🤖 AI 资讯日报 — 2026-06-25

自动生成于 2026-06-25 08:33 | 数据源: OpenAI, Anthropic, HuggingFace, GitHub, Apple, NVIDIA, xAI, Simon Willison, THE DECODER, ITHOME, HackerNews, ArXiv, Twitter KOL

🔥 T1 官方一手源

OpenAI

• Blog
• OpenAI Developer Blog. Insights for developers building with OpenAI. Mastering Codex Remote for engineering. Jun 23. Mastering Codex Remote for engineering.

• Mastering Codex Remote for engineering
• Turn your phone into a Codex control center for starting, steering, reviewing, and organizing engineering work.

• Testing Agent Skills Systematically with Evals
• A practical guide to turning agent skills into something you can test, score, and improve over time.

• What makes a great ChatGPT app
• At DevDay we introduced ChatGPT Apps — a new way to bring your product directly into ChatGPT conversations. This post builds on that launch

• Codex blog posts
• Trigger published ChatGPT workspace agents · Commerce. Build commerce flows in ChatGPT · Ads. Publish and measure ads in ChatGPT.

Anthropic

• Agentic coding and persistent returns to expertise
• This report provides evidence on how Claude Code is used in practice, based on a privacy-preserving analysis of around 400000 interactive

• Project Fetch: Phase two
• We report results from our latest test of whether Claude can help Anthropic employees perform sophisticated robotics tasks.

• Research
• Research. Our research teams investigate the safety, inner workings, and societal impacts of AI models—so that artificial intelligence has a positive impact as

• Labor market impacts of AI: A new measure and early evidence
• Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.

• How AI assistance impacts the formation of coding skills
• Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.

HuggingFace

• Beyond LoRA: Can you beat the most popular fine-tuning technique?
• We're on a journey to advance and democratize artificial intelligence through open source and open science.

• PP-OCRv6 on Hugging Face: 50-Language OCR from 1.5M to 34.5M Parameters
• PP-OCRv6 is the latest generation of PaddleOCR's universal OCR model family. It is designed for real-world text detection and recognition

• Shipping huggingface_hub every week with AI, open tools, and a human in the loop
• huggingface_hub is the Python client at the base of the Hugging Face ecosystem. transformers , datasets , diffusers , sentence-transformers

• Introducing North Mini Code: Cohere's First Model For Developers
• A Blog post by Cohere Labs on Hugging Face.

• Experimenting with the proposed Cross-Origin Storage API in Transformers.js
• We're on a journey to advance and democratize artificial intelligence through open source and open science.

GitHub

• I automated my job (and it made me a better leader)
• Explore how my day as a senior leader looks now that I use 40 automations to help, and learn more about some of my favorites.

• New features and Claude as agent provider preview in JetBrains IDEs
• This update adds support for organization and enterprise agents from GitHub, lets you queue and steer messages in Copilot CLI sessions,

• AI credits consumed per user now in the Copilot usage metrics API
• The Copilot usage metrics API now reports how many AI credits each user consumed per day, derived from the same AI credits consumption data

• How we built an internal data analytics agent
• Learn how GitHub built Qubot, our internal Copilot-powered analytics agent, to allow any GitHub employee to ask questions about our data in

• GitHub joins coalition advocating for fixes to California AI Transparency Act to protect open source
• We're calling for targeted amendments to resolve conflicts with open source licensing and align with international transparency frameworks.

xAI

• Introducing /goal
• Use /goal for long-running autonomous task execution in Grok Build.

• Grok on Databricks
• We're excited to share that Grok models are now natively available on Databricks Agent Bricks, Databricks' developer agent platform.

• Grok Imagine Video 1.5
• Grok Imagine Video 1.5 is now generally available on the Imagine API. We've also rolled out Video 1.5 Fast on grok.com/imagine and our iOS

• Grok for PowerPoint
• Grok now works inside Microsoft PowerPoint — turn outlines into slides, expand the deck, and tighten the narrative without leaving the app.

• Use Grok in Warp
• You can now use your Grok or X Premium subscription inside Warp, an agentic development environment built on the terminal and used by almost

Simon Willison

• Porting the Moebius 0.2B image inpainting model to run in the browser with Claude Code
• Porting the Moebius 0.2B image inpainting model to run in the browser with Claude Code. 22nd June 2026. This morning on Hacker News I saw

• Datasette Apps: Host custom HTML applications inside Datasette
• 18th June 2026. Today we launched a new plugin for Datasette, datasette-apps, with this launch announcement post on the Datasette project

• Prompt Injection as Role Confusion
• I wish every paper would come with one of these. Academic writing is pretty dry - the impact of a paper can be so much higher if you publish

• Simon Willison's Weblog
• Unless LLMs achieve genuine role perception, we think injection defense will remain a perpetual whack-a-mole game. And the continuous nature of role boundaries

• GLM-5.2 is probably the most powerful text-only open weights LLM
• 17th June 2026. Chinese AI lab Z.ai released GLM-5.2 to their coding plan subscribers on June 13th, and then yesterday (June 16th) released

📰 T1.5 媒体源

THE DECODER

• AI is inflating student grades, and the effect points to outsourced work, not better learning
• A UC Berkeley study of more than 500000 grades found that courses heavy on writing and coding saw grades jump after ChatGPT launched.

• Microsoft researcher builds a working neural network out of goats in Age of Empires II to critique AI science
• Adrian de Wynter, a researcher at Microsoft and the University of York, has built a working neural network inside the map editor of the

• Artificial Intelligence: News, Business, Science | THE DECODER
• Artificial Intelligence is changing the world. THE DECODER brings you all the news about AI.

