Micheal Jordan Reddio ♦️
983 posts

Micheal Jordan Reddio ♦️
@Jordan58016764
绿油油的大韭菜(互关必回)Reddio user ♦️


很多人错过Sahara AI第一轮封闭测试,这很正常,78万人只有1万人有资格。错过没关系,第二轮开始报名了:hi.saharalabs.ai/get-started ✅今天我们主要来讲清楚:为什么一定要关注Sahara AI?不然还会错过第二轮 🔸先说结果:Sahara AI是 $TAO 之后,币安布局的另一个压舱石类AI平台。 1⃣ 怎么关注到Sahara AI的? 关注到Sahara AI并不是因为他获得了Binance labs + Polychain+ Pantera三大机构领投的4300万美元融资,虽然他在整个Web3 AI赛道融资规模排第5,但之前并没有引起我的注意。 直到Sahara近期AMA请来Claude AI母公司Athropic的联合创始人,我才知道意识到Sahara在行业的地位。 Claude AI是传统AI中ChatGPT最大竞争者,他在写作领域甚至还要高于ChatGPT,而Athropic就是Claude AI的母公司,相当于OpenAI与ChatGPT的关系。 举个通俗的例子,Sahara邀请到Athropic,就相当于一个初创的社交公司第一场讨论会邀请的是字节跳动。 🔸透过现象看本质: 传统AI遇到很大的瓶颈,一个是算力成本,另一个是数据成本,纵使平台有很强的算法能力,也没办法摆脱算力和数据的巨大负担,再加上传统AI变现能力不佳,很多公司开始寻找新的出路。所以Athropic与Sahara AI探索是有原因的。 在Web3做分布式算力的IO、Aethir,做算法的Bittensor,做数据的Ocean 都通过自己的方法解决了AI的三座大山(算力、算法、数据)困境,并且项目方已经找到了变现方法。 2⃣现在AI遇到的瓶颈 上文说过AI有三座大山,也可以说是三大支柱——数据+算力+算法=AI 🔹数据就相当于AI的燃料 🔹算力是AI的引擎 🔹而算法则是AI的灵魂。 巨头都没办法解决以上所有问题,比如OpenAI与Reddit签署了一项价值6000万美元的协议。 还有前段时间Meta为了改善LLM(大语言模型)的多样性,开始收集Facebook、Instagram英国用户的数据,当时因为让用户选择“退出”而不是“同意”,引起轩然大波。 巨头公司尚且在数据和算力上都有这样的困惑,那中小公司怎么办呢?他们不但在算力上遇到很大的瓶颈,在算法和数据上同样有很大的瓶颈。 任何想要在AI赛道中脱颖而出的创业公司,这三要素必须齐全,或者在其中一项中展现出独特的领先优势。 目前Web3 AI的解决方案有: 🔹算力:IO、Aethir、Hyperbolic、Privasea 🔹算法:Bittensor、Sahara AI 🔹数据:Vana、Sahara AI、Ocean、PINAI、PublicAI、Lightworks、DIN、QnA3 .AI 、Aggregata、Arkham 上面说道如Meta、OpenAI由于在数据上遇到很大的瓶颈,这是因为传统AI无时不刻都在爬取用户的数据(文案、图像、音像、购物、旅游、工作……),巨头通过这些数据创造了巨大的收益,但并没有分给用户,并且这里面很多数据都是用户的隐私,用户不愿意授权是正常的,所以出现meta这样霸王的操作。 如果Meta在Sahara AI上分别构建一个Facebook和Instagram的DLP,这不仅确保了数据隐私,还激励更多用户主动贡献数据,会是一种更好的选择。 3⃣Sahara AI有什么特点 AI=算力+算法+数据,任何想要在AI赛道中脱颖而出的创业公司,这三要素必须齐全,或者在其中一项中展现出独特的领先优势。 而Sahara AI在“三大支柱”中满足算法+数据,他具体是做什么呢? 官方说:SaharaAI是一个多层架构的AI区块链应用平台,专注在AI时代建立一个更加公平透明、能够将数据进行价值归属并分润给用户的AI发展模式,解决传统AI系统中的隐私、安全、数据获取以及透明度等痛点。 通俗来讲:SaharaAI是一个AI数据和算法平台,用户任何数据都可以当作“AI资产”(AI Assets),用户可以将这些“资产”上传到Sahara AI的数据市场,Sahara AI通过内部特殊机制来给提供者确权,这些数据可以给需要的AI团队训练AI,用户根据数据的质量来获得奖励。 SaharaAI包含应用、交易、数据和执行四层架构,为AI生态系统发展提供了强大的基础。 🔹应用层:提供安全保险库、去中心化AI数据市场、无代码工具包和SaharaID等工具。 🔹交易层:Sahara区块链,通过权益证明(PoS)共识机制来确保网络的效率和稳定性,让即使在有恶意节点存在的情况下也能达成共识。 🔹数据层:管理链上和链下的数据。 🔹执行层:抽象了保险库、AI模型和AI应用的操作,支持各种AI训练、推理和服务范式。 ✅ 总结 看到SaharaAI 就让我想起:4月掀起小高潮,在币安上线龙头项目Bittensor( $TAO FDV:98.8亿美元),这已经成为AI 产业压舱石的Beta。 简单来讲,Bittensor是由一个个较小的、专门的子网组成的中心网路运作,每个子网都致力于AI 的不同产业。 Sahara AI 作为囊括应用层+数据市场+模型算法的平台型项目,特别符合币安投资标的,Binance Labs在Sahara AI上投入这么多,大概率是将其当作下半场能接棒 $TAO 的新Alpha 龙头。 所以如果你错过了第一轮测试网,记得申请第二轮,不管是作为数据提供者还是验证者,都有激励:hi.saharalabs.ai/get-started



