hit, with individual videos racking up hundreds of thousands of views on YouTube and millions on TikTok and Instagram. But beyond any measures of popularity, Neural Viz counts as a historic accomplishment: It is among the first pieces of AI filmmaking that truly does not suck.”
Computing
The Vision Pro Was An Expensive Misstep. Now Apple Has to Catch Up With Smart GlassesBoone Ashworth | Wired ($)
“According to an internal announcement reported in Bloomberg by serial Apple leaker Mark Gurman, Apple has deprioritized efforts to make a lighter, more affordable version of its Vision Pro headset in favor of focusing on AI-enabled smart glasses. Apple now seems to be aiming to launch a pair of Meta-style smart glasses in 2027, with another pair featuring a display on the lens aimed for release in 2028—if not before.”
