DeepMind CEO Demis Hassabis says Google will eventually combine its Gemini and Veo AI models

In the last view PossibleLinkedin co-founder Reid Hoffman, Google Deepmind CEO Demis Hassabisin, Google has finally planned to combine it Twins With AI models I see Video-producing models to improve the concept of the previous physical world.

“We are always in the basic model to be multimodal from the beginning,” Hassabis said: ” [is because] We have a vision for a universal digital assistant, auxiliary assistant … In fact it helps you in the real world. “

The AI ​​industry gradually moves to the models of “Omni” models, which can understand and synthesize the forms of many media. Google’s Newest Gemini models sound Like pictures and text, Openai’s standard model in Chatgpt can create pictures locally – of course Studio Ghibli Style Arts. Amazon has Also announced their plans Starting a “any” model at the end of this year.

It is a lot of training information, pictures, videos, voice, text, etc. in these Omni models. Video data for Hassabis, VEO is mainly a platform that Google has a platform from YouTube.

“Basically watching youtube videos – many YouTube videos – [Veo 2] You can understand, you know, the physics of the world, “Hassabis said.

Google, in the previously TechCrunch, “Some” can do “some” “some” models “in accordance with YouTube creators with YouTube content,” he said. Google has expanded the terms of service Last year allows the company to provide more information to bring up AI models.