AI models are increasingly showing multimodal capabilities, suggesting they might develop more complex experiences in the future.

The convergence of reasoning models and multimodal models will lead to a new kind of computer interface that can reason, generate real-time video, and provide personalized experiences.

As AI models improve, the cost of video generation is expected to decrease significantly, allowing for more accessible and efficient production of high-quality content.

Recent advancements in video generation models have drastically reduced the time needed to simulate complex scenes, demonstrating the power of data-driven processes.

The models are getting smarter, faster, cheaper, and safer with each iteration. The AI models you're using today are the worst you'll ever use, and the future will be very different from today.

We're currently experiencing a Cambrian explosion in AI, with advancements in processing text, pixels, videos, and audio, leading to new applications and models.

We're in a period of high uncertainty regarding the evolution of AI model capabilities, particularly in multimodality, which will continue to improve.

The foundation models today represent the world in text, and we are expanding to include images and video to create a more comprehensive understanding of our environment.

The future will likely see more researchers embedded in every product team, as fine-tuning models for specific use cases becomes a core workflow for building products.