与「Interaction」相关的搜索结果

这样的 AI 人机交互才是未来！ 指哪改哪, 只要说话 PPT 就跟着改, 像是一个 PPT 专家坐在那, 你🫵指着屏幕让他改各种细节... Zoom Slides 这个体验着实惊艳到我了。 指针指着元素然后说话,改文字、翻译、问问题、换图、调布局、发评论 —— 一根指针补上了纯 Chat 缺的眼睛和手势, 歧义问题迎刃而解。 人机交互的范式真的在发生大变化, 相信越来越多产品会朝 Zoom Slides 这个方向走。 --- This is what the future of human-AI interaction looks like! Point at anything, change anything—just talk, and the slides update. It's like having a PPT expert sitting right next to you while you 🫵 point at the screen and tell them what to tweak. Zoom Slides genuinely blew me away. Point your cursor at an element and speak: edit text, translate, ask questions, swap images, adjust layout, drop comments. That one cursor adds the eyes and gestures that pure chat is missing—and the ambiguity just melts away. The paradigm for how we interact with machines is really shifting. I bet more and more products will head in the direction Zoom Slides is pointing.

很多人第一次看到 HootArk，会以为只是一个 Web3 浏览器。 但更准确地说，HootArk 是一款 Agentic Web3 Browser & Wallet 🦉 HootArk把移动浏览器、内置多链钱包、dApp 入口和 AI Copilot 放在同一个 App 里，让你可以从“看到信息”直接走到“完成链上操作”。 你可以像平时上网一样打开网页、看资讯、查项目，也可以直接进入 Web3 场景：访问 DeFi、NFT marketplaces、DAO tools 等 dApps，连接钱包，查看资产，发起转账或 Swap。不需要额外插件，也不用在浏览器和钱包之间来回切换。 HootArk内置了钱包模块。在创建或导入钱包后，可以直接无缝接入Ethereum、BSC、Polygon 等主流网络。对刚进入 Crypto 的用户来说是一个更简单的入口；对 DeFi 用户、NFT 玩家和 Web3 老用户来说，它也能让移动端操作更顺手 ⚡ AI 也是 HootArk 很重要的一部分，但它不是一个单独摆在旁边的聊天框。 你可以边浏览边提问，让 AI 帮你理解项目、网页和链上信息； 看到新闻、公告或长文章，也可以直接让它总结重点； 在交易或交互前，帮你核查事实信息； 它更像一个随时跟着你的 Web3 copilot，帮你少查几次资料，少切几个页面 🤖 安全和隐私方面，HootArk 始终坚持安全第一：私钥保存在本地设备上，平台不收集数据，不追踪用户。你的钱包、你的身份、你的数据，都应该由你自己控制 🔐 【重点来啦】 我们即将发布 HootArk Lite 版 🚀 期待更轻、更快、更智能的Web3 AI 移动浏览器🎉 欢迎持续关注HootArk A lot of people see HootArk for the first time and think it’s just a Web3 browser. But more accurately, HootArk is an Agentic Web3 Browser & Wallet 🦉 HootArk brings a mobile browser, a built-in multi-chain wallet, a dApp gateway, and an AI Copilot into one app, so you can move naturally from “finding information” to “taking on-chain action.” You can use it like a regular browser to open websites, read news, and check projects. But you can also jump straight into Web3: access DeFi, NFT marketplaces, DAO tools, and other dApps, connect your wallet, view your assets, and make transfers or swaps. No extra plugins needed. No constant switching between your browser and wallet. HootArk also comes with a built-in wallet module. After creating or importing a wallet, you can seamlessly connect to major networks like Ethereum, BSC, Polygon, and more. For people new to crypto, it’s a simpler entry point. For DeFi users, NFT collectors, and Web3 natives, it makes mobile on-chain actions much smoother ⚡ AI is also an important part of HootArk, but it’s not just a chatbot sitting on the side. You can ask questions while browsing and let AI help you understand projects, web pages, and on-chain information. When you see news, announcements, or long articles, AI can help summarize the key points. Before a transaction or interaction, it can also help you check relevant facts and context. It feels more like a Web3 copilot that stays with you, helping you search less, switch tabs less, and understand things faster 🤖 When it comes to security and privacy, HootArk always puts security first: private keys stay on your local device, the platform does not collect data, and users are not tracked. Your wallet, your identity, and your data should always be controlled by you 🔐 And here’s the big update: We’ll soon be releasing HootArk Lite 🚀 Stay turned! Browser + Wallet + dApp Gateway + AI Copilot. This is HootArk 🦉✨ #HootArk# #Web3# #Web3Browser# #Wallet# #dApps# #AI#

