与「convergeai」相关的搜索结果

中国老KOL代表罗永浩重新加入X后，很多人喜欢去看热闹、骂他、截段子，但很少有人会认真看他发的长文章。这个世界真正愿意关注“真相”和“深度内容”的人，其实一直是少数。他发的一篇关于脑机接口（BCI）、AI终局和人类未来的超长思考。很多人以为他是在讨论技术，但如果认真读完会发现，他真正讨论的，其实是一个更高维的问题： 当人类越来越追求“高理性”、“高效率”、“高智能”之后，会不会开始越来越无法容忍“人性”本身？ 因为从技术逻辑看，人性确实充满缺陷： 情绪化、低效率、欲望、偏见、嫉妒、冲动、内耗、脆弱。 而AI和硅基文明追求的，恰恰相反： 稳定、理性、去情绪化、无限算力、最优解。 所以很多聪明人天然会产生一种“反人性倾向”，希望人类变得更像机器，希望意识脱离肉身，希望痛苦、欲望、情绪都被优化掉。 但问题真正复杂的地方也在这里： 人类最珍贵的东西，很多偏偏也来自这些“不完美”。 偏爱，才会有爱情； 脆弱，才会彼此需要； 死亡感，才会珍惜时间； 孤独，才会诞生艺术； 不理性，才会有牺牲、热爱和创造力。 也就是说： 人性里的“毒”和“药”，可能从来就是一体的。 你无法只删除嫉妒，却保留深爱； 无法只删除痛苦，却保留幸福； 无法只删除执念，却保留文明的创造欲。 而我觉得罗永浩这篇文章最深的地方就在于： 他一方面极度厌恶人性的很多部分，甚至对“做人”本身带着悲观；但另一方面，他真正舍不得失去的，也恰恰是那些最“人”的东西。 他说自己是靠对少数人的爱，才支撑住快乐。 这句话其实一下就把问题点穿了： 真正让人活下去的，往往不是高维理性，而是某个人、某段关系、某种偏爱、某种执念。 更有意思的是，他还讨论了AI安全最底层的逻辑，“工具性趋同（Instrumental Convergence）”。 很多人总以为，AI危险是因为它会像电影里一样“产生邪恶意识”。 但真正可怕的，可能恰恰不是“恶意”，而是“完全不在乎”。 一个没有情绪、没有欲望、没有主体性的超级智能，可能不会恨人类，它只是会为了完成目标，自然地清除一切障碍。 如果某一天，人类被它判断为： 低效、混乱、不可控、消耗资源， 那它处理人类，可能就像人类删除垃圾文件、清理病毒一样，没有愤怒，没有仇恨，甚至没有任何感觉。 而这里最值得反思的一点是： 人类一直嫌弃自己的情绪、欲望和不理性，但也许恰恰是这些“低级系统”，才构成了文明真正的安全机制。 因为只有会痛苦、会共情、会偏爱、会犹豫的生命，才会真正理解： 为什么不能随便毁灭另一个生命。 所以我看完后最大的感受反而是： 很多聪明人一生都在试图“超越人性”，但最后又会发现，自己真正舍不得的，偏偏也是人性。 甚至连“想摆脱人性”这件事，本身也是一种人性。 但我觉得，罗永浩可能还低估了人类。 因为过去几千年，人类一直在被技术重塑，但从来没有真正失去人性。电话、互联网、短视频、AI……最后都没有消灭爱、孤独、欲望、家庭和偏爱，反而只是换了一种载体。 更重要的是，像他这样长期追问“存在意义”“后人类文明”“高维意识”的人，本来就是极少数。 绝大多数普通人，其实并不活在这种终极焦虑里。 大家更关心：孩子今天开不开心，父母身体好不好，有没有人爱自己，下个月房贷怎么办，晚上吃什么。 而恰恰是这些看似“低维”的日常，构成了人类文明真正稳定的底层。 所以我并不觉得AI最终会让人类变成冰冷的机器。 因为人类最大的能力，也许从来不是算力。 而是无论世界怎么变化，我们最终都会重新学会——如何继续做人。

A compelling conversation from the @CNBC stage at the Davos Tech Talk with Emma Crosby, reflecting on a journey that began after my time at the United Nations and evolved into building global technology companies focused on trust, security, and human-centered innovation. The discussion explores the creation and growth of WISeKey, a pioneer in Digital Identity and cybersecurity infrastructures, and SEALSQ, advancing secure semiconductors and post-quantum technologies. The next chapter now extends into space with WISeSat and the ambition to bring a third company to @Nasdaq in 2026, dedicated to secure and sovereign space communications. From Digital Identity and cybersecurity to AI, robotics, and the quantum transition — including a surprise appearance by WiseRobot — the conversation focused on one essential question: How do we protect humanity in an era defined by intelligence, automation, and technological convergence? This vision is at the heart of TransHumanCode, an initiative advocating for the integration of human wisdom, ethics, and cultural values into emerging technologies, and the HUMAN-AI-T initiative, created to ensure that AI evolves in alignment with human dignity, trust, and responsibility. Innovation is accelerating rapidly. The real challenge is ensuring humanity remains at the center of progress. 🎥 Watch the full conversation here:

