Responsive Advertisement

What is ChatGPT? Benefits and Limitations - Learning Hub


We have seen in movies the rise of artificial intelligence and the robots fighting with human beings and claiming to destroy the human race and take over the world. Well I can’t say robots but we are almost close to that with the increase in the need of artificial intelligence in our daily lives or more specifically the generative AI.




Generative AI is basically an artificial intelligence that generates new content according to the need of the customers rather than simply analyzing or acting on the pre existing data.

Chat GPT (Chat Generative Pretrained Transformer) created by OpenAI has been trending in last few weeks and gained a lot of popularity in a short span of time being launched on 30th November 2022.

You might have heard of Dall-E created by OpenAI which is a similar generative AI which creates image based on simple written texts. The new version Dall-E 2 can receive any input in form of text and turn them into photorealistic images that might have never existed before. Another example of such generative AI is Vall-E which can replicate any voice or audio in just matter of seconds.

Now before talking about ChatGPT it is important that we know about OpenAI. Well OpenAI was founded in 2015 by Sam Altman and the revolutionary Elon Musk. Although Elon Musk left the company in 2018 to focus more on Tesla, he remained a major donor and adviser of OpenAI. The business goal has been to advance digital intelligence to benefit the humanity.



Now this is not the first go at AI chatbot. Microsoft launched an chatbot Tay in 2016 which was basically made to chat with the college students for an entertainment purpose. However in the first 24 hours, it had to be stopped as Twitter claimed its texts as rude, racial and misogynistic. Meta had also launched Blenderbot 3 in but again it had to shut down due to the similar reasons. It was said to provide false information. Even to the question like President of USA, it answered Donald Trump.

However OpenAI is certainly not going to make that mistake again and hence it has employed moderation API (Application programming interface), which assists developers showing them when things go against their content policy like illegal or unsafe information. Obviously it is not cent percent accurate, but it certainly helps.

ChatGPT is an GPT 3 which can create some content at an incredible speed. It is basically a model which interacts with the consumer in an interactive way. Its dialog format is made to answer the questions asked by a person, challenge incorrect premises and reject inappropriate requests. But that’s not all, it corrects grammar, convert complex concepts into simple one, converts movies into emojis and even fixes the bugs in your programming codes.

As Times of India puts it, ChatGPT is ‘stunningly lifeline conversational language system’ and ‘world’s first truly useful chatbot’.


Well ChatGPT is presently at its early stage of development, and due to that it has very limited knowledge base and mostly connected to 2021.

Now while ChatGPT is very much useful for the mass, it has been a threat to the jobs of many people. It can create any programming code in just minutes. So what is need of the programmers? As this generative AI can produce human-like written texts, it can push the journalists out of their work.

For example, the Dall-E 2 now has been linked to Shutterstock which can create any image as requested thus pushing the online photo seller out of their job. However Shutterstock CEO Paul Hennessy has a different view. “I think there are two choices in the world. Be the blacksmiths that are saying, ‘cars are going to put us out of the horseshoe making business’, or be the technical leaders that bring people, maybe kicking or screaming, into the new world.”



Well, according to some people, it lacks the critical and ethical thinking skills that are required. This will give OpenAI an opportunity to fix the loopholes.

Some people may think that if ChatGPT can answer all our questions and even make the code itself for us, while Google will just give you the ways to learn them from different websites, why would anybody use Google? Well, people would. Because, we don’t know the sources of the answers that ChatGPT provides us. So there may be some incorrect or inappropriate information.  And this is the biggest problem to the generative AI tool, because anything that we read – we should know its sources, otherwise we would just be misled believing everything it says.

That's it for today. Hope you got to learn something new.

 


Post a Comment

Previous Post Next Post