← 首页
24AUG
AI前沿每日脉动
AI Frontier Pulse · 中英双语版 Bilingual Edition
2026.08.24 · 周一刊
10 位 Builder17 条推文0 期播客1 篇博客
ChatGPT WorkApplied AIOpen WeightsTokens per WattTrust & Access
Richard Liu · 2026
Curated by Richard Liu · Snapshot follow-builders-2026-08-24-v2
今日头条 · Model And Harness Co-design01 / 15
今天最强信号：GPT-5.6 Sol 只用两项运行设置变化，就在 ARC-AGI-3 达到 SoTA。模型能力不再能脱离 harness、compaction 与跨 context 执行单独评估。The strongest signal: two runtime setting changes put GPT-5.6 Sol at the top of ARC-AGI-3. Model capability can no longer be separated from harness design, compaction, and cross-context execution.
@thsottiauxX 原文7.4K ❤ · 316 RT · 1.1K 💬
Thibault Sottiaux：Reset has been propagated to accounts and we landed some fBuilder 动态 · Source Excerpt
Thibault Sottiaux 原文要点：Reset has been propagated to accounts and we landed some fixes to usage for things mentioned yesterday as issues we found.
Source-grounded excerpt from Thibault Sottiaux: Reset has been propagated to accounts and we landed some fixes to usage for things mentioned yesterday as issues we found.
@thsottiauxX 原文7.4K ❤ · 316 RT · 1.1K 💬
Thibault Sottiaux：2026 is the year companies start seriously caring about moBuilder 动态 · Source Excerpt
Thibault Sottiaux 原文要点：2026 is the year companies start seriously caring about model efficiency and reliability as it becomes critical infrastructure
Source-grounded excerpt from Thibault Sottiaux: 2026 is the year companies start seriously caring about model efficiency and reliability as it becomes critical infrastructure
@thsottiauxX 原文5.4K ❤ · 155 RT · 441 💬
02 / 15
行业分析 · Science, Evals, And Credibility02 / 15
Peter Steinberger：cli is nice, having UI visualizations and your team where Builder 动态 · Source Excerpt
Peter Steinberger 原文要点：cli is nice, having UI visualizations and your team where you work is nicer.
Source-grounded excerpt from Peter Steinberger: cli is nice, having UI visualizations and your team where you work is nicer.
@steipeteX 原文1.1K ❤ · 44 RT · 89 💬
Ryo Lu：leaving leaving is the process from fullness into emptinesBuilder 动态 · Source Excerpt
Ryo Lu 原文要点：leaving leaving is the process from fullness into emptiness from the familiar circle of life the routines undone the streets i knew by heart the people i kept meeting the names…
Source-grounded excerpt from Ryo Lu: leaving leaving is the process from fullness into emptiness from the familiar circle of life the routines undone the streets i knew by heart the people i kept meeting the names…
@ryolu_X 原文808 ❤ · 31 RT · 38 💬
03 / 15
产品设计 · Experience Becomes A Force Multiplier03 / 15
AI 降低执行成本，却没有消除品味与经验：资深创始人的判断继续复利，设计工具也必须把审美、易用性和乐趣一起产品化。AI lowers execution cost without eliminating taste or experience. Seasoned judgment compounds, while design tools must productize aesthetics, ease, and fun together.
@rauchgX 原文703 ❤ · 33 RT · 50 💬
Guillermo Rauch：Our philosophy on extending 𝚏𝚡: open protocols.技能化工作流 · Source Excerpt
Guillermo Rauch 原文要点：Our philosophy on extending 𝚏𝚡: open protocols.
Source-grounded excerpt from Guillermo Rauch: Our philosophy on extending 𝚏𝚡: open protocols.
@rauchgX 原文703 ❤ · 33 RT · 50 💬
Garry Tan：Prediction: systems of record will need to become AI harneBuilder 动态 · Source Excerpt
Garry Tan 原文要点：Prediction: systems of record will need to become AI harnesses or face replacement by agents
Source-grounded excerpt from Garry Tan: Prediction: systems of record will need to become AI harnesses or face replacement by agents
@garrytanX 原文567 ❤ · 29 RT · 101 💬
04 / 15
设计系统 · Harness, Design.md, Routing, Mobile04 / 15
Guillermo Rauch：Intelligence is getting cheaper.模型网关 · Source Excerpt
Guillermo Rauch 原文要点：Intelligence is getting cheaper.
