Last night, almost out of curiosity, I installed Claude for Excel . I wasn’t trying to run a demo or prove a point. I simply had a real task at hand:to compare the financial performance of Hong Kong’s UGC-funded universities —revenue, expenditure, surplus, and the trend across multiple academic years. Normally, this would mean: jumping between annual reports, copying figures from PDFs and news articles, cleaning inconsistent tables, and manually building comparisons in Excel.

When internet first started, one of the grueling tasks was to design web page. While the advent of chatbot LLM has allowed us to write simple programming code to display simple web page, there lacks aesthetics. Google has just changed that by the announcement of the Gemini 3 model this week. Wow. In the latest announcement, which is featured here in this paper: https://generativeui.github.io/ (Please go click and try some of the demo.), we can see that just by typing in s

Dear Friends, By now, most of us have used a public AI chatbot. But for many business leaders, a key question remains: "How can I trust it with my  data? How do I stop it from 'hallucinating' and giving me answers that sound good but are factually wrong?" This is the central problem we must solve. We are all looking for a way to move AI from a public "toy" to a private, reliable "tool." Quick Summary I've been prototyping a "Hybrid AI" model that I believe is the solution. It

In our last post, we explored the powerful user interfaces of LLMs and encouraged you to start playing with them. But what if you could...