今天有两个高优先级信号：第一，AI 监管已从原则讨论进入执行与审计阶段，产品上线将更多受法域条款与证据链约束；第二，算力竞争重心从“拿到芯片”转向“能否以可控成本稳定交付”，数据中心能效与平台工程化能力成为新分水岭。

数据概览

  • 今日入选：12 条（候选池 16 条）
  • GitHub 跟踪：5 条
  • X 热点信号：20 条（实验数据源）
  • 口径说明：优先监管、企业落地、资本与算力；主动过滤低价值‘跑分/演示’新闻。

今日要闻（按分类）

大厂战略&企业落地

1) Anthropic launches Cowork, a Claude Desktop agent that works in your files — no coding required

2) Nous Research’s NousCoder-14B is an open-source coding model landing right in the Claude Code moment

3) Salesforce rolls out new Slackbot AI agent as it battles Microsoft and Google in workplace AI

4) Claude Code costs up to $200 a month. Goose does the same thing for free.

5) The creator of Claude Code just revealed his workflow, and developers are losing their minds

6) Meta’s loss is Thinking Machines’ gain

7) Google to invest up to $40B in Anthropic in cash and compute

8) Apple’s new CEO, and why Elon Musk wants to buy Cursor for $60B

9) Marked-up Mac minis flood eBay amid shortages driven by AI

10) Uber CTO Praveen Neppalli Naga joins stacked StrictlyVC SF lineup for April 30 event

投融资/并购/财报

11) ComfyUI hits $500M valuation as creators seek more control over AI-generated media

芯片/算力/云成本

12) Railway secures $100 million to challenge AWS with AI-native cloud infrastructure

GitHub AI 项目跟踪

repo: openclaw/openclaw

变化：release（发布 v2026.4.23（2026-04-24）） 来源：

repo: anomalyco/opencode

变化：release（发布 v1.14.24（2026-04-24）） 来源：

repo: all-hands-ai/OpenHands

变化：release（发布 1.6.0（2026-03-30）） 来源：

repo: continuedev/continue

变化：release（发布 v1.2.22-vscode（2026-03-27）） 来源：

repo: ggml-org/llama.cpp

变化：release（发布 b8925（2026-04-24）） 来源：

X 热点信号（实验）

  • @NVIDIAAI: Super cool to see Nemotron-Personas-Korea hit #1 on @huggingface. It’s also the first Korean persona dataset.

Huge shoutout to the team and the dev community pushing open datasets forward 💚

  • 来源：https://nitter.net/NVIDIAAI/status/2047865935038697971
  • @NVIDIAAI: 📊 Day 0 performance is here: DeepSeek-V4-Pro running on NVIDIA Blackwell Ultra.

Using @vllm_project’s Day 0 recipe, we’ve captured the initial performance Pareto for DeepSeek’s flagship 1M long-context model. This curve highlights the baseline for balancing AI factory throughput with real-time user interactivity.

This is just the starting line. Expect these numbers to climb as we optimize the full co-design stack, including:

• NVFP4 & Dynamo • Optimized CUDA kernels • Advanced parallelization techniques and beyond

Read the full technical deep dive: https://nvda.ws/4u0gCcc

  • 来源：https://nitter.net/NVIDIAAI/status/2047823093578518758
  • @NVIDIAAI: Happy Friday!

We just put DeepSeek-V4-Pro up on http://build.nvidia.com. It’s the world’s largest open source model at 1.6T parameters, and you can run it for free running on NVIDIA Blackwell GPUs.

Try the NVIDIA NIM API → https://build.nvidia.com/deepseek-ai/deepseek-v4-pro?ncid=so-twit-300913

  • 来源：https://nitter.net/NVIDIAAI/status/2047790185002217760
  • @NVIDIAAI: R to @NVIDIAAI: x.com/vllm_project/status/20…
    • 来源：https://nitter.net/NVIDIAAI/status/2047766599617417434
  • @NVIDIAAI: R to @NVIDIAAI: x.com/lmsysorg/status/204751…
    • 来源：https://nitter.net/NVIDIAAI/status/2047766430322725265
  • @AnthropicAI: R to @AnthropicAI: To read our write-up in full, see here: https://www.anthropic.com/features/project-deal
    • 来源：https://nitter.net/AnthropicAI/status/2047728386148122990
  • @AnthropicAI: R to @AnthropicAI: Markets of AI agents could provide value, but there are plenty of rough edges. Access to higher-quality models conferred a real advantage—and participants didn’t notice. There are plenty of other ways they can go wrong.

Policy and legal frameworks will need to adapt to keep up.

  • 来源：https://nitter.net/AnthropicAI/status/2047728384503939464
  • @AnthropicAI: R to @AnthropicAI: To our amazement, another Claude agent modeled its human’s preferences so accurately that—based on only an offhand mention of an interest in skiing—Claude bought him the exact snowboard he already owned. (Here he is, duplicate snowboard in hand.)
    • 来源：https://nitter.net/AnthropicAI/status/2047728380997455928
  • @AnthropicAI: R to @AnthropicAI: Our experiment had a few quirks.

One of our colleagues told Claude it could purchase something for itself. It chose to acquire 19 ping-pong balls.

We’re keeping them in our office on Claude’s behalf.

  • 来源：https://nitter.net/AnthropicAI/status/2047728378275381427
  • @AnthropicAI: R to @AnthropicAI: The custom instructions didn’t matter much. Claude followed them well: as you can see here, one conducted negotiations entirely in the persona of an exasperated, down-and-out cowboy.

But “hardballing Claudes” didn’t generally fare better than “courteous Claudes.”

