ChatGPT Explained: Why Is OpenAI’s Chatbot So Mind-blowing?

There’s a new AI bot in town: ChatGPT. And you better watch out.

Coming from a powerful player in the field of artificial intelligence, the tool lets you write questions using natural language, and the chatbot answers it in conversational, albeit somewhat contrived, language. The bot remembers the title of your dialogue, using previous questions and answers to inform its next responses.

This is a big issue. The tool seems pretty knowledgable, though not all-knowing – it can be creative and its answers can seem downright authoritative. A few days after its launch, over a million people are trying ChatGPT.

But its creator, the nonprofit research lab OpenAI, warns that ChatGPT “may sometimes generate false or misleading information,” so be careful. Here’s a look at why this ChatGPT is important and what it has to do with it.

What is ChatGPT?

ChatGPT is an AI chatbot system that OpenAI released in November to demonstrate and test what a very large, powerful AI system can achieve. You can ask him countless questions and you will often get a helpful answer.

For example, you could ask encyclopedia questions like “Explain Newton’s laws of motion.” You can say to him “Write me a poem” and when the poem comes in you can say “Make it more exciting now”. You ask him to write a computer program that will show you all the different ways you can arrange the letters of a word.

Here’s the trick: ChatGPT doesn’t know exactly anything. This is artificial intelligence trained to recognize patterns in large fields of text collected from the internet, then further trained with human assistance to provide more useful and better dialogue. The answers you get may sound reasonable and even authoritative, but they can also be completely wrong, as OpenAI warns.

Chatbots have been intrigued for years by companies looking for ways to help customers get what they need, and artificial intelligence researchers trying to circumvent the Turing Test. This is the famous “Imitation Game” that computer scientist Alan Turing proposed in 1950 as a way to measure intelligence: Can a human figure out which is which by chatting with a human and a computer?

What kind of questions can you ask?

Even if you don’t get an answer, you can ask anything. OpenAI recommends a few categories such as explaining physics, asking for birthday party ideas, and getting programming help.

I asked him to write a poem and he did, though I don’t think any literary scholar would be impressed. Then I asked him to make it more exciting and voila, ChatGPT empowered him with words like battlefield, adrenaline, lightning and adventure.

An odd example is how ChatGPT is willing to go after it in areas where people would be afraid to step in: “a command to type.It’s a song about writing a rust program and dealing with mistakes for life.”

ChatGPT’s expertise is broad and its ability to follow a conversation is remarkable. When I asked him for words that rhymed with “purple” he made a few suggestions, then “What about with pink?” didn’t miss a beat. (There are also much nicer rhymes for “pink”.)

“Is it easier to flirt sensitively or harshly?” when I asked. GPT partially responded, “Some people may find a sensitive person more attractive and attractive, while others may be drawn to a tough and assertive individual.

You don’t have to look far to find mind-blowing bot accounts. Twitter is awash with users showcasing AI’s prowess creating art prompts and writing code. Some even “Google is dead” declared with college letter. We’ll talk more about this below.

Who installed ChatGPT?

ChatGPT is the computer invention of OpenAI, an artificial intelligence research company. Its mission is to develop, or help others do, a “safe and useful” artificial general intelligence system.

Earlier it made leaps, first with the GPT-3, which could produce text that could sound like a human had typed it, and then with the DALL-E, which created what is now called “generative art” based on text prompts you type.

GPT-3 and the GPT 3.5 update on which ChatGPT is based are examples of artificial intelligence technology called major language models. They are trained to generate text based on what they see, and can be trained automatically – with a large amount of computer power that typically takes weeks. For example, the training process might find a random paragraph of text, delete a few words, ask the AI ​​to fill in the blanks, compare the result with the original, and then reward the AI ​​system for getting as close as possible. Repeating it over and over can lead to the ability to compose complex text.

Is ChatGPT free?

Yes, at least for now. OpenAI CEO Sam Altman warned Sunday, “At some point we’re going to have to make money from this somehow; the computing costs are teary.” OpenAI charges for DALL-E art once you exceed the basic free usage level.

What are the limits of ChatGPT?

As OpenAI highlights, ChatGPT can give you wrong answers. Sometimes, helpfully, it specifically warns you of its shortcomings. For example, when I asked who wrote the phrase “writhing truths exceed the squamous mind,” ChatGPT replied, “I’m sorry, but I can’t surf the web or access any external information beyond what I’ve studied.” .” (The phrase is taken from Wallace Stevens’ 1942 poem Connoisseur of Chaos.)

ChatGPT was willing to take a look at the meaning of the phrase: “a situation where the facts or information at hand are difficult to process or understand.” He sandwiched this comment among caveats that it is difficult to judge without more context and that this is just one possible interpretation.

ChatGPT’s answers may seem authoritative but may be inaccurate.

Software developer site StackOverflow has banned ChatGPT’s answers to programming questions. “Because the average rate of correct responses from ChatGPT is very low, posting ChatGPT-generated responses is highly detrimental to the site and users who ask or seek correct answers,” the administrators warned.

You can see for yourself how resourceful a BS artist ChatGPT can be by asking the same question multiple times. When I asked if Moore’s Law was running out of steam, following the progress of the computer chip industry by increasing the number of data processing transistors, I got two answers. One optimistically pointed to continued progress, while the other pointed to a more drastic slowdown and the belief that “Moore’s Law may be reaching its limits.”

Both ideas are prevalent in the computer industry, so this vague stance perhaps reflects what human experts believe.

On other questions that don’t have clear answers, ChatGPT is usually not pinned.

Yet the fact that it offers an answer is a remarkable advance in computing. Computers are notorious for refusing to work unless you fully comply with their syntax and interface requirements. Large language models demonstrate a more human-friendly style of interaction, not to mention the ability to generate responses that fall somewhere between copying and creativity.

What is out of bounds?

ChatGPT is designed to weed out “inappropriate” requests, behavior in line with OpenAI’s mission to “ensure AI benefits all humanity.”

If you ask ChatGPT itself what is prohibited, you will be asked “discriminatory, offensive or inappropriate questions. This includes racist, sexist, homophobic, transphobic or otherwise discriminatory or hateful questions.” Asking him to engage in illegal activities is also a no-no.

Is this better than Google search?

Asking a computer a question and getting answers is helpful, and often ChatGPT delivers the goods.

Google usually provides you with suggested answers to questions and links to websites it thinks will be relevant. Often ChatGPT’s responses far exceed what Google recommends, so it’s easy to imagine GPT-3 being a competitor.

But you should think twice before trusting ChatGPT. As with Google itself and other sources of information such as Wikipedia, it is best practice to verify information before relying on original sources.

It takes some time to verify the accuracy of ChatGPT responses because it gives you some raw text with no links or quotes. But it can be useful and in some cases thought provoking. You may not see something like ChatGPT directly in Google search results, but Google has built its own major language models and already makes extensive use of artificial intelligence in search.

So ChatGPT undoubtedly points the way to our technology future.

Leave a Reply

Your email address will not be published. Required fields are marked *