How I researched ChatGPT in a week.

I'll take ChatGPT as an example, and talk about how I quickly cover a track in a week, just for reference.

1) When encountering a new track or concept, the first step is always to figure out the actual definition, i.e. what is ChatGPT, so my first step is to read the wiki of ChatGPT, which has the core information such as definition, information, history, etc. After I figured out the definition, I searched Google (Baidu is of low significance and can be ignored).

2）After the definition is figured out, it is to search Google, Google on the one hand is to see the latest news trends, and on the other hand is to directional search investment institutions, famous blogger's analysis of the article, such as A16Z, Sequoia America and so on (you can also search the founder of the most typical company in this field and the back of the investment institution's statement, for example, in this case it is Sam Altman).

This kind of hot topic these institutions must have a lot of analysis articles, blogs, podcasts and other content, and is filtered, processed after the cognitive output of the summary, probably see 10 - 20 on the almost, to this step in fact, up to a day's time, can have more than the majority of people's knowledge of the country, most of the information you can see in the country is the translation of the information within the scope of the excerpts.

(3) then next step is to search facebook, twitter, substack, will find a lot of related topics of the article, which will encounter a lot of English articles above the translation of the article, but also occasionally some of the better focus on the summary.

In this step, it's actually a good test to see what quality people you've added, or whether you have some active groups, if so, you can easily find good articles in your circle of friends, or see some sharing reports in the groups.

The above three is what we call secondary research, that is, you can not have to communicate with the outside world, purely by yourself to spend time to see what you can learn, these things are good, cognition will already be more than 80% of the people on the market (because the vast majority of people will not really sit down and spend the time to seriously search and read).

After this, we should already have enough understanding of a certain concept, track, formed their own perceptions and assumptions, and understand the market probably what typical companies are doing, so then mainly after the process of verifying assumptions, and people to communicate with the process, the purpose is to obtain non-market information, perception, improve their own judgments, as well as more important to us, to find whether there is a landing, and to participate in it! the opportunity, i.e., what are the good asset targets.

At this time, we must ask ourselves to summarize the simple cognition, as well as to form core judgments and assumptions, and have a list of our own questions, so that we can better carry out the next step, primary research, i.e., to communicate with others.

At this point in time, if you find that you don't have questions or good questions worth asking, it means that your understanding is not deep enough, or you need to train and force yourself to think for a long period of time, and most of these questions come from your previous understanding and accumulation of a certain field (correspondence and comparison is the best way to do this, for example, go to compare the development of the mobile Internet era), and the current understanding of the core crux and framework of the field.

Next, take these questions and assumptions and contact professionals in the industry to talk to them again, to fill in and correct what is now known. I'm not going to expand on this because of the word limit, but this is my research journey, and I hope it can be helpful to all of you!