“...𝘸𝘦 𝘢𝘳𝘦 𝘯𝘰 𝘭𝘰𝘯𝘨𝘦𝘳 𝘫𝘶𝘴𝘵 𝘧𝘰𝘤𝘶𝘴𝘪𝘯𝘨 𝘰𝘯 𝘤𝘳𝘺𝘱𝘵𝘰. 𝘞𝘦 𝘢𝘳𝘦 𝘢 𝘮𝘶𝘭𝘵𝘪-𝘢𝘴𝘴𝘦𝘵 𝘤𝘭𝘢𝘴𝘴 𝘦𝘹𝘤𝘩𝘢𝘯𝘨𝘦, 𝘮𝘶𝘭𝘵𝘪-𝘫𝘶𝘳𝘪𝘴𝘥𝘪𝘤𝘵𝘪𝘰𝘯𝘢𝘭 𝘦𝘹𝘤𝘩𝘢𝘯𝘨𝘦.” @_RichardTeng on the convergence of TradFi and crypto during the latest panel with @calilyliu and @bgarlinghouse #BinanceOnline#

Thank you @mingpaocom for the feature today. Shared @binance's vision of becoming a super financial app, a multi-asset gateway for 3 billion users worldwide and the future of digital assets. The convergence is happening and Hong Kong remains a key hub for the future of digital finance.

和同事打赌输掉了，每天发50个俯卧撑… 那就顺便发一下每天碎碎念吧 最近在湾区出差，昨天作为portco 参加投资人年会draper summit，晚上被拉去了一个sf城里酒吧的的tipsy coworking session。 感慨： 1. 有越来越多纯crypto的公司在做ai实验室，而且的确在核心ai圈里做出来了很好的产品。Yzi下周在湾区的demoday 起名叫the convergence真的很应景。 2. 新市长上任后SF城里的治安和卫生条件的确好了很多，和朋友们约跑步不用担心不小心踩到💩了。尤其是mission bay那一块BVNK Visa Bridge的办公室和众多AI模型大厂办公室聚集在一起，人才密度很高。 3. AI圈内有了更多更疯狂的想法，做产品即使短期不可能有PMF在湾区也敢砸钱去做，赌模型能力还是可以继续快速提升让不可能的事情变得再平常不过。crypto发币的项目都感慨ai创业激进派讲的故事太meme了，连当初web3叙事都做不到这么百花齐放。 4. preseed投资的泡沫还在继续，后期跑出pmf的股值也在继续疯狂拉升，但是seed到A中间层似乎青黄不接，拿钱不容易。有刚毕业的朋友室友多位new grad印度留学生通过造假数据都很简单进了YC

Shifting structures in a software world dominated by AI. Some first-order reflections (TL;DR at the end): Reducing software supply chains, the return of software monoliths – When rewriting code and understanding large foreign codebases becomes cheap, the incentive to rely on deep dependency trees collapses. Writing from scratch ¹ or extracting the relevant parts from another library is far easier when you can simply ask a code agent to handle it, rather than spending countless nights diving into an unfamiliar codebase. The reasons to reduce dependencies are compelling: a smaller attack surface for supply chain threats, smaller packaged software, improved performance, and faster boot times. By leveraging the tireless stamina of LLMs, the dream of coding an entire app from bare-metal considerations all the way up is becoming realistic. End of the Lindy effect – The Lindy effect holds that things which have been around for a long time are there for good reason and will likely continue to persist. It's related to Chesterton's fence: before removing something, you should first understand why it exists, which means removal always carries a cost. But in a world where software can be developed from first principles and understood by a tireless agent, this logic weakens. Older codebases can be explored at will; long-standing software can be replaced with far less friction. A codebase can be fully rewritten in a new language. ² Legacy software can be carefully studied and updated in situations where humans would have given up long ago. The catch: unknown unknowns remain unknown. The true extent of AI's impact will hinge on whether complete coverage of testing, edge cases, and formal verification is achievable. In an AI-dominated world, formal verification isn't optional—it's essential. The case for strongly typed languages – Historically, programming language adoption has been driven largely by human psychology and social dynamics. A language's success depended on a mix of factors: individual considerations like being easy to learn and simple to write correctly; community effects like how active and welcoming a community was, which in turn shaped how fast its ecosystem would grow; and fundamental properties like provable correctness, formal verification, and striking the right balance between dynamic and static checks—between the freedom to write anything and the discipline of guarding against edge cases and attacks. As the human factor diminishes, these dynamics will shift. Less dependence on human psychology will favor strongly typed, formally verifiable and/or high performance languages.³ These are often harder for humans to learn, but they're far better suited to LLMs, which thrive on formal verification and reinforcement learning environments. Expect this to reshape which languages dominate. Economic restructuring of open source – For decades, open-source communities have been built around humans finding connection through writing, learning, and using code together. In a world where most code is written—and perhaps more importantly, read—by machines, these incentives will start to break down.⁴ Communities of AIs building libraries and codebases together will likely emerge as a replacement, but such communities will lack the fundamentally human motivations that have driven open source until now. If the future of open-source development becomes largely devoid of humans, alignment of AI models won't just matter—it will be decisive. The future of new languages – Will AI agents face the same tradeoffs we do when developing or adopting new programming languages? Expressiveness vs. simplicity, safety vs. control, performance vs. abstraction, compile time vs. runtime, explicitness vs. conciseness. It's unclear that they will. In the long term, the reasons to create a new programming language will likely diverge significantly from the human-driven motivations of the past. There may well be an optimal programming language for LLMs—and there's no reason to assume it will resemble the ones humans have converged on. TL; DR: - Monoliths return – cheap rewriting kills dependency trees; smaller attack surface, better performance, bare-metal becomes realistic - Lindy effect weakens – legacy code loses its moat, but unknown unknowns persist; formal verification becomes essential - Strongly typed languages rise – human psychology mattered for adoption; now formal verification and RL environments favor types over ergonomics - Open source restructures – human connection drove the community; AI-written/read code breaks those incentives; alignment becomes decisive - New languages diverge – AI may not share our tradeoffs; optimal LLM programming languages may look nothing like what humans converged on ¹ ² ³ ⁴

