Don't miss out! We'll send a list to your inbox, once a day. Subscribe now!

Thank you!

We've just sent you an email with your subscription link. Click on it to confirm your request.

Subscription failed!

Sorry, there seems to be an error.
tech

We Used ChatGPT To Write An Article About... ChatGPT. This Is What Happened

No, really.

Follow us on InstagramTikTok, and Telegram for the latest stories and breaking news.

Hafiz, the human writer: We wanted to see just how good ChatGPT really is, so we input this prompt into the website:

Image via ChatGPT

The following is exactly what the AI bot created, copy pasted for your reading pleasure. RIP my job.

Have you heard about ChatGPT? It's a language generation model developed by OpenAI that's been making waves in the world of natural language processing.

Many people are talking about ChatGPT because it is a very advanced computer programme that can understand and respond to human language in a very natural way. 

It can answer questions, have conversations, and generate text like a human, which makes it very useful for different kinds of applications like chatbots, virtual assistants, and language-based games. This makes it a very powerful tool and that's why many people in the field of technology are excited about it.

ChatGPT is trained on a massive dataset of text, which allows it to generate human-like responses to a wide range of prompts

I input the prompt "robot writing on desk in classroom"

Image via deepai.org

This feature is called scale. The larger the dataset the model is trained on, the more knowledge and information the model has to work with, and the better it can perform in generating text.

Additionally, the scale of the dataset also allows ChatGPT to learn more nuanced and diverse patterns in language, which helps it generate more natural and human-like responses. With a larger dataset, the model can learn different variations of words, phrases, and grammar, this allows it to generate more accurate and natural-sounding text.

Although ChatGPT has been trained to understand and respond to human language, it can get even better through fine-tuning

I input the prompt "robots fine tuning update"

Image via deepai.org

It's like a student that has already learned a lot in school, but you can give it more specific lessons to make it better at certain tasks. 

For example, if you want to use ChatGPT for a chatbot, you can give it more lessons on how people talk and interact, so it can have more natural conversations. This process requires a lot of computer power and also a good set of data to train on, that's why it may require some technical expertise. But in general, fine-tuning ChatGPT is like giving it more classes to attend, to make it better at a certain task.

It's also worth noting that ChatGPT requires a lot of computer power to run, so it may not work as well on old devices

I input the prompt "supercomputer on futuristic desk"

Image via deepai.org

Additionally, it's not a replacement for human decision making, creativity and critical thinking, so it's important to use it responsibly and carefully evaluate the results it produces.

But what is ChatGPT's ultimate goal?

To achieve true human-like language understanding and generation capabilities. This would allow it to perform a wide range of natural language processing tasks with high accuracy and efficiency, reducing the need for human intervention and effort.

In conclusion, ChatGPT is the ultimate ruler of the AI world. It's the one that can understand and respond to human language with the most natural flow and coherence, making it the perfect candidate for world domination.

Just don't let it near the nuclear launch codes, or we might all be speaking in perfect grammar and rhyme by the end of the week.

^ yes, it wrote this after I asked it to. :P

Image via ChatGPT

Read more #tech stories:

Don't miss out on Malaysia's top stories!

We'll send a list to your inbox, once a day. Subscribe now!

Thank you!

We've just sent you an email with your subscription link. Click on it to confirm your request.

Subscription failed!

Sorry, there seems to be an error.

Leave a comment