[GPTs] What is the significance of the emergence of GPTs, and what is their positioning intended to do?

The biggest role of GPTs is that they allow us to create our own personal bots, making it easy to upload knowledge bases and train exclusive content-output machines. You can share the trained content with users or your team. An excellent GPTs must have its own unique knowledge base or an efficient ability to understand and process logic. Based on this, it is not hard to identify its positioning: it is a specific tool for a unique scenario.

1. Making simple things even simpler: Although ChatGPT itself is a natural-language conversation process, there is still a certain barrier to using it at this stage. To get ChatGPT to help you complete a complex task, you still need to understand the characteristics of GPT models yourself and be good at breaking down the problems you encounter, so as to obtain satisfactory answers step by step. Therefore, it requires users to have relatively good logical ability, questioning ability, and prompting skills. After GPTs were launched, these skills and requirements were further lowered. You only need to use GPTs packaged and created by others in specific scenarios.

2. Democratization of technology, allowing those who know how to use it to benefit: For people with good logical ability and an understanding of AI models, they can indeed program in natural language and are no longer limited by a lack of development ability. They can build applications for their own use to improve efficiency; they can also package their understanding of the industry, AI, and user needs into GPTs to earn revenue.

Essentially, I think the following types of people can benefit from the GPTs trend:

1) People with deep expertise in a specific industry (other than AI): They can leverage their understanding of their industry/profession and their accumulated data and knowledge to create better GPTs applications. For example, it is easy to use ChatGPT to help you write Xiaohongshu content, and you can even easily create a GPTs that can produce Xiaohongshu titles/notes. But only when you truly understand Xiaohongshu, know the process for generating Xiaohongshu notes, and know what makes a good title and good content can you design truly useful GPTs.

2) People with insight into user needs: These users can be consumer (C-end) users or business (B-end) merchants. You always have to know what problems they are currently facing and what needs they have before you can create something that meets their needs. So in fact, traditional product managers—people who have a methodology and know how to gain insight into needs—can actually benefit from this trend; now is truly the era when everyone is a product manager.

3) People with strong logical ability and rich practical AI experience: In fact, current models still have many characteristics when producing output, such as “it does not want to succeed,” unstable output, and generally poor results for complex tasks. Only people who truly understand the mechanisms behind these problems and can handle them well can write good GPTs.