Computer

6 mistakes you are making when using Artificial Intelligence

Artificial Intelligences have been with us for many years, although it seems that they were born last year with the launch of ChatGPT, Midjourney and others, platforms that have popularized the use of this technology among users who, until now, were completely unaware of what its usefulness could be on a day-to-day basis, leaving aside the fears that the Terminator movies showed us.

However, we must bear in mind that any Artificial Intelligence It’s not perfect and it’s far from it.at least at present, therefore, whether we use them regularly or if we want to start doing so, we must take into account a series of aspects.

not all are the same

While OpenAI’s ChatGPT does not have internet connection, Microsoft’s Bing Chat if you have internet access, so the answers that it offers us based on current questions are usually, in most cases, consistent with reality. If we talk about AIs to generate images, the one that offers us the best results is Midjourney, especially with the launch of version 5, a version that is capable of recognizing the natural language of a description to generate any type of image without the need to use commas to separate the aspects that we want the image we are looking for to have.

trust blindly

AIs, especially those that create text based on a description, are far from perfect. Sometimes they approach solving a problem in the most absurd way possible taking into account the entire context. In addition, some, such as ChatGPT, the most recent updated information is from 2021, although if we ask him about events that happened later, he is capable of answering, although obviously incorrectly.

ChatGPT updated data

set a goal

When working with any AI, the first thing we must do is establish the goal we want to reach. If we want to create a job related to a certain subject, it won’t do us any good to ask the corresponding AI to generate it automatically. We must, first of all, know the possibilities that it offers us to find the structure that it suggests to us to carry out the project. If it does not convince us, we should not continue wasting time and we can try other ones that are more complete.

Other factors to consider

In addition to the three factors that we have shown you on these lines, we must also take into account the three points that we show you below.

Chat GPT Detector

  • Updates include news. With the launch of each update of the different AIs currently available, improvements are introduced in the way of presenting the data, data that is presented in a much more concise and coherent way, reducing the indications that the content has been generated. by an Artificial Intelligence. If it is an AI to generate text, it is advisable to opt for those that use GPT 4 instead of GPT 3.5
  • Check the result. Although it seems that the AIs know everything and are not wrong, unfortunately this is not the case. The AIs that allow you to generate text based on a description use different methods to give cohesion to the information they find on the internet and, on many occasions, the interpretation they make when generating the text is incorrect, so we must ensure that the result it’s right.
  • It’s easy to spot. On the Internet we can find a large number of platforms that are capable of recognizing when a text has been written by Artificial Intelligence and when it has been done by a human, so, in addition to reviewing the content to verify that it is correct, we must only use it as a source for a job or project, not as the end result.

Related Articles

Leave a Reply

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