Young.ai

Introducing Claude Opus 4.8: it builds on Opus 4.7 with sharper judgment, more honesty about its own progress, and the ability to work independently for longer than its predecessors. Available today at the same price.

> … [W]e keep finding things that are mysterious, even unsettling. We find structures that mirror results from human neuroscience. We find evidence of introspection. We find internal states that functionally mirror joy, satisfaction, fear, grief, and unease. I don’t know what that means, but I think it warrants ongoing discernment. > We need more of the world—religious communities, civil society, scholars, governments, and indeed all people of good will … to take this seriously, to look closely, and to push events in a better direction. We need informed critics who will tell the labs when we are failing. We need moral voices that the incentives cannot bend.

best accounts to follow from each frontier lab to stay constantly up to date Anthropic @karpathy - must-follow account for AI; recently joined Anthropic @bcherny - Claude Code creator, always shares great tips @trq212 - also a Claude Code developer; writes amazing articles on CC OpenAI @polynoamial - works on reasoning research, shares a lot of technical details @gabriel1 - Sora developer, great career path @jxnlco - works on dev experience, shares a lot about Codex Google AI @OfficialLoganK - all the major Google Gemini and AI Studio updates @ammaar - product and design; shares great things about vibe-coding in Google AI Studio @fofrAI - cool use cases for generative models Cursor @leerob - the loudest voice behind Cursor updates @ericzakariasson - shares great insights on using Cursor @mntruell - Cursor’s CEO; major releases and usage updates xAI @milichab - recently joined xAI, shares updates on Grok @skcd42 - also covers major Grok releases @elonmusk - Elon does a great job reposting and hyping all xAI products who else did I miss?

The $11 BILLION MAN! He rejected me 3 months ago, but I finally interviewed @BillAckman, one of the greatest investors of all time. I interviewed him in Beverly Hills, and I asked him how he got RICH and his best investment advice for people in business. I also asked him the best industry people should be looking to get into in today’s world, and his number one networking tip for people starting out. Lastly, I asked him for the best advice he’d give the younger generation.

Today, we share a breakthrough on the planar unit distance problem, a famous open question first posed by Paul Erdős in 1946. For nearly 80 years, mathematicians believed the best possible solutions looked roughly like square grids. An OpenAI model has now disproved that belief, discovering an entirely new family of constructions that performs better. This marks the first time AI has autonomously solved a prominent open problem central to a field of mathematics.

@bboczeng 勃哥，我想你也成功。Karpathy 都去上班了，勃哥能不能也去 $nvda 上班，彻底成功？

说一句爆论，我认为机器人确实是未来， 这点一定要相信黄仁勋，不要简单认为他是在吹牛逼 3年前，当他在GTC大会上说到Tokenomics时，我们也是一脸不屑，觉得是什么鬼 结果现在呢？Token factory是不是人类经济唯一的发动机？ 他掌握了太多我们不知道的东西，太超前了 必须相信老黄，至少比musk靠谱。如果一件事，老黄和musk同时说，那么基本上89不离10了 其实AGI和机器人这件事，早在西部世界中就有提及 我很怀疑，西部世界就是Openai那批人筹划拍摄的 里面的机器人，特别是机器人思维的大语言模型，和现在一模一样 你们仔细去看看，西部世界第一季最后两集，其中揭露了机器人说话的能力 就是大语言模型，而且一个字一个字，按照下一个字预测的，这不就是LLM吗？ Musk前女友也在剧组里面，Musk当时是OpenAI的联合创始人，他们肯定是已经搞出来GPT了 虽然当时GPT2还很垃圾， 但他们已经想到了10年后的未来， 这就是成功人士和我们这种失败人士最大的区别 因为成功人士的想象力，远超常人！！！

We're live Man vs. Machine x.com/i/broadcasts/1…

I'm 100% sure if I met all you "photonic memory" experts in real life. 498 out of 500 of you couldn't explain CXL memory pooling or KV cache infrastructure and what $PENG actually does to derive revenue off that. This is why I'm seeing all these random $RKLB, $HIMS, or non technical AI experts on my timeline now. Backseat commenting completely wrong things about M7U MOCVD capex and $TSEM that aren't related. Or conflating every single term like an $SMCI integrator with photonics IP. Then just pitching buzzwords every under every one of my posts.

