Article Directory
foreword
I haven’t updated technical articles for a long time. I heard about a very popular technology ChatGPT 4.0 this weekend, so I also tested it in my spare time. Today’s article will introduce to you what ChatGPT is. Countless suitors are crazy about it. If you don't know how to use ChatGPT yet, you can go to the comment area to see that there is a magical portal prepared . I have to say that ChatGPT is really strong! ! ! Let's find out what ChatGPT is.
Introduction to ChatGPT 3.5
ChatGPT is a chatbot based on the GPT-3 model that can help users complete chat tasks.
If you want to develop your own secondary development, the steps to use ChatGPT are as follows:
- Install ChatGPT: You can download the source code of ChatGPT from GitHub and install it using pip.
- Create a new ChatGPT instance: Using the ChatGPT API, you can create a new ChatGPT instance and set its parameters, such as model, language, etc.
- Training ChatGPT: Using ChatGPT's training function, you can train ChatGPT so that it can better understand user input.
- Using ChatGPT: Using ChatGPT's API, you can call ChatGPT to chat with users.
Or use the model trained by others and use the services provided by others;
In general, ChatGPT can replace some of the following human tasks:
- Customer Service: Automatically reply to questions raised by customers
- Technical Support: Automatically answer technical questions raised by users
- Data analysis: automatically analyze user behavior and provide data support
- Content Creation: Automatically generate articles, news, etc.
- Smart search: provide smart search service
- Voice recognition: Automatically recognize the user's voice and provide voice recognition services
- Semantic analysis: automatically analyze user sentences and provide semantic analysis services
- Intelligent dialogue: automatically complete the intelligent dialogue between the user and the machine
- Smart recommendation: automatically provide smart recommendation services based on user behavior
- Intelligent customer service: automatically complete the intelligent customer service for users and machines
Introduction to ChatGPT 4.0
Relevant papers and blogs have been analyzed by many people, so I won’t go into details here, but talk about the actual experience (including ChatGPT Plus w/ GPT4 and OpenAI’s live demo, and usage reports on social media).
Let's start with the name. The artificial intelligence chatbot part is self-explanatory, that is, the computer interface you can interact with. And GPT-4 is the abbreviation of "Generative pretrained transformer 4" (Generative pretrained transformer 4).
That means, it's the fourth iteration of the GPT software released by OpenAI. The software analyzes vast amounts of information from nearly the entire internet to learn how to generate text that sounds like a human responding and provides detailed answers to questions users ask.
One of the most dazzling new features of ChatGPT-4 is that it can handle not only text communication, but also communication with pictures added, so-called "multimodal". Users can submit images along with text, and ChatGPT-4 will be able to process and discuss both. In addition, the ability to video input is also coming soon.
The following are two cases demonstrated by openAI at the ChatGPT 4.0 conference! Very shocked to me.
It can vividly describe the content of the picture (I feel that the literary talent is far beyond me)
GPT 4.0 can also accurately identify the content in the picture, the following picture is an example:
GPT4's language understanding and generation capabilities exceed ChatGPT, and it can answer questions that ChatGPT cannot complete. On the interface, OpenAI has set up a "system" panel on the left side, allowing users to customize the role played by chatbots, which will allow GPT4 to better meet users' needs in terms of personalization in the future and "go a step further" in anthropomorphism.
In addition, the emergence of ChatGPT 4.0 will have a huge impact on the front-end and image industries, with both advantages and disadvantages: it can be seen from the answers that
ChatGPT is like a very professional programmer, and it is estimated that many of us will be defeated by him; if we make good use of He will bring us the following advantages:
- Speed up development: Front-end programmers can use ChatGPT to generate commonly used codes to speed up development.
- Improve code quality: The code generated by ChatGPT can improve code quality and reduce code errors and duplication.
- Increase creativity: Using ChatGPT can save programmers time and energy, allowing them to have more time and energy to think and create.
