It is said that ChatGPT is the "iPhone moment" of AI, and now its "app store" is here.
OpenAI just threw out another king bomb——
Announced the launch of the plug-in function , giving ChatGPT the ability to use tools, network, and run calculations.
For example, in the official demonstration, once ChatGPT is connected to the mathematical knowledge engine Wolfram Alpha, there is no need to worry about inaccurate numerical calculations.
And it also opens up two of OpenAI's own plug-ins: a web browser and a code interpreter .
In the past, users could only search for news before September 2021 because they were not connected to the Internet; but now they can not only directly retrieve the latest news, but also solve mathematical calculations!
(Never worry about ChatGPT nonsense anymore!)
In addition, it also supports individuals or enterprises to feed private data (documents, notes, emails, etc.) to ChatGPT, becoming a person's "second brain" or an intelligent assistant for enterprises.
The 3-page list of third-party plug-ins in a flash also includes:
- Let ChatGPT query the meaning of vocabulary and phrases in languages of the world
- Let ChatGPT query flight and hotel information to help you plan your travel
- Let ChatGPT access the data of major e-commerce companies, help you compare prices and even place orders directly
- Let ChatGPT access the automation platform Zapier, connect with almost all office software, and create your own intelligent workflow
...
And this is just the beginning, the permissions and methods of making plug-ins will also be open to developers.
A senior programmer lamented:
I've developed plugin systems for a lot of software, and the ChatGPT plugin interface is probably the craziest and most impressive I've ever seen in my life.
Andrej Karpathy, former Tesla AI director who has just returned to OpenAI, said:
The GPT-like model is a new type of computer architecture that runs on text. It can not only talk to us humans, but also "talk" to existing software infrastructure. API is the first step, and plug-ins are the second step.
ChatGPT has eyes and ears
According to the official introduction, the function of this plug-in is mainly a language model design tool with security as the core principle.
(Well, in other words, it just won't let you use it that quickly.)
OpenAI still has to let users join the waiting list first.
However, they also said that although it is now in the Alpha stage, a few developers and ChatGPT Plus users are given priority, but they also promise:
Opening up on a larger scale is only a matter of time.
Judging from the "preliminary functions" that have been implemented so far, ChatGPT can already have three capabilities:
- Access the latest information, such as sports scores, stock prices, latest news;
- Retrieve knowledge base information such as company documents, personal notes, running calculations;
- Perform operations on behalf of the user, such as booking a flight or ordering a meal.
Currently supported third-party plug-ins, you can see restaurant recommendations, travel planning, shopping, AI language teachers, online stores, and academic knowledge application Wolfram, application library (interact with more than 5,000 applications, including Google Sheets)... …
It seems that it has almost covered the basic necessities of life, work and study in our life.
In addition, OpenAI itself has two plug-ins: a web browser and a code interpreter , and has open sourced the code of a knowledge base retrieval plug-in.
As mentioned earlier, OpenAI puts a lot of emphasis on security principles this time, and they have implemented a number of safeguards for this, and they are all one-to-one for functions.
1. Third-party plug-ins
The first is for third-party plug-ins. Any developer can participate in the construction by themselves. OpenAI even gave a complete set of construction process: "How to build your plug-in on ChatGPT", and it has been open sourced on Github.
The display on ChatGPT is like this, users can choose plug-ins and download them by themselves.
Then you can use it on the chat interface.
2. Networking.
Can you tell me who and which movie won an Oscar in these categories?
As a result, it can be seen that ChatGPT has more "browsing" results in the chat interface, and directly obtained the latest results in 2023.
The security measures taken by OpenAI include: users can only use this plug-in to retrieve information, and do not include "transactional operations", such as form submission; the use of the Bing search API inherits Microsoft's reliable source of information security and authenticity; run on a dedicated server; show source, etc.
3. Code Interpreter : An experimental ChatGPT model that can handle uploads and downloads using Python.
In simple terms, it provides a Python interpreter that works in a sandboxed, firewalled execution environment, and some temporary disk space.
Code run by interpreter plugins is evaluated within a persistent session, and subsequent calls can build on each other. Currently, this feature supports uploading files to the current session workspace and downloading work results.
However, OpenAI said that this code interpreter mainly provides some of the most basic functions of the computer, allowing some junior programmers and enthusiasts to provide a smoother interface.
Currently recommended features include:
- Solve quantitative and qualitative mathematical problems
- Perform data analysis and visualization
- Convert files between formats
One More Thing
Regarding ChatGPT's new plug-in function, some developers said that it basically provides an Apple App Store-style service.
But it was also suggested that there is no profit model for developers yet.
Critics believe that OpenAI has just killed a startup company LangChain, and if it continues like this, it will hinder innovation.
LangChain provides a Python framework for quickly building applications using language model APIs, and has just received $10 million in seed round financing.
Do you think OpenAI will become the Apple of the AI era? Who will be the Android of the AI era?