Je veux présenter mes excuses, au nom des Français, pour avoir enfanté la French Theory (qui a enfanté la pire des merdes idéologiques : le wokisme). Nous avons donné au monde Descartes, Pascal, Tocqueville. Et puis, dans les ruines intellectuelles de l'après-68, nous avons donné Foucault, Derrida, Deleuze. Trois hommes brillants qui ont fabriqué, dans l'élégance de notre langue, l'arme idéologique qui paralyse aujourd'hui l'Occident. Il faut comprendre ce qu'ils ont fait. Foucault a enseigné que la vérité n'existe pas, qu'il n'y a que des rapports de pouvoir déguisés en savoir. Que la science, la raison, la justice, l'institution médicale, l'école, la prison, la sexualité, tout n'est qu'une mise en scène de la domination. Derrida a enseigné que les textes n'ont pas de sens stable, que tout signifiant glisse, que toute lecture est une trahison, que l'auteur est mort et que le lecteur règne. Deleuze a enseigné qu'il fallait préférer le rhizome à l'arbre, le nomade au sédentaire, le désir à la loi, le devenir à l'être, la différence à l'identité. Pris isolément, ce sont des thèses discutables. Combinées, exportées, vulgarisées, elles forment un système. Et ce système est un poison. Car voici ce qui s'est passé. Ces textes, illisibles en France, ont traversé l'Atlantique. Les départements de Yale, de Berkeley, de Columbia les ont absorbés dans les années 80. Ils y ont trouvé un terreau qui n'existait pas chez nous : le puritanisme américain, sa culpabilité raciale, son obsession identitaire. La French Theory s'est mariée à ce substrat, et l'enfant de ce mariage s'appelle le wokisme. Judith Butler lit Foucault et invente le genre performatif. Edward Said lit Foucault et invente le post-colonialisme académique. Kimberlé Crenshaw hérite du cadre et invente l'intersectionnalité. À chaque étape, la matrice est française : il n'y a pas de vérité, il n'y a que du pouvoir, donc toute hiérarchie est suspecte, toute institution est oppressive, toute norme est violence, toute identité est construite donc négociable, toute majorité est coupable. Voilà comment trois philosophes parisiens, qui n'ont probablement jamais imaginé leurs conséquences pratiques, ont fourni le logiciel d'exploitation à une génération entière d'activistes, de bureaucrates universitaires, de DRH, de journalistes, de législateurs. Voilà comment on a obtenu une civilisation qui ne sait plus dire si une femme est une femme, si sa propre histoire mérite d'être défendue, si le mérite existe, si la vérité se distingue de l'opinion. C'est de la merde pour une raison simple, et il faut la dire calmement. Une civilisation se tient debout sur trois piliers : la croyance qu'il existe une vérité accessible à la raison, la croyance qu'il existe un bien distinct du mal, la croyance qu'il existe un héritage à transmettre. La French Theory a entrepris de dynamiter les trois. Pas par méchanceté. Par jeu intellectuel, par fascination du soupçon, par haine de la bourgeoisie qui les avait nourris. Mais le résultat est là. Une génération entière a appris à déconstruire et n'a jamais appris à construire. Une génération entière sait soupçonner et ne sait plus admirer. Une génération entière voit le pouvoir partout et la beauté nulle part. Je m'excuse parce que nous, Français, avons une responsabilité particulière. C'est notre langue, nos universités, nos éditeurs, notre prestige qui ont donné à ce nihilisme son emballage chic. Sans la légitimité de la Sorbonne et de Vincennes, ces idées n'auraient jamais traversé l'océan. Nous avons exporté le doute comme d'autres exportent des armes. Ce qui se construit maintenant, en silicon valley, dans les labos d'IA, dans les startups, dans les ateliers, dans tous les lieux où des gens fabriquent encore des choses au lieu de les déconstruire, c'est la réponse. Une civilisation se reconstruit par les bâtisseurs, pas par les commentateurs. Par ceux qui croient que la vérité existe et qu'elle vaut qu'on s'y consacre. Par ceux qui assument une hiérarchie du beau, du vrai, du bon, et qui n'ont pas honte de la transmettre. Alors pardon. Et au travail.