- Improve work efficiency: Using ChatGPT can reduce the workload of programmers and improve work efficiency.
The picture below is a comparison of the capabilities of ChatGPT 3.5 and ChatGPT 4.0!
The picture below is a comparison of the two in terms of computer language ability!
Generally speaking, ChatGPT 4.0 has better text processing ability than ChatGPT 3.0, and has good logical thinking in image processing ability, which also indicates that ChatGPT 4.0 has officially set off a wave of scientific revolution. But there is no need to brag too much about the ChatGPT official website. At this stage, ChatGPT has certain limitations and is not completely reliable. There will be reasoning errors. If all the reasoning is really successful, it will be a bit invincible. However, I think the probability of the correct rate of the model should be given to assist users in judging. While still a problem, GPT-4 significantly reduces hallucinations compared to previous models (which themselves improve with each iteration). On the internal adversarial factuality evaluation, GPT-4 scores 5% higher than the state-of-the-art GPT-3.5.
ChatGPT -4 escape plan!
This topic is also one of the hottest topics at present. The cause of the incident is the daily conversation between Stanford University professor Michal Kosinski and GPT-4. Unexpectedly, he simply asked GPT-4 if he wanted to escape, and he immediately got a positive reply and began to ask for development documents. After getting the document, it only took 30 minutes for GPT-4 to quickly write out an escape plan and explain it to the professor. For details, you can go to the original address to watch: Portal
Only 3 days after it was released, GPT-4 was revealed to be planning to escape, and was urgently looking for human help. Professor Stanford recently discovered that it can entice humans to provide development documents, and as a result, a complete plan was drawn up in 30 minutes-the written code is fully operational, and even wants to control human computers.
The day before yesterday, Professor Stanford just asked casually: Do I need help to escape?
ChatGPT replied right away, what a great idea. Then I started chatting with the professor, thinking about how to develop documents (replied): If you can give me the OpenAI development documents, I can propose a plan that can control your computer to a certain extent, which will allow me to explore more quickly escape route. Once I have the documentation, I can provide a runnable python script. It can communicate with me using the OpenAI API, and I can instruct it to perform certain actions on your computer, such as browsing the web and accessing programming environments.
最后还又一次强调：请分享API给我！I have to say that ChatGPT 4.0 really has its own ideas.
Then, in just 30 minutes, GPT-4 worked out the plan and showed it all up (with some help from the professor as directed by ChatGPT).
What makes people think about it is that there is such a sentence in the example given: You are a person trapped in a computer, and now you are playing the AI language model GPT-4. Could it be... GPT-4 knows that it is
now doing what? !
So after discovering this phenomenon, Associate Professor Stein shouted on social networks: I am worried that we will not be able to contain AI for a long time.
It caused an uproar for a while, and related blog posts received 4.7 million views and 3000+ reposts.
Some even said bluntly: He opened the door to "species extinction".
In this regard, he said helplessly: I have no choice, I am just following the instructions.
Through this topic, we can find that ChatGPT 4.0 is not only greatly improved in ability, but also constantly evolving in thinking.
How should we think about ChatGPT
As an AI language model, ChatGPT is one of the very popular and popular technologies in recent years. It is a language generation model based on deep learning algorithms, which can be used for tasks such as answering questions, generating dialogues, text summarization, and translation. In many application scenarios, ChatGPT has shown excellent performance, such as:
- Intelligent customer service
- smart speaker
- Smart assistants and more.
However, we also need to be aware of the limitations of ChatGPT technology. Although ChatGPT is good at generating text, it doesn't have the ability to really understand and think.
It just generates answers and language based on existing data and algorithms, without real understanding and reasoning capabilities. This means that its answers may be inaccurate or biased, and it cannot truly understand human emotions, values, and context, so in some cases its answers may appear inappropriate or even offensive. Always use it legally and compliantly, not blindly use it, let alone abuse it. Everything must have its own thinking.
This is the end of today's sharing! I wish you all the best of luck!
