The release of ChatGPT by OpenAI ushered in a new era of artificial intelligence. With this app, humans can chat with machine learning-based AI to get relevant answers. Technology offers almost endless possibilities for research and content production. But it’s still in development.
The launch of ChatGPT marks a new milestone in artificial intelligence. It is an innovative tool that will create conversations between humans and an artificial intelligence (AI) to provide advice and answers to your questions.
The concept of ChatGPT is simple. It is an online application that connects a human with an AI based on machine learning. When you send the AI a question, it taps into its vast pool of knowledge and provides relevant answers. What makes ChatGPT a very interesting tool is that it can adapt to more complex conversations and provide more accurate responses.
What you have just read is the result of the work of the OpenAI company’s artificial intelligence, GPT-3 (Generative Pre-trained Transformer 3). It is a discussion space with artificial intelligence in writing, open to the general public. Just create an account. Nothing new. But here, the AI is so good it’s scary.
Discussion is natural, intelligent. The computer actually follows the thread of the conversation and uses your responses. It is possible to issue a command without computer code or specific vocabulary, but simply by asking a question.
For example, “Write me an article about your launch.” Even stronger, it can turn your texts and ideas into computer code. He becomes a programming assistant. We are at the beginning of a new relationship with computing that no longer involves lines of code. true democratization.
A multitude of possibilities
What can he do exactly? Here’s your answer: I’m Assistant, a great OpenAI-trained language model. I am a computer program designed to answer your questions and help solve your problems.
And since the November 30th update, the qualitative leap is impressive. With the Playground option, the possibilities are almost endless. It seems your imagination is the only limit.
GPT-3 can write poems, essays, provide arguments for your debates, but also find the chords of your favorite song. He mimics writing styles, makes up stories, corrects his texts. The AI also writes apology emails or birthday cards.
GPT-3 can offer you SMS [Open AI]
One has the impression of speaking with a universal genius. No wonder the model is called davinci-003. It will instantly extract from the millions of texts it has read and digested.
If Google provides links, GPT-3 will talk to you. It reinvents the way we search for our information on the internet.
Don’t need Google anymore?
In spending time with GPT-3, we realized that this is the beginning of a new information access interface. Being able to exchange facilitates the search.
Artificial intelligence surfs for you. It’s fast, efficient, but not always fair. Care is in order. Playground mode is still under development. When he cannot find the information, he tends to lie. As shown in the example below.
In its “playground” part, GPT-3 does not hesitate to invent information, to create false links and even to lie. [Open AI]
GPT-3 even invents internet links that don’t exist to justify its claims.
Keep your critical sense in front of the machine
Technology is risky if not tempered and framed. AI is so powerful that it can be used to deceive Internet users. We always tend to believe in the machine. The answers are convincing and coherent.
It is also difficult to distinguish it from a text written by a human being. And here, just one click is enough to produce a lot of content. This facilitates the spread of false information.
Not to mention students who could write assignments automatically. There is also the risk of copyright infringement, as it is difficult to know what pure copy is in the flood of words produced by AI. This technology can also be used for malicious purposes, such as phishing or spreading computer viruses.
Meanwhile, ChatGPT is starting to moderate. He no longer answers all questions.
ChatGPT moderation is refined over time [Open AI]