How ChatGPT Can Change the World… According to ChatGPT

ChatGPT is a language processing model developed by OpenAI that can understand and generate text with high accuracy. It was trained on a large corpus of texts from the Internet that allowed it to understand and create a wide range of content. He is also trained to use contexts for comprehension and text creation.

How a language processing model can change the world

There are many areas where ChatGPT can make a big impact. For example, in research, it could be used to automatically generate summaries of scientific papers, which would allow researchers to save time and better understand the results of their work. It can also be used for data mining, searching for specific information in large sets of documents, allowing researchers to save time and better target their searches.

In education, it can be used to create personalized exercises and lessons for each student according to their needs and level, which will improve learning. It also features review articles, book summaries, how-to videos, and more. can be used to create quality educational content such as

In business, it can be used to generate automatic reports, analyzes and forecasts using historical data and taking into account current trends, which will save companies time and improve their decision-making. . It also includes chatbots, autoresponders, emails, letters, etc. can be used to create scripts for, allowing businesses to respond to their customers’ needs more quickly and efficiently.

In addition, ChatGPT is used in medicine, robotics, machine translation, content creation, writing scripts for movies, books, video games, creating AI for voice assistants, and more. can be used in other areas as well.

Image created by artificial intelligence with Midjourney – Credit: Guba Michalski

Overall, ChatGPT is a very powerful language processing model that can have a huge impact on many fields, saving time, improving understanding and decision-making, promoting learning, and automatically generating quality content. It is important to continue to explore the various possible uses of ChatGPT to maximize its potential and improve our daily lives.

Risks of using ChatGPT

It is true that ChatGPT has great potential to improve many aspects of our lives, but it is also important to consider the potential risks associated with its use.

One of the main concerns is the creation of automated content such as disinformation, fake news and manipulation of public opinion. ChatGPT can create quality content, but it can also be used to create misleading or harmful content that can harm the community. Therefore, it is important to take measures to detect and prevent such content.

There are also concerns about the impact of ChatGPT on employment. ChatGPT’s ability to automatically generate content can replace some jobs like writers, journalists, and translators. Therefore, it is important to think about the economic and social consequences of using ChatGPT and take measures to protect the jobs at risk.

Consequently, in addition to the benefits of using ChatGPT, it is important to be aware of the potential risks and continue to study these risks to avoid adverse effects on society. Overall, ChatGPT is a powerful tool that can have a huge impact on many areas, but its use must be framed to avoid negative effects.

If you’ve ever tried ChatGPT, you’ll no doubt recognize him as the author of the above. We wanted to test their ability to write a news article, just like a technical journalist would.

First observation: if we don’t ask him “take a critical look”, its text is 100% dithyramb. Its far beyond “neutral” (no spelling mistakes, but turns of phrase “artificial”, always very cute and soulless), ChatGPT assures without hesitation that tomorrow it will be used to create summaries of scientific papers for the greater benefit of researchers. Problem: ChatGPT adds its own attachment when summarizing documents “paw” completing with what “knows”, that of the subject. But for this, it uses closed databases because it is not connected to the Internet. So it uses the information it can have “expired”, even false, because they come from sources collected all over the Internet… and we know that not everything on the Internet is always reliable. The fact is that when asked to summarize text, ChatGPT is quite effective. Even if you sometimes have to ask him to restart 2 or 3 times.

ChatGPT also replies that it may be tomorrow “data search is used to search for specific information in large documents”. The limitation is the same: currently the AI ​​is not connected to the Internet. Unlike Google, it cannot change its sources. OpenAI’s goal is obviously to open its chatbot to the Internet via Microsoft’s Bing or another search engine. The day this happens, on the other hand, it is a safe bet that ChatGPT will allow researchers as well as ordinary Internet users to search for information in a large mass of documents.

