The launch of ChatGPT will have allowed the general public to discover the capabilities of OpenAI models, in particular those based on GPT-3. These technologies have been accessible since 2020 via the publisher’s API. And recently, developers can even integrate ChatGPT technology, based on GPT-3.5, into their applications thanks to the availability of a dedicated API.

The success of ChatGPT heightens expectations – and speculation – around OpenAI’s next language model, namely GPT-4. The latest rumors mentioned a launch during the first quarter of 2023, and this could well be verified… as early as next week.

“We will introduce GPT-4 next week”

During the KI im Fokus event organized by Microsoft, the CTO of the German branch of the company released two important pieces of information.

We will introduce GPT-4 next week, and we will have multimodal models which will offer completely different possibilities – for example videos.

Clearly, the next OpenAI language model would be presented next week – by the publisher, by Microsoft, or jointly by the two companies? That’s pretty good: Satya Nadella and Jared Spataro have planned announcements on artificial intelligence next Thursday, March 16. But we were rather expecting novelties related to the application of AI in the world of work.

A multimodal artificial intelligence?

Another important piece of data, Andreas Braun mentions, in the same sentence, the arrival of multimodal models – without clearly saying whether or not this concerns GPT-4 – that is to say capable of supporting several types of content, including the video.

Whatever happens, this will be a major evolution for artificial intelligence models. These announcements come a week after the presentation of Kosmos-1 by Microsoft researchers, a rather impressive multimodal AI model capable of interpreting the contents of an image.

What we know about GPT-4

At the moment, we know relatively little about GPT-4. On the date of presentation of the model, the CTO of Microsoft Germany seems in any case to confirm the statements of the New York Times. The CEO of OpenAI, Sam Altman, had however tempered the information, specifying that GPT-4 would be released when he was convinced that it could be done. “safely, responsibly”. Also, remember that it will probably take some time between the presentation of GPT-4 and its availability via an API or an interface like ChatGPT.

On the capabilities of GPT-4, Sam Altman also indicated that “People just want to be disappointed and they will be”. An astronomical number of parameters had initially been mentioned: 100,000 billion, or 571 times the size of the GPT-3 neural network. It would seem that the innovations brought by GPT-4 are more at the level of the optimization of the training of the model. We hope to know more next week!

Source: Heise.