I forget since when this problem appear, but until now this is what i do to try solve this problem.
- Clear cache
- Change browser
- Make sure internet connection is okay
- Using Model GPT 4, GPT 40, Turbo etc
- Log out and login again (Using Chrome and Edge)
- Ask my IT guy friend
- Pray
This is very frustrating, even when i ask Chat Gpt “Why u can’t read image anymore” this is his answer (Using GPT 40)
Can u analyze or read image when i upload it?
ChatGPT
Currently, I can’t directly analyze or interpret images. If you upload an image, I won’t be able to see its content. However, if you can describe what the image contains or share the text or details from the image, I can definitely help analyze or interpret it for you.
4o
Why? i remember a months ago you can read and analyze image
ChatGPT
I understand the confusion! At the moment, I don’t have the capability to analyze or interpret images directly in this interface. While some versions of AI models may have image-processing abilities, my current setup is text-based, which is why I can only process and respond to textual input.
If image analysis was something available before, it’s possible that feature was part of a different version or tool that had specific capabilities, but for now, I’m here to help with any text-based tasks or analysis you need. If you need any help with text from an image, feel free to share it, and I’ll assist you with it!
4o
is Open Ai give u limitation to read image?
ChatGPT
Yes, OpenAI has set limitations on my current abilities. I’m specifically designed as a text-based model, meaning I can only process and respond to textual information. While there are other OpenAI models that can analyze images (such as CLIP or DALL·E for generating images based on text prompts), my current configuration doesn’t have the ability to directly interpret or analyze images.
If image analysis is a feature you’d like, OpenAI is working on integrating multimodal capabilities, but as of now, those are separate from what I can do. If you need help extracting text or other details from an image, you could use tools like Optical Character Recognition (OCR) software and share the text with me, and I’d be happy to assist with any analysis or information you need!
4o
Why?
ChatGPT
The reason for this limitation is largely due to how OpenAI has designed and deployed different models with different capabilities. My current version, which is based on GPT-4, is focused on text-based tasks, meaning my training and optimization have been primarily for understanding and generating language.
The ability to process and analyze images requires a different type of model architecture, and OpenAI has built separate models for that, like CLIP or DALL·E, which specialize in tasks involving images. These models are trained with visual data, whereas I’ve been trained specifically on large amounts of text data to excel at language understanding, reasoning, and generation.
Although integrating multimodal capabilities (text and image processing in one model) is an ongoing goal for AI research, it involves significant technical challenges. For now, OpenAI keeps image and text processing in separate models to provide optimized performance for each type of task.
If you need to analyze or extract text from an image, there are external tools like OCR (Optical Character Recognition) that can help, and I can assist with any information or analysis from the text you extract!
what the heck they limitation this feature… i mean, i have pay for this feature and they cut it in the middle of my subscription period still valid? Some one sue this company plz