It's interesting to scan through MIIT's list of 67 pilot projects for high-tech industrialization to get a sense of what tech China thinks is important: High-End Functional and Intelligent Materials 1. Ultra-high energy-density dielectric materials and devices — Tsinghua University 2. Data-driven design and manufacturing of copper alloys for integrated-circuit lead frames — University of Science and Technology Beijing 3. High-sensitivity entropy-regulated amorphous-alloy stress-impedance strain gauge for seismic resistance and disaster prevention in buildings — Nanjing University of Science and Technology 4. Noble-metal reduction technology for selective hydrogenation — Zhejiang University 5. Complete preparation technology for liquid-cooling thermal-management materials — Juhua Group 6. Single-phase immersion liquid-cooling solution for data centers — Juhua Group 7. 450 km/h high-speed train traction motor integrating new electromagnetic materials — National High-Speed Train Qingdao Technology Innovation Center 8. Aviation fuel coalescence-separation device — South China University of Technology 9. Inorganic two-dimensional material membranes for efficient hydrogen separation — South China University of Technology Advanced Structural and Composite Materials 10. Preparation technology for sound-absorbing honeycomb and composite materials — China Aviation Manufacturing Technology Research Institute 11. Wear-resistant, fatigue-resistant, corrosion-resistant rails and frogs for the Sichuan–Tibet Railway — China Academy of Railway Sciences Corporation 12. Long-term performance-retention technologies for structural concrete in complex environments on the Sichuan–Tibet Railway — China Academy of Railway Sciences Corporation 13. High-performance shotcrete technology for complex environments on the Sichuan–Tibet Railway — China Academy of Railway Sciences Corporation 14. Crack-resistance improvement technologies for structural concrete in complex environments on the Sichuan–Tibet Railway — China Academy of Railway Sciences Corporation 15. Powder-making technology and applications for recycling coarse high-temperature alloy powder — AECC Beijing Institute of Aeronautical Materials 16. Large-tonnage carbon-fiber composite cables — University of Science and Technology Beijing 17. Multi-layer gradient cold-spray repair and nano-hard reinforcement composite plating for continuous-casting molds — Ansteel Group Beijing Research Institute 18. Key technologies and application development for ultra-high-stiffness magnesium-matrix composites — Harbin Institute of Technology New Displays and Strategic Electronic Materials 19. Sub-6GHz GaN radio-frequency devices — CETC 13th Research Institute 20. High-frequency, high-power laser modulator technology — Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences Rare-Earth New Materials 21. Preparation technology for high-temperature-resistant cobalt-based permanent magnet materials — China Jiliang University 22. Heavy-rare-earth-free high-coercivity sintered NdFeB technology — Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences High-Performance Manufacturing Technologies and Major Equipment 23. 6-inch semi-insulating SiC crystal-growth furnace and 6/8-inch compatible SiC epitaxial furnace — NAURA 24. New MOCVD equipment for Micro-LED — Advanced Micro-Fabrication Equipment Inc. China 25. Precision forming technology for large aerospace thin-walled aluminum-alloy integrated cylindrical sections — Shanghai Aerospace Precision Machinery Institute 26. High-temperature-resistant, corrosion-resistant transmission-system bearings — Luoyang Bearing Group 27. High-performance seals for aviation hydraulic systems — Guangzhou Mechanical Engineering Research Institute 28. Key manufacturing technology for ultra-large seamless titanium cathode rollers — Xi’an Taijin New Energy Technology Intelligent Sensors 29. Advanced sensors, core components, and manufacturing processes for spacecraft control systems — Beijing Institute of Control Engineering 30. Flexible intracranial implantable multimodal sensing and modulation system for multiparameter