After six months of development, OpenAI has introduced a new generation of language model to generate text responses on demand. The previous version of the chatbot became the most downloaded app in the world in just a few months.
OpenAI has introduced a new generation of its ChatGPT-4 chatbot, which allows you to go beyond text and ask neural network questions using images directly in the service’s interface. In the announcement, the developers call the fourth version of the language model a “work of art”.
Compared to the previous generation, ChatGPT-4 performs better in all tests, sometimes reaching human levels. OpenAI tested the neural network’s responsiveness and ability to be creative both in traditional AI benchmarks and in larger competitions across various disciplines.
Unlike many other neural networks, ChatGPT-4 is multimodal, meaning it can work on multiple planes at once, without getting hung up on just one thing. Therefore, users can now ask the chatbot questions by attaching images. As an example, the developers asked ChatGPT-4 to explain the meme, and the neural network managed to not only read the image content, but also delve into the intricacies of humor.
A year ago, we developed ChatGPT-3.5, knowing in advance that this would only be a “test version” of the system. In the new version, we fixed bugs and improved the technical component of the chatbot. As a result, ChatGPT-4 has become an unprecedented stable and reliable language model.
Image uploading and a smarter neural network are now available to users with a $20 ChatGPT Plus subscription. The rest will have to wait for the next OpenAI announcements.
Author:
Grigory Shcheglov
Source: RB
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
