OpenAI, the creators of ChatGPT, recently showed off a new version of ChatGPT, GPT-4. AI is able to solve complex problems and communicate with users. However, the most important new feature is the GPT-4’s ability to interpret images.

Previously, OpenAI demonstrated Dall-E, which allowed users to write a description, and the program created the corresponding image. However, GPT-4 represents a significant step forward in this regard.

During the presentation, OpenAI showed how AI can work with different types of data provided, such as text, programming snippets, and different languages, including Portuguese and English. The GPT-4 is able to distinguish between languages ​​as well as understand programming codes.

During the presentation, the AI ​​also demonstrated its ability to identify funny elements in an image and explain why those elements are funny. For example, in an image of a squirrel holding a camera, GPT-4 identified the situation as funny because squirrels don’t normally use cameras. In addition, the AI ​​was also able to identify and describe in detail the image of an astronaut exploring an alien planet with a spaceship in the background.

The GPT-4’s ability to interpret images is due to its neural network, which is trained on large amounts of data to predict what should happen next. With the amount of text and images available for cross information, AI is able to identify details and relationships that cannot be easily identified by humans.

In short, the new version of ChatGPT, GPT-4, represents a significant advance in AI’s image interpretation capabilities. Technology opens up new possibilities for more natural and intuitive interaction with machines.

…..