Microsoft officials have confirmed some details regarding the rumored GPT-4, including its release date and major upgrade. And according to the details, it won’t take months or weeks to arrive.
Surely, we can expect it to be more advance than ChatGPT and GPT 3.5, but as the information expressed about it, it seems way better than our expectation as it will be multimodal. Let’s delve into the details below.
Open AI’s GPT-4 To Have Multimodal Language Models
The GPT-4 AI technology, the successor of GPT-3 AI technology that powers popular ChatGPT, will be developed by Open AI, and its launch will take place this week.
And there is no path to doubt this information as it was revealed by Andreas Braun, CTO of Microsoft Germany, in an event named “AI in Focus – Digital Kickoff“.
As we all know, the AI-powered ChatGPT and Bing Chat currently only allow text inputs to interact with users, and also it shows all replies and results in the form of text, and it could change soon.
With multimodal capabilities, GPT-4 will allow users to interact in various methods such as text, images, sounds, and maybe even videos too.
For example, it will be able to transcript the audio file and can extract the text from the given image.
Besides, this GPT-4 will have the ability to generate AI videos with the given text by users, which will also be the primary feature of ChatGPT’s upcoming app.
With all those features, there will be some other improvements, such as generating human-like answers, providing results more speedily, and more.
Earlier, Microsoft also revealed that it crossed 100 million daily active users on its Bing search engine. The major reason behind this success is the recently launched Bing Chat powered by GPT-3.5.
However, Andreas Braun hasn’t mentioned anything about GPT-4 implementation in Bing Chat, so we can’t say exactly when its user we’ll get all those features that we mentioned above.