GPT-4o could revolutionize AI assistive technologies

On May 13, 2024, OpenAI introduced its latest language model called GPT-4o. It will be exciting!

A robot (left) and a human (right) sitting at a table in a restaurant, the robot pointing at the human, a keyboard in front of it, and the human holding a cell phone in its hands — Mid-flight

AI and ChatGPT in particular have now reached the population. ChatGPT allows you to communicate with your computer in natural language and answer your questions.

the new OpenAI's GPT-4o model was introduced on May 13, 2024. It's what's called a multimodal model, which means it can process text, speech, images, and videos.

So far only language processing has been published, and image processing options will gradually be opened. But what will be exciting are the audio-visual capabilities, from real-time language translations to live support. Many things should be possible when learning.

From an accessibility perspective, the following video caused a stir: Andy from Be My Eyes described his surroundings liveDo not – as before – take a picture and wait for a description.

There are already applications that describe the environment in real time, such as Seeing AI or Lookout, but not in this natural way and with this quality. I'm excited to see when these features will be released.

It's possible that Be My Eyes has been around again since the beginning, as was the case with GPT-4 and image descriptions.

It is not yet possible to fully predict what new use cases will arise for AI-based assistance technologies.