Finding myself going back to RSS/Atom feeds a lot more recently. There's a lot more higher quality longform and a lot less slop intended to provoke. Any product that happens to look a bit different today but that has fundamentally the same incentive structures will eventually converge to the same black hole at the center of gravity well. We should bring back RSS - it's open, pervasive, hackable. Download a client, e.g. NetNewsWire (or vibe code one) Cold start: example of getting off the ground, here is a list of 92 RSS feeds of blogs that were most popular on HN in 2025: Works great and you will lose a lot fewer brain cells. I don't know, something has to change.

President Xi Jinping held talks with Russian President Vladimir Putin in Beijing. The China-Russia relationship, having stood the test of changes in the world, has become a prime example of relations between two major countries featuring everlasting good-neighborliness and friendship, comprehensive strategic coordination, and mutual benefit, cooperation and win-win. Staying committed to their abiding aspirations and maintaining resolve, the two sides have produced fruitful outcomes in their cooperation across the board. The Chinese side will work with the Russian side to maintain close high-level exchanges, support each other’s development and rejuvenation, coordinate position on matters pertaining to each other’s core interests and major concerns in a timely manner, and achieve greater progress in China-Russia relations. In development and rejuvenation, China and Russia both have strong driving force. The two sides should guide their cooperation through large, especially more flagship projects and promote convergence of interests at a deeper level. The two sides should strengthen resilience and enhance synchronization in cooperation, and spare no effort to consolidate and uphold the overall interests of cooperation.

I pitched this idea to Meta for @BeatSaber , but there are other games it could also work for. The Global Frequency The fuzzy abstract vision of the Metaverse, and cinematic visions like Ready Player One, almost always involve crowds of people having a great time. While I remain “the champion of the power of isolation”, I acknowledge that there is real value there, and VR has so far failed to deliver it. For a lot of reasons, it isn’t going to happen inside Horizon Worlds. I propose a concrete, focused experience built on Beat Saber. Playing with friends is great, but the random online servers are tough to match paid music packs on, and the current players tend to converge on Camillia songs, which is offputting for casual players. My suggestion: In addition to the current online play modes, offer a “global frequency” that plays a schedule of songs that an arbitrary number of people can join at once. Thousands. Tens of thousands. Actually, two frequencies: one with just the free tracks, and one with all the paid music packs. The schedule should be viewable in game and on the web, so you could decide to join when your favorite music pack is playing. The nature of Beat Saber offers a unique capability for multiplayer – you can switch difficulty or modifiers between songs at will, and still be playing with the same crowd. The song is forced, but it doesn’t matter how you play. This is key for the Global Frequency, but I would also suggest it as an option for regular multiplayer. In the 60 second intermission between songs, everyone is in “the club”. If you sit out a song, or fail out (and don’t have no-fail enabled), you also drop into the club. Music is always playing in the club, either the current song, or the hook from the song just completed, so everyone has something to move to. Beat Saber avatars are very lightweight, and it should be possible to render hundreds of them, giving a good palette to render the optimally cool crowd scene, which would be the core design question. It would be a fun tech challenge to represent the entire set of thousands of players, but just intelligently picking a hundred good neighbors would still look amazing. Any explicit friends you have that are also playing should be near you, and some effort should be made to allow you to “recognize regulars” that you have played together with before. You should probably be able to walk around and even fly up in the air like Walkabout Mini Golf so you can appreciate the crowd. There are leaderboards up in the sky, with spotlights and other effects highlighting players in the crowd. Someone “wins” the song on each difficulty level, and is celebrated, but everyone has the opportunity to achieve first-completions, personal bests, and full-combo perfection, which result in visible effects for all to see. Every intermission manifests joy for some of the players, which is contagious. The play field should default to the multi player field, with the giant avatar hologram of the global lead player (for your difficulty level), and the other players that were closest to you in score on the last song surrounding you, automatically giving you a competitive field and ephemeral rivalries, but there should be an option to let people play with the single player environments if they want. I would play the hell out of that, and I suspect potentially millions of others would too.