ChatGPT — In-Depth FAQ’s

This is a live document and will be updated as and when OpenAI makes improvements to ChatGPT.

Follow me to get live updates on this article.

Is ChatGPT going to be free forever?

ChatGPT is free to use during the initial research preview. Given this official wording from OpenAI, I do not think ChatGPT will be free forever.

How long will ChatGPT be available for free?

Based on my research online, ChatGPT is costing OpenAI approximately $3.0 million per day. It seems likely that this service will not remain free for an extended period of time.

Will ChatGPT be available as a subscription service?

OpenAI has not yet announced a specific plan for monetizing ChatGPT, but they have indicated that they will begin charging for the service in the future. It is possible that they will offer a subscription service with various pricing tiers that offer different amounts of prompts and response size limits to accommodate different user’s needs.

How much would ChatGPT subscription service cost?

The server costs for ChatGPT, which currently serves approximately 3 million users per day, are estimated to be around 3 million dollars. This works out to be approximately $90 million per month. In order to break even on these costs, OpenAI would need to charge a subscription fee of $30 per month. However, this does not take into account other expenses such as development, research, and other business costs. If the number of users remains at 3 million active monthly users, it would be fair for OpenAI to charge $50 per month for the service. However, if the number of users were to increase to 20 million, the subscription price could potentially be reduced to a more reasonable $20 per month.

ChatGPT’s response is stopping in the middle without completing. How should I request ChatGPT to provide me the complete response?

To continue a previous answer that was cut off, the best prompt would be to restate the question or topic that you were discussing, and then provide any additional context or details that may be relevant. This will help me understand the context and pick up where I left off in my previous response.

For example, if the previous answer was cut off while discussing the process for baking a cake, you could prompt me to continue by saying something like: “Can you please continue explaining the steps for baking a cake? I remember you mentioned mixing the ingredients, but what comes next?”

Alternatively, you can simply ask me to continue the previous answer by saying something like: “Can you please continue the previous answer?” This will allow me to review the previous conversation and provide a response that builds on my previous remarks.

Is ChatGPT good enough to replace software engineers’ jobs yet?

There are several reasons why artificial intelligence systems like ChatGPT are not yet advanced enough to fully replace the role of software engineers.

First, these systems are trained on information that is already available on the internet, which means that they may not have access to the most up-to-date information. In order to overcome this limitation, the systems would need to have access to a separate, updatable memory of world knowledge, rather than relying on information that was available at the time of training.

Second, software engineers have a broad skillset that includes not just writing code, but also designing, building, testing, and deploying software systems. These systems are complex, and it is unclear whether ChatGPT or similar systems are capable of handling this level of complexity.

Finally, ChatGPT is a dialogue system, which means that it is biased by the context of previous conversations. This context bias, combined with the statistical nature of the system, can lead to different solutions being output for the same problem, some of which may be incorrect.

While it is possible that artificial intelligence systems like ChatGPT may be able to assist software engineers and increase their productivity, they are not yet advanced enough to fully replace the role of software engineers. It is also possible that these systems may pose a threat to the demand for junior programmers whose main duties consist of writing code to implement certain functionalities.

How does ChatGPT work?

GPT-3 is a language model developed by OpenAI that uses machine learning techniques to generate text. It is trained on a large dataset of text and is able to predict the next word in a sequence based on the context of the words that come before it. This allows it to generate coherent and seemingly human-like text.

ChatGPT is a variation of GPT-3 that has been specifically designed for dialogue. It uses a method called Reinforcement Learning with Human Feedback (RLHF) to improve its ability to engage in conversation. RLHF involves providing the model with human feedback on its output in order to guide it towards desired behavior. For example, if ChatGPT generates a response that is inappropriate or not helpful in a conversation, a human can provide feedback indicating that the response was not appropriate, and ChatGPT will use this feedback to learn how to generate more appropriate responses in the future.

Why does the AI seem so realistic and lifelike?

These models were trained on large amounts of data from the internet written by humans, including conversations, so their responses may sound human-like. It is important to keep in mind that this is a direct result of the model’s design (i.e. maximizing the similarity between outputs and the data the models were trained on) and that these outputs may be inaccurate, untruthful, and potentially misleading at times.

Can I trust that the AI is telling me the truth?

ChatGPT is not connected to the internet and may occasionally provide incorrect answers. It has limited knowledge of events after 2021 and may also occasionally give harmful instructions or biased content. OpenAI recommends verifying the accuracy of responses from the model. If you find an answer to be incorrect, please provide feedback by using the “Thumbs Down” button.

Who can view my conversations?

As part of our commitment to responsible AI, OpenAI reviews conversations to improve our systems and ensure that the content complies with our policies and safety requirements.

Will OpenAI use my conversations for training?

Yes, your conversations may be reviewed by OpenAI’s trainers to improve their systems.

Can OpenAI delete my data?

Yes, please follow the data deletion process at this link: https://help.openai.com/en/articles/6378407-how-can-i-delete-my-account

Can you delete specific prompts?

No, you are not able to delete specific prompts from your history. Please do not share any sensitive information in your conversations.

Can I see my history of threads? How can I save a conversation I’ve had?

Currently, it is not possible to view your conversation history, but OpenAI is looking into this feature.