🪧 AI Tools: Less Debate, More Action Every time a new model is released, it immediately divides into two factions: 1. Adventist faction Damn it, the modeling ability is too strong. Hang all the lords 2. The opposition The level is too low, I can't bear to look directly at it. It's incomparable to XX I understand, they are all for the sake of traffic, without any advantages/disadvantages, they still need to be artificially created, otherwise there will be no debate; Without debate, where does the heat and traffic come from. The question is, how many rounds of field testing did this model/product undergo before reaching a definitive conclusion The new model is amazing&the new product is terrible Less debate, more action. A model is a tool, there is no right or wrong distinction, but there may be a distinction between superiority and inferiority, or a suitable audience and scenario. It's not appropriate to compare Gemini 2.5 at 9.9/month with OpenAI Deep Research at 200. As soon as the Deep Research based on Gemini 2.5 Pro driver came out today, I conducted a test: 1. Retrieve English materials and gather as many quotes as possible from Charlie Munger 2. Perform secondary processing on the generated document, add Chinese and English subheadings, and generate a JSON format document Overall, it looks quite good. Although I failed the first test because I asked it to generate a JSON format document directly, I finally worked hard and gave me a blank JSON document. But this did not affect my ability to successfully split the workflow on the second attempt. Report to report, process to process. That's good. Subsequently, I added the generated JSON document to my "One Word" website so that I could listen and absorb the wisdom of looking up old words and phrases in my spare time. For example, in the screenshot: Long term wisdom You don't need to be talented, you just need to be slightly smarter than the average person for a long time. ——Charlie Munger | One Word