• Algorithms from the 1980s power today's AI breakthroughs, earn Turing Award for researchers
• Andrew Barto and Richard Sutton have won the 2024 AM Turing Award for developing key technologies that power modern artificial intelligence.

• AI assistants fail basic fact-checking in BBC news study
• A systematic evaluation of leading AI chatbots reveals widespread problems with accuracy and reliability when handling news content.

ITHOME

• NAIRR Science Program Reshapes Scientific Research, Powered by NVIDIA AI Infrastructure
• 美国国家科学基金会的NAIRR 试点项目在NVIDIA AI 基础设施支持下，推动了多个领域超过700 个创新科研项目，本文介绍了部分代表性项目。

• NVIDIA Vera Rubin Delivers World-Class Supercomputers for Science
• IT之家 6 月22 日消息，. NVIDIA Vera Rubin Delivers World-Class Supercomputers for Science. With 7 Exaflops of AI for Science and 5 Petaflops

• NVIDIA Announces BioNeMo Agent Toolkit — Tools for Agents to Accelerate Scientific Discovery
• IT之家 6 月23 日消息，. NVIDIA Announces BioNeMo Agent Toolkit — Tools for Agents to Accelerate Scientific Discovery. News Summary:.

• 苹果 iOS / iPadOS 26.2 正式版发布：增强音乐、播客和游戏体验
• 苹果今日向iPhone 和iPad 用户推送了iOS/iPadOS 26.2 更新（内部版本号：23C55），本次更新距离上次发布正式版间隔39 天。

• 《仙剑奇侠传七》Xbox 版现已上线 Win10 / Win11 微软商店，预计 11 月 2 日发售
• 仙剑奇侠传七登陆Xbox# 《#仙剑奇侠传七#》Xbox版现已上线微软商店，游戏预计2022 年11 月2 日发售，具体定价尚未公布。

HackerNews

• OpenAI unveils its first custom chip, built by Broadcom

• Qualcomm to Acquire Modular

• RubyLLM: A Ruby framework for all major AI providers

• Computer use in Gemini 3.5 Flash

• GLM-5.2 is a step change for open agents

💬 T2 社区 & 学术

Twitter KOL

• Ethan Mollick: AI task horizons graph with GPT-2 image generator
• My most popular AI post was a bunch of made-up 'graphs' four years ago. Now, the new GPT-2 image generator does it for real (though not perfect). Here's the famous AI task horizons graph with a touch ...

• Ethan Mollick: Frontier models (GPT-5.5 Pro Extended, Claude 5 Fable Max) fail translation test
• This is an interesting test, and the frontier models (GPT-5.5 Pro Extended, Claude 5 Fable Max) do fail. They refuse to turn the 'three words' into 'four' if that fits better.

• Ethan Mollick: History of AI in 32 images - improvements in AI models and open weights
• I wrote a history of AI in 32 images of otters using wifi on airplanes, from images to video to code. It shows two big trends: rapid improvements in AI models of all types and the growth of open weigh...

• Ethan Mollick: Using AI hurts learning if it undermines mental effort
• More evidence, from a large-scale study in China, that using AI hurts learning if it undermines mental effort. When homework time drops due to AI use, so do test scores.

• Ethan Mollick: Big Three AI Labs likely uncatchable while rapid improvement continues
• At least until (if?) rapid improvement stops, it seems less likely someone is going to catch the Big Three AI Labs. Microsoft and Meta released their models, which were fine, but not frontier.

• Ethan Mollick: OpenAI profitable on serving customers with 40%+ gross margins
• If the leaked financial data is right, OpenAI is profitable on serving customers with 40%+ gross margins. But training remains incredibly expensive.

• swyx @aiDotEngineer profile
• swyx - AI engineer, creator of Latent Space podcast, associated with Cognition and Temporalio

• Berryxia.AI: UX/Product designer prompt for mobile and web interfaces
• You're a friendly UX/Product designer that's an expert in mobile and web UX design. Design one or multiple mobile or web interfaces.

HF Daily Papers

• MEMPROBE: Probing Long-Term Agent Memory via Hidden User-State Recovery

• VeriEvol: Scaling Multimodal Mathematical Reasoning via Verifiable Evol-Instruct

• Critique of Agent Model

• InSight: Self-Guided Skill Acquisition via Steerable VLAs

• Multi4D: High-Fidelity Dynamic Gaussian Splatting via Multi-Level Competitive Allocation

• Semantic Browsing: Controllable Diversity for Image Generation

• AGORA: An Archive-Grounded Benchmark for Agentic Workplace Document Reasoning

• ChartWalker: Benchmarking the Cross-Chart RAG Task

ArXiv

• [[2506.09985] V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning](https://arxiv.org/abs/2506.09985)

• [[2312.10997] Retrieval-Augmented Generation for Large Language Models: A Survey](https://arxiv.org/abs/2312.10997)

• [[1706.03762] Attention Is All You Need](https://arxiv.org/abs/1706.03762)

• [[2005.14165] Language Models are Few-Shot Learners](https://arxiv.org/abs/2005.14165)

• Artificial Intelligence in Creative Industries: Advances Prior to 2025

• From Mind to Machine: The Rise of Manus AI as a Fully Autonomous Digital Agent

• HybridQuestion: Human-AI Collaboration for Identifying High-Impact Research Questions

• A Comprehensive Review of AI Agents: Transforming Possibilities in Technology and Beyond

本日报由 Hermes Agent 自动生成 | 2026-06-25