According to new reports, OpenAI has begun rolling out a more streamlined approach to how people use ChatGPT. The new system will allow the AI to choose a model automatically, letting you run Python code, open a web browser, or generate images with DALL-E without extra interaction. Additionally, ChatGPT will now let you upload and analyze files.
Tech. Entertainment. Science. Your inbox.
Sign up for the most interesting tech & entertainment news out there.
The new features are part of the latest beta updates for ChatGPT Plus members, and many subscribers already have access to the change. Multimodal support is a considerable change because it doesn’t require manually changing what model you want to use in the responses ChatGPT gives you.
On top of multimodal support, ChatGPT Plus users will also get one of the more office-oriented features from ChatGPT Enterprise, which lets you upload files like PDFs to the chatbot. From there, ChatGPT can analyze the files and provide information about them – like summarizing the data, answering questions about it, and even generating visualizations based on prompts.
Open AI’s ChatGPT start page. Image source: Jonathan S. Geller
This is, of course, just one way that ChatGPT continues to improve as OpenAI works to make the large language model even more impressive. Another recent change saw OpenAI give ChatGPT full internet access, allowing it to browse the web without any kind of additional plugins being needed.
While there have been plugins for ChatGPT that allow many of these things to happen, the fact that OpenAI is adding them in natively to the software is huge for people who utilize ChatGPt every day. In fact, the ability to upload files to ChatGPT and have the system analyze them will no doubt open new doors for users who weren’t utilizing plugins before.
With the continued improvement of ChatGPT, some are even calling for a ChatGPT-powered phone or laptop, which would allow the AI to fully integrate with the operating system and improve how it works.