March 10 (Portaltic/EP) –
The next generation of the OpenAI language modelGPT 4, will be presented next week as confirmed by Microsoft, and in addition to improving interaction, it will incorporate the generation of videos.
The current version, GPT 3.5, power the ChatGPT ‘chatbot’ and the new Bing search engine based on artificial intelligence. This improvement compared to GPT 3, which means that in addition to understanding and generating natural language and code, as explained in Open AI.
The most advanced model is known as the GPT 3.5 Turbo, and it is updated with data up to September 2021, and according to the company, it is optimized for conversations and completing tasks.
The next generation of this language model is expected under the name of GPT 4, and as has been announced by the CTO of Microsoft Germany and Lead Data & AI STU, Andreas Braun, at the ‘AI in Focus – Digital Kickoff’ event, will be released next week.
GPT 4 will be a multimodal model, that is, users will be able to interact in multiple ways, not only through text, but also with images and sounds. Also you can generate videosas collected in gizmochina.
The implementation of GPT 4 will allow Microsoft streamline response delivery and make responses more detailed and humaneas sources close to these plans pointed out to Semafor, at the beginning of February.