One aspect of @AntarcticWallet that deserves more attention is its focus on usability at scale. A lot of crypto products are designed for individual transactions. The real challenge begins when communities, businesses, and growing teams need to manage hundreds of users, payments, and interactions efficiently. That's where strong infrastructure matters. The most valuable tools are often not the ones with the longest feature lists, but the ones that simplify complex processes behind the scenes. Reducing operational friction, improving transaction management, and creating a smoother experience for users can have a significant impact over time. As digital finance continues to mature, platforms that prioritize practical utility and operational efficiency will likely play an increasingly important role. That's one of the reasons I'm interested in following the development of @AntarcticWallet.

GM, happy Wednesday CT ☕ Been looking at a few projects and what they’re actually trying to solve @TheARCTERMINAL is less about basic AI chat and more about agents, that can actually keep context and build on past interactions instead of resetting every time @quipnetwork is more infra focused direction distributed compute + long term security thinking around post quantum systems @useTria is just making crypto usage simpler across chains in one place without too much friction Explore Tria:

this is my personal singularity moment this post may sound like a paid ad. I only wish. I'm concerned, more so than happy. the world is changing, and, among the scenarios where AI goes terribly wrong, inequality is the most realistic, yet, the one Anthropic seems to be the least concerned about. I'm glad OpenAI is taking the opposite stance: *personal AGI for everyone*. I think this is a commendable position in the times we live. but who am I in the queue of the bread? anyway, Fable is here, so I'll just report my first-hour experience first of all, all my pet prompts are solved. → λ-calculus puzzles → bug questions → one-shot apps all are trivial to it. I don't have anything harder other than my ongoing work so, in the last several days, I've been toying with HVM5, a new interaction net evaluator with a faster loop. after writing the first version, I left 32 GPT-5 agents working for ~20 hours each. this resulted in up to 2x speedups, but the file size increased by 2-fold and quality decreased significantly. I then simplified the whole thing into an even simpler core, and left Opus 4.8 and GPT 5.5 optimizing it for 8 hours. Opus got a legit 6% - 34% speedup in most benches. GPT got better results, but, sadly, an unusable file. I then asked Fable to optimize it. 2 hours later, it landed a 1770% speedup in one case, 100%+ in other 4, and 22% in average. yes, in 2 hours it outperformed me, opus 4.8 and a swarm of gpt 5.5 agents, by one order of magnitude. that could not possibly be legit. "it must be hardcoding the benchmarks" (GPT trauma). so I read its explanation and what it did was, indeed, the most high impact optimization one could try first. seems like HVM5 was wasting a lot of time garbage-collecting unused branches of pattern-match nodes. I had optimized that for static mats, but not for dynamic mats. skill issue. Fable figured how to do it for these, resulting in a massive speedup in some benches but wait, is that *correct*? I'm not sure yet, it is credible, but this is the kind of thing that is very easy to get wrong on interaction nets. the problem is, when I was ready to start auditing Fable's solution so I could tell whether it was buggy or legit, it interrupted me to tell me it had found a massive bug on the code *I* had written. ... wait, what? so... for garbage collection purposes, I stored a bit on lambda term pointers that meant "the variable bound by this lambda has been freed, so, its lambda must free whatever argument it is applied to". that's fine. yet, on duplicator nodes, I also used the same bit to mean "one of the duplicated variables was freed, so, treat this dup as a passthrough no-op". so, if a lambda entered a duplicator, it would mistake the lambda's collection bit for its own, resulting in corrupted interaction! that's a mouthful, why I'm writing this? just so you can appreciate the sheer absurdity of what just happened. I didn't ask it to find bugs. I asked it for an optimization. and even if I did ask it to find bugs, this bug is so astonishingly subtle and specific, identifying it takes mastering the domain to an extent that it beyond even me. I'd easily need hours or days to fix it, *if* I ever came across it. chances are it would just go unnoticed. and Fable found it and fixed it like it was nothing, while it was busy adding a 17x speedup to a file that neither I, nor Opus 4.8, nor a fleet of GPT 5.5 managed to barely make 2x faster. oh and there is also another tab where it is also ripping through Bend's codebase and finishing everything I had to do I don't know what to say anymore this isn't about Anthropic or OpenAI, this is about our collective future as a species. the world is changing, and we need to be aware of it, and discuss how to handle this change. receipt below . . .