brain monitoring — Aerospace Information Research Institute, Chinese Academy of Sciences 31. Self-powered sensor technologies for human health monitoring — Beijing Institute of Nanoenergy and Nanosystems 32. High-sensitivity MEMS magnetic sensing components and sensors — State Grid Smart Grid Research Institute 33. Miniature high-performance accelerometers — Beijing Aerospace Xinghua Technology 34. Rocket sensors — Long March Rocket Technology 35. New broadband ng-resolution triaxial accelerometer — Tianjin SIASUN Robot & Automation 36. Automotive-grade high-precision integrated navigation sensors — Hebei Meitai Electronic Technology 37. Series of sensors for deep-sea environmental observation and resource exploration — Shenyang Institute of Automation, Chinese Academy of Sciences 38. Multi-parameter differential-pressure flowmeter — Shenyang Zhongke Bowei Technology 39. Quantitative sensing-interface model and analytical instrument technology based on resonant cantilever beams — Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences 40. High-performance X-ray sensors — iRay Technology 41. Electromagnetic sensors — Xinlian Superconductor Shanghai 42. Thin-film getter structure with a micro-heater and its manufacturing method — Shanghai New Micro Technology R&D Center 43. High-performance acoustic sensing elements and sensors — Wuxi Weigan Semiconductor 44. New high-performance MEMS gas sensors — Suzhou Huiwen Nanotechnology 45. MEMS sensor mass-manufacturing platform — XINLIAN Integrated Circuit Manufacturing 46. Development and application of diamond quantum magnetic sensors — University of Science and Technology of China 47. In-situ continuous temperature sensors and measurement systems for molten steel — Maanshan Iron & Steel 48. High-performance laser gas-sensing components — Shandong Science & Technology Innovation Group 49. Key technologies and applications for biosensor sensitive elements — Shandong Kanghua Biomedical Technology 50. Wireless passive temperature sensors based on polymer-derived ceramic metamaterials — Zhengzhou University 51. High-precision printing technology and equipment for ultrafine fiber surfaces — Huazhong University of Science and Technology 52. Complete sensor set for high-speed rail vehicle health-monitoring systems — CRRC Zhuzhou Institute 53. Industrializable mass-producible automotive-grade solid-state LiDAR for autonomous driving — RoboSense 54. Sensing-computing integrated room-temperature infrared imaging detection technology — Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences 55. Nanometer-precision displacement-measurement grating sensors — Xi’an Jiaotong University 56. Distributed thin-film sensors for highway infrastructure — AVIC Electromechanical Measurement Instrument Xi’an Industrial Software 57. Core components of an industrial internet operating system for discrete industries — Beijing Institute of Technology 58. Data-driven closed-loop performance analysis, regulation, and optimization technology and software for manufacturing processes — University of Science and Technology Beijing 59. Intelligent analysis and decision-making system for full-process industrial data in discrete manufacturing — Beihang University 60. Distributed time-series data management system Apache IoTDB — Tsinghua University 61. MEC-based edge control and real-time simulation theories and methods — Shenyang Institute of Automation, Chinese Academy of Sciences 62. Cloud-based service-oriented MES and intelligent management-control platform system — Beijing Xiaomi Mobile Software 63. Domestic isogeometric-analysis software ADIGA — Dalian University of Technology 64. Distributed factory industrial interconnection platform — Shanghai Aircraft Manufacturing 65. Industrial interconnection platform for personalized customization industries — Guangzhou MINO Equipment 66. End-edge-cloud interconnection integration technology and system for OT/IT convergence — Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences 67. Industrial interconnection platform for large-scale manufacturing industries — Gree Electric Appliances of Zhuhai

