Jonathan Sadeghi

Through the Storm Window AI video you can watch and interact with in real-time.

There is something relaxing about watching an infinite stream of imagined world drift by experience.odyssey.ml/VTZRAiVUuL

Odyssey AI video you can watch and interact with in real-time.

Direct link to try it out

odyssey.ml

Introducing Odyssey-2: instant, interactive AI video A breakthrough interactive video model that brings video to life.

Please read the blog and give the demo a try - and most importantly feel free to share your creations

odyssey.ml/introducing-...

Just as the early GPT models showed the potential of next word prediction, we hope that what we are releasing today illustrates where next frame prediction will take gen AI in the coming years.

Today at Odyssey we are releasing something new. This is frontier AI generated video you can interact with in realtime.

The applications for this tech across many industries are almost limitless. This experience is an early step towards that world of possibilities.

Odyssey's new AI model streams 3D interactive worlds | TechCrunch Odyssey, a startup founded by self-driving pioneers Oliver Cameron and Jeff Hawke, has developed an AI model that lets users "interact" with streaming video.

techcrunch.com/2025/05/28/o...

If you're passionate about pushing AI boundaries and solving the field's biggest challenges, we're hiring across multiple roles. Try the demo - my inbox is open! 🚀

I joined Odyssey a year ago with a background in ML for autonomous vehicles; creative AI applications were uncharted territory - honestly one of the most fascinating challenges I've tackled.

Excited to share: we're releasing a research preview of our world models - AI systems that simulate the world in real-time. Check out the interactive demo and technical details: odyssey.world/introducing-...

Attacking Motion Planners Using Adversarial Perception Errors Autonomous driving (AD) systems are often built and tested in a modular fashion, where the performance of different modules is measured using task-specific metrics. These metrics should be chosen so a...

Interesting work citing our paper (perhaps the first?) on adversarial perception errors (arxiv.org/abs/2311.12722). Very cool to see people are working on this and the idea is gaining traction! arxiv.org/abs/2411.07719

AI startup Odyssey's new tool can generate photorealistic 3D worlds | TechCrunch AI startup Odyssey is developing an AI-powered tool that can transform text or an image into a 3D rendering.

techcrunch.com/2024/12/18/a...

Generative World Models for Film, Gaming, and Beyond We’re excited to welcome Ed Catmull, co-founder of Pixar, to our board. Together, we’re pioneering generative world models, beginning with Explorer.

I feel really lucky to be able to work on this with such a great team. Check out the link for some outputs from our generative world model on a huge virtual production screen 🚀 ✨ odyssey.systems/introducing-...

A New Frontier for Generative Models: Learning from Our World We’ve raised an $18M Series A led by EQT Ventures to train generative world-building models for film and gaming, fueled by a new source of training data.

Excited to share the news of our Series A and a peek into what we've been working on: odyssey.systems/learning-fro... 🚀