(1/n) Introducing ChainOpera AI (ChainOpera.ai), the L1 Blockchain and Protocol for Co-owning and Co-creating Decentralized AI Agents for Humanity, backed by Federated AI OS and Platform. What's ChainOpera AI? ChainOpera AI offers the blockchain and protocol for co-owning and co-creating decentralized AI apps and agents. 1. The blockchain and protocol empower the AI community to drive innovations in AI agents and apps while co-owning them and restoring data sovereignty to everyday users. It enables a full-lifecycle, scalable, sustainable, and trustworthy AI economy, rewarding all participants—from data and GPU resource suppliers to model developers, AI agent/app creators/developers, and end-users who can monetize their private data to improve community-driven AI. The protocol also features customizable token economics to fit token preferences from agent/app creators/developers, including buyback and burn mechanisms, staking options, and governance with community DAOs. The protocol operates on our federated AI OS and platform. 2. The federated AI OS incorporates Decentralized Identifiers (DIDs) for users to manage their data access, privacy, security, and sovereignty across diverse agents and apps with interoperability and cross-agent/app compatibility. Its innovative AI terminal rewards users for their data contributions, which help build better community-driven large language models (LLMs) with private data. Additionally, it simplifies agent and app development by offering model serving and AI agent workflow, creation, and management through APIs, CLIs, and low-code UIs. 3. The federated AI platform, the world's first decentralized machine learning platform, allows all AI resource contributors (data, model, and GPUs) to participate in the serving of AI agents and apps and get rewarded. Its many years of technological advances in enterprise AI services in TensorOpera, FedML, ScaleLLM, and edge-cloud hybrid model serving make it the only model service provider in the web3 industry that can stably provide cost-effective, scalable, and fast model-serving APIs based on decentralized resources. 4. Community-built Agents. ChainOpera team is also developing a few flagship AI agents and applications to showcase how community AI resources can drive the innovation of AI agents and apps, and promote private data sovereignty within the protocol. The community can co-own and co-create diverse autonomous and socialized AI agents (e.g., AI influencers, game players, AI companions, teachers) that interact with Internet services such as YouTube, TikTok, Twitter, and Discord. Each agent has its own token issuance, and users can even create digital twin-like agents based on their private data. 5. Flagship AI Agent App - AI Terminal. On the AI app side, our flagship ChainOpera AI Mobile App features an AI Terminal with a "Type to Earn" system that benefits humanity by allowing users to share their private data and receive rewards for LLM and GenAI model training. The AI Terminal also functions as a powerful AI agent that simplifies trading of AI coins on the ChainOpera platform and other meme coins. The AI Terminal is your personal companion that represents you and completes various tasks in your life and work. Eventually, the AI Terminal will evolve into your personal AI operating system.



New Incentivized Testnet: OpenLedger 🦑 @OpenledgerHQ, the data blockchain for AI, is offering rewards for users who leverage their unused bandwidth to participate in the incentivized testnet. Run worker nodes, provide data, earn rewards. ➡️ coinlist.co/openledger-tes…












