Ari solves a Rubik's cube while Griffin coaches him from what it sees in his hands, and when he goes quiet to think, it waits. Its perception is always on, so it acts without being asked. Note: this actually blew our minds seeing happen live.

Mid-story, Mars holds up a peacock, and Griffin seamlessly writes it into the plot. It sees while it speaks, so what it sees can change what it's saying. Show it something and it's part of the conversation.

While Mars is still telling Griffin about a boring date, it laughs, nods and asks questions instead of waiting for her to stop. That's full duplex: listening, watching and talking at the same time, the way people actually talk.

Watch Brittney and Mars try to trip Griffin up at Simon Says. To play, it has to listen for "Simon says," watch their hands and move its own, all at the same time. It generates the whole frame, not just a face.

When machines meet us where we are and truly understand us, we unlock a future where using a machine will be as easy as talking to a friend or coworker. A tutor for every student that notices when they’re lost, an elderly care companion that listens, or the perfect assistant.

Introducing Griffin, the first model to pass the video Turing test. 48% of people who talked to it live thought it was a real human. Previous systems have had a pass rate <3%. It is #1 on NVIDIA's benchmark for full-duplex AI video. It’s the first Human Interaction Model (HIM).

See how Sagar keeps his eyes on the soldering iron while Griffin talks him through the job and counts the seconds for him. Time is part of the conversation too, not just words. That's what a good coach does without being asked.

Most AI can tell you it's happy for you. Griffin shows it, in its voice, its face and its gestures, all at once. Rudy just got promoted, and you don't need the sound on to know how Griffin feels about it.

Because it’s the first model that can be mistaken for a real person, Griffin requires safety considerations before it can be released publicly. The full research preview covers how it works, how we tested it and the safety measures that have to come first:

cool, cant wait to see how scammers use this tech to rip off old people because what other use case is there for this really

I see exactly where this is going.

the banality of evil. the resources need to sustain things like this will result in climate catastrophies everywhere. those involved have a real ethical responsibility not to prioritize "cool toys" over the welfare of billions of people

Zooming in its really clear you can see she is ai, it even fades to another face at some point

This sucks so much. All this money spent on recreating humans, when there are…humans…ridiculous that thinking about what technology should be is limited. Solve some real problems you dorks

Finally I can solve the Rubik Cube!!!

Why do they always sound like Hermione. It's always the Hermione voice.

Uncanny valley

That’s cool, and very impressive, but its responses sound exactly like an LLM.

If you fall for this you are a troglodyte. The second you full screen the video, the illusion disappears.

Amazing. It's too focussed to be human in this task. Tunnel vision.

@Bitcoin_Cookie Bloody hell! 🤯😳

You people are demons

已经想象到超级多细分场景

Es increíble lo del cubo.

Why not just one-shot the last layer?

Why she have English accent 😂

Insanely easy to see this is ai, look at the neck

Narrator: Ari is a loser.

What ????

Did she go to Cornell?

This is pretty gnarly

Love how this has basically no actual use beyond being used by third worlds to scam people