Source-grounded excerpt from Guillermo Rauch: Intelligence is getting cheaper.
@rauchgX 原文325 ❤ · 17 RT · 33 💬
Peter Steinberger：I added the rotation USB protocol to and told my claw to lBuilder 动态 · Source Excerpt
Peter Steinberger 原文要点：I added the rotation USB protocol to and told my claw to look around- now it's having fun with my 360 webcam.
Source-grounded excerpt from Peter Steinberger: I added the rotation USB protocol to and told my claw to look around- now it's having fun with my 360 webcam.
@steipeteX 原文208 ❤ · 11 RT · 20 💬
Nikunj Kothari：Mother in law is visiting from India..Builder 动态 · Source Excerpt
Nikunj Kothari 原文要点：Mother in law is visiting from India..
Source-grounded excerpt from Nikunj Kothari: Mother in law is visiting from India..
@nikunjX 原文170 ❤ · 0 RT · 12 💬
Madhu Guru：How to build great evals - part 7.Builder 动态 · Source Excerpt
Madhu Guru 原文要点：How to build great evals - part 7.
Source-grounded excerpt from Madhu Guru: How to build great evals - part 7.
@realmadhuguruX 原文110 ❤ · 8 RT · 3 💬
05 / 15
深度博客 · Claude For Apple Foundation Models05 / 15
一个用户体验可以由多种模型共同完成：设备端负责快速本地任务，Claude 接手需要更深推理与工具调用的步骤。One user experience can use the right model for each step: local models for fast tasks, Claude for deeper reasoning and tools.
Claude Blog原文链接
Claude 通过 Swift package 接入 Apple Foundation ModelsClaude Joins Apple's Foundation Models Framework
Claude Blog 发布新的 Swift package，让 Apple 开发者在同一个 Foundation Models 工作流里组合设备端模型与 Claude。设备端模型处理快速、本地的摘要或抽取；复杂的多步推理、代码生成、网页搜索与数据分析可交给 Claude，并将 typed Swift values、streaming 和 tool calls 返回同一个 SwiftUI 界面。
Claude Blog introduces a Swift package that combines Apple's on-device models with Claude in one Foundation Models workflow. Local models handle fast private tasks, while Claude takes multi-step reasoning, code generation, web search, and data analysis, returning typed values and streamed tool results to SwiftUI.
Claude Blog原文链接Jun 08, 2026
06 / 15
播客深度 · Compute Meets The Physical World06 / 15
PODCAST DEEP DIVE
AI 的下一轮竞争不只发生在模型层，也发生在冷却、电网、数据中心、供应链与每瓦 token 数构成的物理系统里。The next AI race is not only at the model layer, but across cooling, grids, data centers, supply chains, and tokens per watt.
07 / 15
播客理念 · Industrial Compute At Scale07 / 15
算力上线即被消耗New Compute Is Consumed Immediately
新增容量上线即被工作负载吸收，过早放慢建设通常会带来负面意外。
New capacity is consumed immediately, and slowing the build too early repeatedly creates negative surprises.
关键指标：每瓦 token 数The Metric Is Tokens Per Watt
定制芯片与模型工作负载协同设计，目标是在电力约束下最大化每瓦可生成的 token。
Custom silicon and model workloads are co-designed to maximize tokens generated per watt.
数据中心必须新增电力Data Centers Must Add Power
大规模 AI 基础设施需要同步投资发电、输电、变电站与配套容量。
Large-scale AI infrastructure requires new generation, transmission, substations, and supporting capacity.
算力策略是 All of the AboveThe Strategy Is All Of The Above
多种云、芯片、数据中心与供应路径并行推进，减少单点约束。
Clouds, chips, data centers, and supply paths advance in parallel to reduce single-point constraints.
08 / 15
快讯速览 · Builder Signals08 / 15
Guillermo Rauch：Intelligence is getting cheaper.模型网关 · Source Excerpt
Guillermo Rauch 原文要点：Intelligence is getting cheaper.
Source-grounded excerpt from Guillermo Rauch: Intelligence is getting cheaper.
@rauchgX 原文325 ❤
Peter Steinberger：I added the rotation USB protocol to and told my claw to lBuilder 动态 · Source Excerpt
Peter Steinberger 原文要点：I added the rotation USB protocol to and told my claw to look around- now it's having fun with my 360 webcam.
