ChatGPT - What do we have for Future?
How would ChatGPT perform in the long run? And what we can expect.
Introduction
Ah, ChatGPT is the latest craze that has taken the internet by storm. It's a technology that promises to revolutionize the way we interact online. And no, we are not talking about communicating with fellow humans, but with computers.
What we are talking about is an AI Chatbot - called ChatGPT.
But what is so special about ChatGPT? After all, what could be so revolutionary about a chatbot that can generate text based on a given prompt?
Let's take a closer look and see what ChatGPT is all about.
What is so special about ChatGPT?
Let's see a small snippet of ChatGPT. Shall we?
Woah, that seems pretty self-explanatory. If you are not new to chatbots and it's working, this may just be a fun result to see. After all, there are many chatbots available, and many of them give the same 'quirky and fun' answer.
However, below are some awesome results that took the internet by storm, and made ChatGPT a household name in the Tech and AI community.
How does ChatGPT work?
Well, before we dive in, I'll tell you what ChatGPT is. It's Chat Generative Pre-trained Transformer.
What does that mean?
A Generative Pre-Trained Transformer (GPT), or in the case of ChatGPT - GPT-3, is a language model that relies on deep learning to generate human-like texts based on a given text-based input. In this AI model, a user “feeds” a sentence, and the transformer (aka the main component of this deep learning model) creates paragraph-based information extracted from publicly available datasets.
GPT-3 (Generative Pre-trained Transformer 3) is a language model that was created by OpenAI. The company is known for its cutting-edge artificial intelligence research laboratory in San Francisco. GPT-3 contains around "175-billion" parameters and can produce human-like text. The model was trained on large text datasets with hundreds of billions of words.
That's all great and impressive. However, shortly after its release on November 2021, many people speculated about its future.
And that's what we will be looking at next
What is the supposed future of ChatGPT?
This is the fun part.
Take a look at the following news articles:-
and
ChatGPT is a fascinating technology, but it certainly has its disadvantages.
Many people from the creative industry - like that content writing, fear that this tool would soon replace their job. You can now instantly create a full article with just by few simple commands that you give to ChatGPT. Many content creators are already using this tool and integrating it with their workflow.
If this is good or not, efficient or inefficient, depends on the users.
I asked Matt Thomas, a professional technical writer/copyrighted on LinkedIn about his views on this matter. "The creative possibilities it offers to the writer" was his favourite feature of ChatGPT.
Interestingly, when I asked him about his opinion on the disadvantage of such AI chatbots - he said the following.
"I think it depends on whether you’re using it to generate ideas or just churn out mediocre content for you ".
And he is right. It's not surprising to see content writers directly using the generated text in their writing, and sharing their work as their own on social media. While this is not wrong per se, it does affect the credibility and quality of the content.
...there is more to it
Apart from the human factor, we also have some technical limitations.
For starters, it's NOT the most reliable technology. Yes, I said it. It can often give inaccurate or inappropriate responses.
Here's one ChatGPT instance that I reacted to on my LinkedIn.
And there are more -
With this, it is apparent that ChatGPT still has a long way to go. And the fact that an AI is unable to do calculations is quite an unmerited instance.
Technical geeks would argue that the fault is on our side. The reason being us giving bad or unclear prompts. While this may be right to some degree, an AI built for "human" interactions must also be capable to take a "humanly" style prompt as well.
Apart from this, it takes huge maintenance costs to keep such an up and running.
" I think the usage of conversational LLMs like ChatGPT is here to stay though the tech giants would be burning a lot of money to keep it running in the short run", says Abhishek Mishra, a technical lead working at intel. The cost of maintaining a chatbot like ChatGPT depends on the complexity of the tasks it is expected to perform.
For instance, a basic chatbot that can respond to customer inquiries on a company website may require minimal maintenance costs, such as hosting fees for the server and occasional software updates. This cost increases as the demand for computation and traffic increases.
On the other hand, more advanced chatbots (like ChatGPT) that are used for more complex tasks, such as natural language processing, may require more substantial investments. In addition to the hosting fees, businesses may need to invest in powerful hardware, such as graphics processing units (GPUs) and specialized software. They may also need to hire a team of developers to design, build, and maintain the chatbot.
Recently it is estimated that $ 1 billion was received by OpenAI from Microsoft. Many other companies are also investing in this firm. ChatGPT owner OpenAI predicts that they will be able to generate a revenue of $1 billion by the end of the year 2024.
Currently, OpenAI has around 120 employees. Hence the workforce is still relatively low when compared to the increasing demand.
Ultimately, the cost of maintaining a chatbot like ChatGPT will depend on the tasks it is expected to perform and the resources available to the business.
As and when the demand increase, and traffic skyrockets, the organization has to put huge capital into scaling its Ai chatbot.
As I am writing this, ChatGPT is still in the testing phase (ironically, many people don't know that). However, as this tool scales, it could start charging money from users to can't out heavy processing
That's why Mishra posted this on his LinkedIn a few days back -
Conclusion
It's just recently that ChatGPT was released and it has a long way to go. Many companies see new opportunities in this evolving tech.
However, only time will answer whether this is a success or failure. Personally, though, I'm pretty optimistic and excited about how this technology will evolve.
Let’s see what the coming days have to prove.
See you on the next post.