OpenAI’s chat bot has plenty of fun use cases, but can ChatGPT read images? GPT-4 is the latest Generative Pre-Trained Transformer created by the artificial intelligence company OpenAI. Having already caused a scene, breaking boundaries with its renowned chatbot Chat-GPT, the research & development firm has impressed once again with the intelligent capabilities of GPT-4.

Following the success of GPT-3.5, GPT-4 takes things to the next level. It's been described as “more reliable, creative, and able to handle much more nuanced instructions than GPT-3.5”.

Essential AI Tools

Can GPT-4 read images? In short, yes.

Yes, GPT-4 can read images. According to OpenAI, “GPT-4 can accept a prompt of text and image”. What this means is that after inputting an image into GPT-4, it will output natural language, code, instructions, or artificial opinions as a response to a photo. This is called multimodality – when an AI can translate between different forms of digital media. In this case, we’re talking about image input to text output, but this also applies the other way around – something we expect to see in GPT-5.

So yes, GPT-4 can prepare a cohesive response that uses both sets of data, text and image, as an input. This development means that GPT will now have the ability to interpret graphs and predict complex patterns, one of the biggest leaps for this multi-modal AI yet. This has huge implications for both business and creative sectors, as users can engage in increasingly complex discussions with this piece of artificial intelligence.

However, despite developments in its ability to read images, GPT is still limited to text and cannot yet create an image as a response as it is a language-based model. If you are interested in AI software that can create images based on text prompts, OpenAI also created DALL-e

Can ChatGPT read images?

Yes, ChatGPT can read images. As OpenAI’s flagship product, the worlds most popular AI chat bot is constantly being improved with new features, including the multimodality required to pull off this particular trick.

ChatGPT currently uses GPT 4 as its LLM (Large Language Model). That means that anything GPT 4 can do, ChatGPT can do too!

Can I put images into GPT-4 yet?

GPT-4 has officially been released, and you can now have a go at using the brand-new model via Chat GPT if you have a Chat GPT Plus subscription. Otherwise, the next best way to access GPT-4 is by signing up for the OpenAI API waitlist while the company gradually rolls out the latest update in its line of research.

About the Author

Marla Broadway

Marla writes across a wide range of topics across PC Guide, including AI, PC hardware, and news on the latest tech releases. She's a passionate writer that's interested in the future of technology.