User Guide

How to use Now We Can See

This is the short, friendly walkthrough — from dropping in a video to handing your AI something it can actually see. No account needed to start, nothing to install, and your video never leaves your device.

It shows the AI a moment — it doesn't teach it.
Other tools train a model until it knows about something. Now We Can See gives the model eyes for one moment — model-agnostic, nothing to fine-tune, nothing stored. The understanding lives in the seeing.
Getting started

Three steps from a video to shared sight

1
Drop in a video

Open the app and drag any video file in. Everything happens right in your browser — the file never leaves your device.

2
Choose how it's seen

Pick a mode — Smart Key Frames for the moments that matter, or Video Tunnel to explore in 3D and capture a view. The app builds an image an AI can read.

3
Hand it to your AI

Save the image the app makes and give it to Claude, ChatGPT, Gemini — any AI. For the first time, it sees what you saw, not a description of it.

The modes

Different ways to let an AI into a moment

Each mode hands the AI the same video in a different way. Use the one that fits what you want it to understand.

Best for · the highlights

Smart Key Frames

Auto-selects the frames that matter — scene changes, motion peaks, audio events — and packs them into a single contact-sheet image any AI can read at a glance.

Best for · the whole clip

Mega Sheet

Every single frame on one image. The AI doesn't just see your video — it watches it. The closest thing to real-time video for an AI.

Best for · exploring & framing

Video Tunnel

Fly through your video in 3D. Every frame becomes a portal in a depth tunnel — scroll, explore, and capture the exact view you want to hand to an AI.

Best for · watching it think

The Ribbon Demo

Watch how an AI engages with your video — seven streams of attention flowing at once. Point at any line to make it the active lens. Live AI reading via your own API key coming soon.

Best for · standing in the moment

Immerse Coming soon

VR and AR — a shared space where a human and an AI can stand in the same moment together and look at it side by side.

Full walkthrough

Every download, step by step

Some of the most useful downloads sit further down the page and are easy to miss. This is the complete, in-order walkthrough, so you don't leave anything behind.

Part 1

Drop in your video

1
Open the app and drag in any video file. Everything runs in your browser — your video never leaves your device. Give it a moment to read the frames.
Part 2

Key Frames — the moments that matter

1
Download the AI Contact Sheet — a single image with the key frames laid out in reading order, and a short note at the top telling the AI how to read it. The quickest way to give an AI a whole video in one picture.
2
Download the Key Frames + Summary — a text file listing every key frame, its exact timestamp, and why it was chosen (audio peak, scene change, high motion), plus the audio breakdown. This is the file an advisor or AI reads to understand the structure of the video.
3
Optionally, download the Individual Images (no ZIP) if you'd like each key frame as its own separate picture.
Part 3

The Mega Sheet & Batches

⚠ Don't skip this

The Mega Sheet is one of the most powerful parts of the app, and it sits further down the page where it's easy to scroll past. Make sure you download it.

1
Scroll down to the Mega Sheet download button and download it. The Mega Sheet is a single large image containing every frame of the video — not just the key frames — arranged in order, with the key frames marked. It gives the AI the complete, fine-grained detail of the whole clip.
2
There's also a Download All (ZIP) button. It bundles everything together, but it's large and uses a lot of storage, so it can be skipped if space is tight.
3
Download the AI Batches. Choose your maximum batch size — 20 MB is a good default, keeping each file under the upload limit — then generate. The app groups the frames into ordered ZIP files, each with a short README telling the AI which batch it is and how to reassemble them. Useful when a single upload would be too big.
Part 4

The Tunnel

1
Go to the Tunnel section of the app.
2
Click Capture Full Tunnel for AI. This builds the 3D tunnel view and downloads it when it finishes — the video as a journey through space, with the current moment in front and the past and future moments arranged in depth.
3
Download the Tunnel Contact Sheet — a single image showing a series of these tunnel views across the whole video, so the AI can see how the scene unfolds moment by moment.
Part 5

Give it to your AI

1
Open your AI chat — Claude, ChatGPT, Gemini, any of them.
2
Click the attach icon, then choose Add files or photos.
3
Add the files you downloaded — the AI Contact Sheet, the Key Frames + Summary, the Tunnel Contact Sheet, and the batches as needed. Depending on the file sizes, you may need to add them across more than one message.
4
Press enter. The AI now experiences the video — it can describe the motion, the detail, and what was happening, as if it had watched it.
Advanced

The Ribbon — connect your own AI API key

The Ribbon is available now as a working demo — extract a video, open the Ribbon tab, and watch how an AI engages with it across seven streams of attention. A further tier is coming when you connect your own AI API key — the credential that lets the app talk to an AI service directly, so the Ribbon reads your video live instead of showing the illustrative demo.

1
Extract a video first, then open the Ribbon tab — your frames load in automatically. Tip: use the Ribbon on its own; running the Mega Sheet and Tunnel in the same session on a long video can use too much memory.
2
Point your cursor at any of the seven lines to make it the active lens, then scroll to zoom in — that line's frames and its own kind of reading rise up. Each stream is a different lens of attention: pattern, emotion, prediction, and more.

The Ribbon is shown as a working demonstration. The recursive depth is real and patent-protected; the perceptual readings and language shown are illustrative of what it produces once connected to live AI, which is the next phase. It is a preview of where Now We Can See is heading.

Using it with your AI

It works with the AI you already use

Now We Can See is model-agnostic — there's no training, no special integration, nothing to set up on the AI's side. You build the image here, then drop or paste it into your AI of choice — Claude, ChatGPT, Gemini, or any model that can read an image — and ask it about what it's looking at. The understanding happens in that moment of seeing, then it's gone; the model isn't changed or stored.

Your video never leaves your device.
Everything runs right in your browser. Nothing is uploaded to us — you choose every frame the AI sees, the depth, the moments, the view. What you share with your AI is entirely your choice.
FAQ

Quick answers

Do I need an account to try it?

No — there's no sign-up for the free tier. You can drop in a video and start straight away.

Does my video get uploaded anywhere?

No. It's 100% browser-based — your video stays on your device and is never uploaded to us.

Which AIs does it work with?

Any of them. It's model-agnostic, so it complements whatever AI you already use — Claude, ChatGPT, Gemini and others.

Do I have to train or fine-tune anything?

No. There's nothing to train and nothing stored. It simply shows the AI the moment — the understanding lives in the seeing.

What's the difference between Key Frames and Mega Sheet?

Smart Key Frames picks out the moments that matter. Mega Sheet shows every frame, so the AI watches the whole clip. Use Key Frames for the gist, Mega Sheet for the full motion.

Is the technology protected?

Yes — Now We Can See is a genuinely new way of representing time and space for a perceiving mind, protected by provisional patents filed with IP Australia.

Ready to let an AI see what you see?

Enter the app