惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

D
DataBreaches.Net
B
Blog
博客园_首页
C
Check Point Blog
Microsoft Security Blog
Microsoft Security Blog
MyScale Blog
MyScale Blog
P
Proofpoint News Feed
Engineering at Meta
Engineering at Meta
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
B
Blog RSS Feed
M
MIT News - Artificial intelligence
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
WordPress大学
WordPress大学
宝玉的分享
宝玉的分享
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
The Cloudflare Blog
量子位
V
V2EX
Y
Y Combinator Blog
Hugging Face - Blog
Hugging Face - Blog
Martin Fowler
Martin Fowler
Recent Announcements
Recent Announcements
I
InfoQ
博客园 - 【当耐特】

Hacker News - Newest: "AI"

AI can't read an investor deck AI as an attorney? Student uses ChatGPT, Gemini to sue UW over alleged racial discrimination Hacking MCP Servers in AI Systems – The Rug Pull: Tool Changes After Approval GitHub - MeepCastana/KubeezCut: Free Web based video editor Can AI judge journalism? A Thiel-backed startup says yes, even if it risks chilling whistleblowers Coming soon: 10 Things That Matter in AI Right Now DARPA built an AI to fact-check enemy weapons claims What explains heterogeneity in AI adoption? When AI Meets Muscle: Context-Aware Electrical Stimulation Promises a New Way to Guide Human Movements - Department of Computer Science AI Changed How We Build. It Did Not Change What Matters. Linux rules on using AI-generated code - Copilot is OK, but humans must take 'full responsibility for the… Meta spins up AI version of Mark Zuckerberg to engage with employees Code Mode: Let Your AI Write Programs, Not Just Call Tools | TanStack Blog GitHub - Delavalom/graft: Go framework for building AI agents. Type-safe tools, multi-provider (OpenAI, Anthropic, Gemini, Bedrock), zero vendor SDKs. India's TCS tops estimates, says new AI models did not dent services demand Gen Z's fading AI hype Strong feeling: we are in a folded AI reality GitHub - machinarii/total-recall-catalog: A reference catalog of latest knowledge retrieval, memory & RAG systems GitHub - mensfeld/code-on-incus: Give each AI agent its own isolated machine with root, Docker, and systemd. Active defense detects and stops threats automatically.. Quantization, LoRA, and the 8% Problem: Benchmarking Local LLMs for Production AI Iran war: We spoke to the man making Lego-style AI videos that experts say are powerful propaganda Powell, Bessent discussed Anthropic's Mythos AI cyber threat with major U.S. banks GitHub - immartian/bellamem: Persistent belief-graph memory for AI agents. Retrieves decisive context by importance — not recency, not RAG, not /compact. recursive-mode: The Repo-Native Operating System for AI Engineering After the attack on Sam Altman's home, will AI CEO's go on the offensive? The biggest advance in AI since the LLM Opus 4.6 vs GPT 5.4 One Prompt Unity World Generation Test “AI polls” are fake polls Client Challenge Can AI be a 'child of God'? Inside Anthropic's meeting with Christian leaders
"Using Obsidian with AI"
sspaeti · 2026-04-24 · via Hacker News - Newest: "AI"

Everyone is using Obsidian for AI, or wants to use it in order to get more productive. But I think it’s a dead end.

The reasons are clear: Obsidian notes are open on your local disk in an open format, Markdown, and accessible to everyone, including your AI agents. This makes it as great and easy an integration as possible. But should you actually use AI with your notes? And if so, what are the use cases or ways you should interact with it?

  1. First of all, if you use Vibe Code Agents with Obsidian, use the Obsidian CLI. It will be much faster in returning your files, searching, or doing an action than grepping through the files.
  2. I’m a strong proponent of not generating AI summaries or other on-the-fly generated text in my vault, and this is what I try to do as well. The reason is simple: over time, I don’t know anymore if the content was written by me or AI-generated, and my own, much more valuable thoughts get diminished by AI Slop.
    1. Also, when searching for something, you need to fight through the noise of generated stuff. If you only have your own writing, it’s all valuable, or at least there’s a reason why you noted it down. There was a conviction, idea, or something that moved you.
    2. Yes, it’s good in certain areas, and mostly on-the-spot summaries seem great, but every time I come back to them, I find them very average. It doesn’t help me, as it didn’t highlight the things I would, and I need to reread anyway.
    3. What I do sometimes with Obsidian Webclipper, if I create a new note, is summarize it in one sentence or a few. I clearly mark it as AI-generated and even put it into a quote, so it’s clear to me also in 5 years that it wasn’t mine. If I have added my own thoughts, I just remove it.
  3. Use it as search, like finding related notes. I think this might be the best use case. But don’t use it for tagging etc., because again, all of a sudden, all your relations and connections don’t count anything anymore, as they are not done by you. The power lies in the deliberately created graph of notes, your very own Second Brain.
    1. I don’t use it yet. I might when having a local model, as I have lots of sensitive notes that I don’t want to upload anywhere. Plus it takes away the thinking part of my My Obsidian Note-Taking Workflow. I don’t want to replace my human thinking yet. But I use a mathematical or local model with Obsidian Smart Connections, or the Graph Analysis plugin.

I’m pretty sure if you go all in using AI to generate and create your notes, you will end up with less clarity and fewer insights. You might start fresh very soon, as you won’t get value from it, and stop maintaining or adding ideas.

I think if you have an urge to do something, do it in a separate vault, or just use a database, e.g. DuckDB, and do all the fun stuff with AI. But don’t mix your precious notes and Zettelkasten for it. Not yet, at least, but probably never.


Origin: Li by mehdio, Obsidian
References: Writing is Thinking!