AI

Lately, there is a need of private chatbot service as a complete alternative to OpenAI’s ChatGPT. So, I decide to implement one at home and make it accessible to everyone in my household alongside with my network printer and NAS (OpenMediaVault). In the past, I used to recommend people using Llama series for English tasks and Qwen series for Chinese tasks. There was no open-source model that’s strong enough in multilingual tasks comparing to proprietary ones (GPT/Claude)....

As a lifelong self-learner, I use all sorts of methods to learn new things, and AI is what I’m currently into. Although I’ve been in the game since 2022, my background wasn’t focused on AI. So, like everyone else, I had to do some “AI For Dummies” level study in order to get more involved. Below is a list of learning materials that I find very helpful for myself to get started with and might also be helpful for someone else in the same situation....

Recently, I encountered a scrolling problem on ChatGPT web like this. I don’t use any userscript or extensions for OpenAI’s products and this problem occors in my unhardened Chrome based browser which I do not want to switch from. Based on the workaround in the threads, I made an userscript out of it which solves the problem by running in tampermonkey. (function () { document.querySelectorAll('html *').forEach(function(node) { var s = getComputedStyle(node); if (s['overflow'] === 'hidden') { node....

the “Open Source AI Manifesto” Mark Zuckerberg has been hated by Richard Stallman for decades and he is the only person who appears on both cover image of the video essays: How the Internet was Stolen and How AI was Stolen by Then & Now. However, things has changed recently after his adoption of the Fediverse and open-sourcing of Llama continuously. In Zuckerberg’s latest “Open Source AI Manifesto”, he stated 5 needs of open-source:...

AI-Generated content can be fun or “slop” according to Simon Willison, but also can be malevolent due to its abuse in phishing attacks. Some of my readers may have already known that recently I’m working on a side project , which based onchatgpt-html and uses LLMs to detect phishing emails. I think at some point, the tool should be able to detect phishing attempt from video content too. Because deepfake technology is so accessible nowadays and its generated content can be quite convincing....