In a recent interview with OpenAI CEO Sam Altman, he said in an interview at the Davos Forum that his priority now is to launch the next generation of large models, which may be called GPT-5, which "can do more, more things" than existing models.
Sam Altman emphasizes multimodal capabilities, which are the fusion of different forms of input and output, such as speech, image, and finality. This advancement could change the way we interact with AI, making it more intuitive and versatile.
The launch of GPT-5 will be a key step towards achieving artificial general intelligence.
GPT-5's language processing capabilities will be greatly improved and can handle more complex tasks; Its range of knowledge will also be expanded and its comprehension will be stronger. This means that GPT-5 can override and replace many patching schemes designed for GPT-4 limitations.
Seizing the opportunities brought by GPT-5 is the more important direction.
Sam Altman said in an interview that as models become smarter, less and less training data is needed.
You don't need a lot of data to train a high-quality model, the key is the quality of the data and the training method of the model.
For example, humans don't need to read all biology textbooks to have a good understanding of biology, they just need some high-quality data and deep thinking.
OpenAI CEO Sam Altman plans to raise billions of dollars to build AI chip factories.