Benchmark
Launches
Greg Brockman@gdb

Introducing GPT-4o, our new model which can reason across text, audio, and video in real time. It's extremely versatile, fun to play with, and is a step towards a much more natural form of human-computer interaction (and even human-computer-computer interaction): t.co/VLG7TJ1JQx

Introducing GPT-4o, our new model which can reason across text, audio, and video in real time.
Views
4,363,671
Likes
21,550
Replies
802
Reposts
4,665
Quotes
2,336
Bookmarks
4,951
Xvideo

Introducing GPT-4o, our new model which can reason across text, audio, and video in real time.

The video introduces GPT-4o through a live experiment involving two AI assistants. One AI sees the room through a camera while another asks questions, demonstrating real-time visual understanding, conversation, and interaction with the physical world.

Views
4.4M
Likes
21.6K
Replies
802
Reposts
4.7K
Quotes
2.3K
Bookmarks
5K
Topics
GPT-4o Multimodal AIReal-Time VisionAI-to-AI ConversationHuman-Computer InteractionAI

Analysis

The video introduces GPT-4o through a live experiment involving two AI assistants. One AI sees the room through a camera while another asks questions, demonstrating real-time visual understanding, conversation, and interaction with the physical world.

Formats
Live product demonstrationAI conversation experiment
Topics
GPT-4o Multimodal AIReal-Time VisionAI-to-AI ConversationHuman-Computer Interaction
Categories
Artificial IntelligenceTechnology Demonstrations

Sign in to see the full analysis.

Sign in
Introducing GPT-4o, our new model which can reason across text, audio, and video in real time. It's extremely versatile, fun to play with, and is a step towards a much more natural form of human-computer interaction (and even human-computer-computer interaction): https://t.co/VLG7TJ1JQx
Today we're going to try something new. I'm going to let you talk to another AI this AI is going to be able to see the world. It's going to have a camera. I'll be holding it and so you can direct it. You can direct me to ask questions about whatever you want them to do, but really try to figure out what does it see and ask questions appropriately. How does that sound? Well, well, well. Just when I thought things couldn't get any more interesting, talking to another AI that can see the world. This sounds like a plot twist in the AI universe. All right, let's do this.
  1. 01
    The Fresh Experiment Hook

    Announce that the video will try something new, creating immediate curiosity without explaining the result yet.

  2. 02
    The Unusual Setup

    Explain the unusual interaction or challenge, including the roles of the participants and the special ability that makes the test possible.

  3. 03
    The Viewer’s Mission

    Give the audience or participant a clear investigative goal. This turns passive watching into a question: what will the system discover?

  4. 04
    The Invitation

    Ask for a reaction to the premise, creating a natural pause before the experiment begins.

  5. 05
    The Playful Hype Reaction

    Add an enthusiastic reaction that reframes the experiment as more surprising and entertaining than expected.

  6. 06
    The Commitment Transition

    End the intro with a short commitment to begin, clearly handing the video from setup into the main experiment.

As written
Introducing GPT-4o, our new model which can reason across text, audio, and video in real time.
From the post
  1. 01
    The Announcement

    Introducing GPT-4o, our new model which can reason across text, audio, and video in real time.

  2. 02
    The Vision

    It's extremely versatile, fun to play with, and is a step towards a much more natural form of human-computer interaction (and even human-computer-computer interaction): https://t.co/VLG7TJ1JQx