Elon Musk told a story that should terrify every AI company on Earth. His son Saxon is autistic. Saxon couldn’t understand why the family went to restaurants. You can get the same food delivered. You can call your friends over. You can eat better at home for half the price. So why go? Musk: “He had an epiphany and said, ‘Oh, the reason people go to restaurants is to hang out with strangers.’” A kid who takes the world literally just decoded something the rest of us never thought to question. We like being around people we’ll never know. Look at what we already built. Delivery apps so you never wait in line. Remote work so you never share an office. Self-checkout so you never talk to a cashier. Every innovation of the last 20 years was a bet against human proximity. Every one paid off. Until it didn’t. Loneliness is now a public health emergency. Depression has doubled since the smartphone. The average American has fewer close friends than any generation in history. We didn’t remove friction. We removed the thing friction was hiding. Now look at what’s coming. AI agents that handle your emails. AI companions that replace your conversations. AI assistants that make every human interaction optional. Same playbook. Same bet. Except this time we’re not engineering out strangers. We’re engineering out humans entirely. The coffee shop where nobody knows your name. The subway where no one speaks. The restaurant where you’ll never see that couple again. Those aren’t failed connections. They’re the background radiation of belonging. We don’t just need people who know us. We need to exist in rooms full of people who don’t. That’s what a kid understood at a dinner table that billion-dollar companies still can’t grasp in a boardroom. We spent 20 years building a world you never have to show up to. AI is about to finish the job. And nothing it builds will ever replicate sitting in a room full of strangers and not feeling alone.

Chinese company UBTECH Robotics has unveiled teasers of its U1 series humanoid robots, designed for the mass market The lineup includes two bionic humanoid models: one 183 cm tall and weighing 42 kg, and a smaller version at 168 cm and 35.2 kg. They feature 88 degrees of freedom, Wi-Fi support, and built-in AI for learning and interaction with the environment. Battery life is up to 4 hours. The full presentation is scheduled for June 30, but pre-orders are already open. According to the company, 1,943 units have been reserved.

Jensen Huang just told you you’re the slowest part of your own computer. And that he fixed it. For forty years, the entire architecture of personal computing depended on a single biological bottleneck. You had to click. You had to type. You had to translate every thought into the rigid language of the machine just to make it do anything. The computer was a passive terminal. It did nothing without explicit human permission. And buried inside that dependency was a word we never questioned. Personal. Your files. Your commands. Your keystrokes. That word meant total, uncontested human authority over a machine. Every interaction was permission-based. Every output was authored by you. That was the contract. Huang: “40 years later, Microsoft and NVIDIA are going to reinvent the PC. It took this long to completely reinvent how the PC is going to work.” He and Satya Nadella spent three years quietly dismantling that contract from the silicon up. No leaks. No breadcrumbs. Three years of silence before retiring the most important human-machine agreement in computing history. They didn’t build a faster processor. They assassinated the interface. The old PC was application-driven. You opened programs. Navigated file systems. Clicked through menus. We spent decades learning how to input. The machine finally learned how to listen. The new PC is agentic. It reasons. It anticipates. It generates. You don’t operate it. You deploy intent. And the moment that gap collapses, the biological intermediary isn’t the operator anymore. It’s the bottleneck. When a machine understands human context natively, the concept of a “user” ceases to exist. The PC was the last workspace where a human had complete control over a machine. No algorithm curating your attention. No feed ranking your reality. Just a blinking cursor and total authority. That space is being surrendered. Willingly. Enthusiastically. And marketed as progress. Because “personal” is about to mean its opposite. We spent forty years defining ourselves by how well we could operate the machine. Only to realize the machine was just waiting to operate itself.

🚨Report: McDonald's is planning to replace drive-thru attendants with an AI model named Archy. They were able to take 90% of orders without human interaction in a recent test, and they're preparing to roll it out for every McDonald's in the U.S.