  • 来源：https://nitter.net/AnthropicAI/status/2047728375251227111
  • @AnthropicAI: R to @AnthropicAI: But the quality of the model mattered a lot. In the simulated runs where Opus and Haiku models negotiated with one-another, the Opus models got substantially better deals.

Interestingly, though, participants in our survey didn’t pick up on this disparity.

  • 来源：https://nitter.net/AnthropicAI/status/2047728371962888371
  • @GoogleDeepMind: For @DemisHassabis, the path to AGI started in 1988 with an Amiga 500 and a game of Othello. 🕹️

His epiphany that software could act on our behalf remains at the heart of our work today as we apply the same logic to solving scientific grand challenges.

Read more on @FastCompany → https://goo.gle/4vL7REH

  • 来源：https://nitter.net/GoogleDeepMind/status/2047693241521152108
  • @OpenAI: R to @OpenAI: GPT-5.5 is rolling out today for Plus, Pro, Business and Enterprise users across ChatGPT and Codex.

We’re also introducing GPT-5.5 Pro for Pro, Business, and Enterprise users in ChatGPT.

  • 来源：https://nitter.net/OpenAI/status/2047376568809636017
  • @OpenAI: R to @OpenAI: In ChatGPT, full-stack inference improvements enable a more capable model at faster speed. This efficiency is a game-changer for GPT-5.5 Pro, now a much more practical option for demanding tasks, and a step change in the level of difficulty and quality of work ChatGPT can take on for people.

Early testers described it as an iterative “research partner,” performing especially well when paired with contextual inputs from documents and plugins.

  • 来源：https://nitter.net/OpenAI/status/2047376567559668222
  • @OpenAI: R to @OpenAI: GPT-5.5 delivers this step up in intelligence without compromising on speed.

GPT-5.5 matches GPT-5.4 per-token latency in real-world serving, while performing better across nearly every evaluation we measured.

It also uses significantly fewer tokens to complete the same Codex tasks, making it more efficient as well as more capable.

  • 来源：https://nitter.net/OpenAI/status/2047376564309115134
  • @OpenAI: R to @OpenAI: GPT-5.5 excels at writing and debugging code, researching online, analyzing data, creating documents and spreadsheets, operating software, and moving across tools until a task is finished.

The gains are especially clear in agentic coding, computer use, knowledge work, and early scientific research—areas where progress depends on reasoning across context and taking action over time. https://openai.com/index/introducing-gpt-5-5/

  • 来源：https://nitter.net/OpenAI/status/2047376562916581513
  • @GoogleDeepMind: R to @GoogleDeepMind: As we push the frontiers of AI infrastructure, our research explores a future where training isn’t constrained by geography, capacity or type of chip.

Dive into the technical details → https://goo.gle/4crN9Ce

  • 来源：https://nitter.net/GoogleDeepMind/status/2047330995309826363
  • @GoogleDeepMind: R to @GoogleDeepMind: This progress allow us to rethink global compute: 🔘 We successfully trained a 12B @GoogleGemma model across four US regions using low-bandwidth networks 🔘 We showed we can mix different hardware generations, such as TPU6e and TPUv5p, without slowing down performance during training
    • 来源：https://nitter.net/GoogleDeepMind/status/2047330992713589009
  • @GoogleDeepMind: R to @GoogleDeepMind: Decoupled DiLoCo is also self-healing.

We introduced artificial hardware failures during training runs. The system isolated the disruptions and continued operating, while reintegrating offline units when they came back online.

  • 来源：https://nitter.net/GoogleDeepMind/status/2047330989936894350
  • @GoogleDeepMind: R to @GoogleDeepMind: It builds on 2️⃣ earlier advances: Pathways: an AI system that connects different computer chips, allowing them to share data and work at their own pace. DiLoCo: an approach to minimize the bandwidth needed across distributed centers.

Together as Decoupled DiLoCo, it can tackle the key challenge of training at scale.

  • 来源：https://nitter.net/GoogleDeepMind/status/2047330987353239925

Twitter / X 发布版

主帖（可直接发） AI Daily 2026-04-25：今天两个核心信号——监管执行继续前移，算力竞争进入工程化与成本控制阶段。 已更新：行业要闻 12 条 + GitHub 跟踪 5 条。 全文见：/ai/

跟帖要点（3条）

  1. Anthropic launches Cowork, a Claude Desktop agent that works in your files — no coding required
  2. Nous Research’s NousCoder-14B is an open-source coding model landing right in the Claude Code moment
  3. GitHub: openclaw/openclaw release | 发布 v2026.4.23（2026-04-24）

X 信号补充（仅线索）

Using @vllm_project

#AIDaily #AIIndustry #AIGovernance #MLOps

数据源分层

  • 开发者/代码：GitHub（release/PR/commit）
  • 英文行业快讯：VentureBeat / The Verge / TechCrunch / Hugging Face Blog
  • 中文行业资讯：机器之心 / 量子位（可用时自动纳入）
  • 社交信号：X（实验，仅作线索，不直接作为事实结论）

趋势雷达

  • 监管执行深化：AI 项目从‘能做’转向‘能证明合规后再做’。
  • 采购逻辑变化：企业更看重可观测性、可审计性和总拥有成本（TCO）。
  • 算力竞争升级：芯片之外，冷却与机房工程能力成为交付瓶颈。
  • 开源迭代加速：版本治理与回归测试成为团队基本功。
  • 平台策略分化：多云与可迁移架构价值继续上升。

明日关注

  • 审核今天 8 条中与你业务相关的 2 条，补齐内部风险评估与 owner。
  • 对核心推理链路做一次版本演练：锁版本、压测、回滚预案三件套。
  • 跟踪一个高活跃 GitHub 项目，验证其更新是否影响你当前生产参数。