Source-grounded excerpt from Peter Steinberger: I added the rotation USB protocol to and told my claw to look around- now it's having fun with my 360 webcam.
@steipeteX 原文208 ❤
Nikunj Kothari：Mother in law is visiting from India..Builder 动态 · Source Excerpt
Nikunj Kothari 原文要点：Mother in law is visiting from India..
Source-grounded excerpt from Nikunj Kothari: Mother in law is visiting from India..
@nikunjX 原文170 ❤
Madhu Guru：How to build great evals - part 7.Builder 动态 · Source Excerpt
Madhu Guru 原文要点：How to build great evals - part 7.
Source-grounded excerpt from Madhu Guru: How to build great evals - part 7.
@realmadhuguruX 原文110 ❤
Peter Yang：There are two types of AI evals - tops down and bottoms upBuilder 动态 · Source Excerpt
Peter Yang 原文要点：There are two types of AI evals - tops down and bottoms up.
Source-grounded excerpt from Peter Yang: There are two types of AI evals - tops down and bottoms up.
@petergyangX 原文43 ❤
Peter Yang：I've been working with Char, my (human!) assistant from @o技能化工作流 · Source Excerpt
Peter Yang 原文要点：I've been working with Char, my (human!) assistant from @oceanstalent for 6 months.
Source-grounded excerpt from Peter Yang: I've been working with Char, my (human!) assistant from @oceanstalent for 6 months.
@petergyangX 原文21 ❤
09 / 15
数据洞察 · Snapshot09 / 15
今日数据概览Today Stats
收录 Builder：10 位
总推文数：17 条
播客节目：0 期
博客文章：1 篇

最高互动：Thibault Sottiaux：Reset has been propagated to accounts and we landed some f · 7.4K ❤
第二高互动：Thibault Sottiaux：2026 is the year companies start seriously caring about mo · 5.4K ❤
The snapshot includes 10 builders, 17 tweets, 0 podcast, and 1 blog post.
follow-buildersSnapshot follow-builders-2026-08-24-v2
5 条关键洞察5 Key Takeaways
ChatGPT Work 正从任务助手变成可调度上下文、应用与协作的新计算入口。
ChatGPT Work is becoming a computing surface that orchestrates context, apps, and collaboration.
模型越强，企业越需要连接数据、监管、人类决策点与现实反馈回路的 applied AI 层。
Stronger models increase demand for an applied-AI layer spanning data, regulation, human decisions, and real-world feedback.
开放权重从开发者偏好升级为产业领导力与生态扩散战略。
Open weights are moving from developer preference to an industrial strategy for leadership and ecosystem diffusion.
算力竞争的关键约束转向物理世界：电力、冷却、建设速度与每瓦 token 数。
Compute competition is increasingly constrained by power, cooling, build speed, and tokens per watt.
跨应用执行放大了价值，也把 Gmail、日历与办公数据的信任问题推到产品核心。
Cross-app execution expands value while making trust around Gmail, calendars, and office data a core product issue.
10 / 15
本周之声 VOICE 01
只要允许模型持续推理，并借助 canonical compaction 跨多个 context window 工作。
“You just have to allow it to reason and work over multiple context windows.”
@thsottiauxX 原文7.4K ❤ · 316 RT · 1.1K 💬
11 / 15
本周之声 VOICE 02
最好的路径，是赋能科学家，而不是试图由我们自己解决所有问题。
“The best way to do this is for us to empower scientists.”
@steipeteX 原文1.1K ❤ · 44 RT · 89 💬
12 / 15
本周之声 VOICE 03
我总会先创建 design.md，写清颜色、字体以及其他设计准则。
“I always create a design md file with guidelines for color, typography, and more.”
@steipeteX 原文208 ❤ · 11 RT · 20 💬
13 / 15
播客之声 PODCAST VOICE
AI 是一场“50 年的一夜成功”。
“AI's fifty year overnight success.”
14 / 15
AI前沿每日脉动
AI Frontier Pulse · 2026.08.24
本期收录 10 位 Builder · 17 条推文 · 0 期播客 · 1 篇博客
ChatGPT Work · Applied AI · Open Weights · Tokens per Watt · Trust & Access
感谢阅读 · Thank You For Reading
Richard Liu · AI前沿每日脉动 · 2026 · Snapshot follow-builders-2026-08-24-v2