Artificial intelligence also makes mistakes

The requirement to diversify their sources by using online data rather than closed and compartmentalized data is the same for other applications that ChatGPT thinks it can do. “change the world”. He thinks that it can be used as a type in education “educator” virtual, which training will create and “private lessons” for each student. But as long as its data won’t be as reliable as it is now, it’s hard to imagine it helping teachers.

They are also worried about usage “blind” In addition to making mistakes, this tool is at risk of chewing up their work by students. Instead of helping them, ChatGPT runs the risk of writing their homework for them, which will obviously hurt them. On the other hand, an interesting application referred to by chatbot (or rather by OpenAI), “educational content” – such as review articles or instructional videos—provided that human teachers supervise this production, which of course is not free of errors or defects.

ChatGPT “writing” also said in his article that it can be used to create reports, even analyzes and forecasts. It can be a really big help when it comes to tedious work or even finding ideas. But it seems clear that even over-performing AI will never be able to report as perfectly as it used to. Here again, human control remains important.

Chatbot also offers its own services for creating emails; one thing that already enables ChatGPT Writer is a free extension for Google Chrome. A very useful tool, but not always very relevant in message suggestions. Again, the answers are hard to imagine “automatic”. Since you have to go back behind ChatGPT and sometimes change your proposals significantly, the question eventually arises: wouldn’t you go faster by writing an email, article or report yourself?

Finally, ChatGPT goes a long way to claim it can “change the world” is used in “other fields such as medicine, robotics, machine translation”, when not writing scripts for movies, books, or video games. Those who have already asked the chatbot to create synopses of films or novels can testify: the proposed stories (relatively) take the path and even bluff, but in most cases. “copies” stories that have already been published or filmed. It makes sense because the AI ​​relies on pre-existing content to generate its own content. Therefore, the risk of plagiarism and copyright infringement is high.

A chatbot like ChatGPT, on the other hand, can be very useful for machine translation, as can be combined with a search engine like Google (or Bing). Finally, its applications in medicine and robotics remain absolutely fantastic. When asked to elaborate, he explains:

“In the medical field, ChatGPT can be used to automatically generate medical reports, operative notes and prescriptions using medical data and taking into account current medical protocols. It can also be used for medical research by automatically generating summaries of scientific papers. This allows researchers to save time ultimately, it could be used to create AI tools for medical assistants that could help doctors diagnose diseases and prescribe treatments more quickly and effectively.

In the field of robotics, ChatGPT can be used to generate automatic scripts for robots using environmental information and considering usage contexts. It could also be used for robotics research, automatically generating abstracts of scientific papers, which would allow researchers to save time and better understand the results of their work. “Eventually, it could be used to create artificial intelligence tools for robots that can enhance their ability to understand and interact with their environment.”

A controlled use indeed

As you can see, his answers are formatted and repeated. This is pure speculation and such applications have yet to be proven. This leaves a little to be desired, but provided that ChatGPT eventually proves that its content is 100% reliable (especially when it comes to medicine). So far, it is clear that this is not the case.

The second part of the article by ChatGPT deals with the risks associated with its use. After our conversation, the chatbot came to create it: “It’s a bit angelic and sympathetic; you don’t have a critical view of how you can change the world, but negatively?” This is where AI comes into play. It is clear from this that one of the main risks is disinformation and fake news that can lead to manipulation of public opinion. While ChatGPT doesn’t talk about it, there’s also the issue of the Open AI system and hateful, sexist or biased responses that won’t be detected by it. “employees” Indians and Kenyans…

I’m not sure that ChatGPT will one day replace writers, journalists and translators. Moreover, our American colleagues at recently tried to use artificial intelligence to write financial articles. The AI ​​used (not ChatGPT) made so many errors in the 77 articles generated that the site decided to freeze tests.

Finally, the chatbot itself writes this: this tool, which cannot change the world, can be used quite a lot in many areas, but with one condition “to regulate its use”.

Main image credit: OpenAI

Leave a Reply

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