OpenAI revealed another evolution of research in the field of artificial intelligence (AI) with the announcement of the launch of model GPT-4. After launching ChatGPT in November, a service similar to a chatbot, based on the GPT-3.5 language model, many eyes were on the Californian startup, waiting for further developments.
The wait came to an end this Tuesday, when the company presented the new language model. OpenAI explains that it is a “multimodal model”, that is, it will accept images and text from users, but will only respond with text. The image is already an improvement over GPT-3.5, which can only respond to text interactions.
ChatGPT brings chat to the Internet. The limits, dangers and promises of the new phenomenon of artificial intelligence
This article is exclusive to our subscribers: subscribe now and benefit from unlimited reading and other benefits. If you are already a subscriber, log in here. If you think this message is an error, please contact our customer service.
Source: Observadora