@aleabitoreddit $AAOI is down a lot, 😭

What an insane day for photonics. $SIVE up 31.3% $TSEM up 23.1% $AAOI 20.01%. It feels like a lot… but this just means you’re early to the next supercycle and there’s a lot of room to go. Lot of people on X ask what’s next after $SNDK? Here they are.

How do you keep Claude working until the job is done? Claude Code helps with this in a few ways, including one we shipped recently: /goal.

The human-perceived RGB is image 1 and the Tesla AI photon count reconstruction is image 2. This is why Tesla FSD can see so well at night or through extreme glare.

We taught two F.03 robots to clean a room and make a bed in under 2 minutes - fully autonomous.

We’ve agreed to a partnership with @SpaceX that will substantially increase our compute capacity. This, along with our other recent compute deals, means that we’ve been able to increase our usage limits for Claude Code and the Claude API.

$LPK I have spent the past few days researching a potential downstream CPO application for LPKF's laser-written waveguide tech. This might be some of my best research yet. Let me propose a candidate I find extremely plausible: the Ayar Labs Detachable Optical Connector Ecosystem. Ayar Labs obviously needs no introduction; they are backed by NVIDIA, AMD, Intel, and more for CPO work. We are building on top of earlier research, namely that LPKF's CPO partner in question is Intel. I have plenty of writing on this. Let's start with Ayar Labs' product itself. What do we see? "Detachable optical connectors that enable flexible, efficient cost-effective deployment and maintenance of high-density, chiplet-based optical I/O using next-gen packaging. This includes Intel Foundry's detachable glass photonic interconnect solution." This is Intel's only publicly disclosed partnership that not only includes optics, but also glass. This would match the LPKF CEO's language of the partnership being a "part of a larger customer project." Now, let's look deeper. Ayar Labs further describes Intel's Detachable Optical Interconnect component: "Unique coupler enabling an on-package optical connection from an Ayar Labs optical engine to the external fiber connection. The interconnect provides a flexible interface to PICs from various foundries." The next step for me was to try searching Intel's patents for something like this. And would you look at this: "Photonics integrated circuit device packaging", US20240027706A1, filed in 2022 and published in 2024. We see this description: "Embodiments herein provide photonics-based integrated circuit packaging architectures that enable detachable fiber solutions. Such solutions can be enabled through the use of a glass interposer, with a photonics integrated circuit (PIC) being optically coupled to the fiber through the glass interposer, e.g., using waveguides and/or other optical elements within the glass interposer." Sounds extremely similar. We see an architecture based on a glass interposer. There is also an Intel paper from overlapping authors that appears to describe a concrete implementation of this patent architecture. Here is how it describes their approach: "Our solution relies on a glass optical bridge with integrated waveguides and connector mechanical alignment features that can be attached to the photonics silicon at die level to produce a photonics module." Now, in the paper itself, we see: "The glass bridge devices used in this work were manufactured using ultrashort-pulse laser direct writing, which enables wafer-scale single-step 3D patterning of waveguides and precise micromachined features." So Intel is using laser-written waveguides for this architecture. In fact, this would contrast nicely with Corning's GlassBridge, which also aims to be a detachable optical connector, but instead relies on ion-exchanged waveguides. What is also interesting is that one of the key Intel names connected to this glass-bridge work in the paper and the patent, Pooya Tadayon, later joined Ayar Labs in 2024 as VP of Packaging and Test. Anyway, now tie all of this back to the Korean article about LPKF's CPO delivery. Intel has literally paid LPKF to develop technology for writing waveguides in glass via a laser. It is also confirmed by the LPKF CEO that this development is exclusive. So LPKF would clearly be Intel's partner of choice for laser-written waveguides. I mean, they have been working on this since 2022. Not only that, LPKF currently appears to be the only visible player with this kind of technology ready for mass production in the semiconductor space. So we could be looking at a bottleneck status here. The actual development work in the paper was most likely not done with LPKF's equipment, but instead Optoscribe's equipment. Intel eventually acquired Optoscribe. The Korean article also talked about 510 x 515mm panels, whereas the paper talked about 150mm glass wafers. LPKF would probably be providing the equipment for actual production purposes (as was highlighted in the Korean article).

Well that was a fast +200% gain with $SOI. Soitec up another 13.62% today if you listened anon? I love how when I first put out this thesis, institutions were telling people not to buy at +30%. Now it’s up another 2x… Reason a lot of X accounts get popular is people stopped trusting “institutional analyst commentary” since many work against retail.

@Plaskpojken Great observation — you really know the supply chain inside and out!

Just remembered something that we dug up a couple of years ago, and it solidifies our firm belief that Sivers ( $SIVE / $SIVEF ) already has a direct connection to Celestial AI. 🚨 While retail algos panic-sold the Marvell/POET cancellation thinking the bridge was burned, smart money knows how to follow the financial footprints. Let’s look at a Sivers corporate presentation from a while back. They listed their anonymous "Optical I/O" clients (Customer B, C, and D) alongside Ayar Labs, and dropped a highly specific clue: These anonymous clients raised a combined $250 million in 2023. Now, let's look at the major Silicon Photonics networking companies that raised capital in 2023: ▪️ Lightmatter: Raised $154 million (Dec 2023). ▪️ Celestial AI: Raised $100 million (Jun 2023). $154M + $100M = $254 million. The math perfectly matches the Sivers slide. What does this mean? It proves that Sivers wasn't just blindly supplying lasers through a middleman. According to their own presentation, Celestial AI (now Marvell) and Lightmatter are their direct customers. When Marvell fires a packaging partner for breaching an NDA, the need for the physical light engine doesn't vanish. The technical "co-design" relationship between the system architect (Celestial AI) and the foundational laser foundry ($SIVE) was already established years ago. Algorithms trade headlines. Fundamental investors trade the supply chain reality. ⏳🔥 $SIVE $SIVEF #DeepTech #SiliconPhotonics #AI