Frontier Daily (Aug 26): 18 items from GitHub (OpenAI), Anthropic, Hacker News and more + X supplements

Every day Laojin pulls the official channels and compiles the past 48 hours of AI news into a short brief: curation only, no commentary, with the original link and publish time attached to every item.

1. openai/openai-dotnet: The official .NET library for the OpenAI API

OpenAI has released the official .NET library for the OpenAI API, written in C#.

Source: GitHub (OpenAI) · 2026-08-25 · Original

2. Funding better evaluations of AI’s impact on wellbeing

Anthropic is launching a $5 million grant program to fund independent research into AI's impact on users' wellbeing.

Source: Anthropic News（官网监测） · 2026-08-25 · Original

3. Introducing the Anthropic Economic Index

Anthropic announced the new Anthropic Economic Index with data on AI use in occupations.

Source: Anthropic News（官网监测） · 2026-08-25 · Original

4. How much of HN is AI?

A post analyzing how much of Hacker News content is AI-related, with 179 upvotes and 148 comments.

Source: Hacker News · 2026-08-25 · Original

5. AI is hitting entry-level jobs hardest, Stanford study finds

A Stanford study finds AI is hitting entry-level jobs hardest, with 76 upvotes and 65 comments on Hacker News.

Source: Hacker News · 2026-08-25 · Original

6. The full stack behind abundant intelligence

OpenAI CFO Sarah Friar explains how advances across chips, compute, models, and products compound to deliver more useful intelligence at scale and lower cost.

Source: OpenAI Blog · 2026-08-25 · Original

7. Jalapeño’s first results show industry-leading speed and efficiency in AI inference

OpenAI's custom inference chip Jalapeño delivers faster, more power-efficient AI inference with higher throughput and lower latency.

Source: OpenAI Blog · 2026-08-25 · Original

8. [AINews] Andrew Ng gets into AI Engineering

Andrew Ng relaunches DeepLearning.ai with a focus on AI engineering, based on analysis of over 10,000 job postings and interviews.

Source: Latent Space · 2026-08-25 · Original

9. llm-anthropic 0.27

llm-anthropic 0.27 adds compatibility with the anthropic v1.0.0 Python library, which switched from httpx to httpx2.

Source: Simon Willison's Weblog · 2026-08-24 · Original

10. Quoting Drew Breunig

Drew Breunig reflects on how Fable's high cost led to thinking about what work goes where, marking the end of the free lunch.

Source: Simon Willison's Weblog · 2026-08-23 · Original

11. OpenAI Cuts GPT-5.6 Sol API Pricing by Over 20%

OpenAI is dropping API and credit pricing for GPT-5.6 Sol by over 20% for the next 3 months while introducing new credits for eligible ChatGPT Work and Codex plans.

Source: X (Grok x_search, @OpenAI) · within 48h · Original

12. DeepSeek Launches Multimodal V4-Flash-Vision-Exp on API

DeepSeek releases experimental multimodal V4-Flash-Vision-Exp matching text performance of V4-Flash and nearing Opus-4.8 on agent benchmarks, plus new Files API and Harness support.

Source: X (Grok x_search, @deepseek_ai) · within 48h · Original

13. Mistral AI Partners with HUMAIN for Regional AI Development

Mistral AI announces strategic collaboration with HUMAIN to develop localized frontier models focused on cybersecurity, voice, and Arabic performance across Saudi Arabia and the region.

Source: X (Grok x_search, @MistralAI) · within 48h · Original

14. Anthropic Demonstrates Claude's Breakthrough in Protein Binder Design

Anthropic shows Claude autonomously designing successful protein binders at 22-35% success rate versus 10-15% industry norm, with open-sourced prompts and data to advance drug development.

Source: X (Grok x_search, @AnthropicAI) · within 48h · Original

15. Meta AI Previews WildArtifactBench for Multimodal Agents

AI at Meta introduces WildArtifactBench, a new evaluation framework using human and agentic preference judges for real-world multimodal agent tasks, releasing 10 tasks to the community.

Source: X (Grok x_search, @AIatMeta) · within 48h · Original

16. Qwen3.8-27B Tops Code Arena in Its Size Class

Alibaba's Qwen3.8-27B achieves #9 overall on Code Arena WebDev as the only model in its size in the top 10, reshaping the Pareto frontier and continuing strong adoption in research papers.

Source: X (Grok x_search, @Alibaba_Qwen) · within 48h · Original

17. Google DeepMind Advances Continual Learning in Persistent Game Worlds

Google DeepMind partners with game creators to tackle continual learning, deep memory, long-horizon planning, and multi-agent dynamics in living universes for new gameplay and real-world applications.

Source: X (Grok x_search, @GoogleDeepMind) · within 48h · Original

18. xAI Increases Included Grok Usage in Cursor

xAI raises the included usage limits for Grok models in Cursor permanently as demand surges following the Grok 4.6 launch, building on prior expansions.

Source: X (Grok x_search, @leerob) · within 48h · Original

---

_Sources: official blogs of OpenAI, Google DeepMind and Qwen; official RSS of Simon Willison, Latent Space, Import AI and One Useful Thing; Anthropic site monitoring, top AI discussions on Hacker News and open-source updates from key AI orgs on GitHub. X sources are retrieved live via Grok x_search with per-link verification. This is a source digest — no commentary from Laojin._