Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Apparently not satisfied with its take on this world, Google is on trial the staff his DeepMind research laboratory to build generative models that are able to simulate the physical world. The project — which will be led by Tim Brooks, one of the principals who helped build OpenAI’s video generator, Sora — will be a critical part of the company’s attempt to achieve general artificial intelligence, he said. job listings in relation to the new team.
Brooks, who joined DeepMind later escape from OpenAI in October, and his team has “ambitious plans to make massive generative models that simulate the world.” According to the role descriptions, the effort to build world models “will fuel numerous domains, such as visual reasoning and simulation, planning for embodied agents, and real-time interactive entertainment.” If you are willing to take one of these roles, maybe you can understand what these vagues mean and come back to us.
A model of the worldput as simply as possible, it typically seeks to simulate how the world actually works. Generative models like Sora are able to replicate things they’ve seen before in their training data, they have no real understanding of why this thing happens. So he can successfully generate a video of a person throwing a baseball, but he does not understand the physics of what is happening. The world’s models aim to arm the machine with enough information to really analyze how an action takes place and the likely outcome.
Meta AI scientist Yann LeCun described world patterns in this way during a speech at the Hudson Forum earlier this year: “A world model is your mental model of how the world behaves… You can imagine a sequence of actions that you can take, and your world model will allow you to predict what the effect of the action sequence on the world”.
World models are difficult to build for a number of reasons, including the massive amount of computation required to run a model and the lack of sufficient training data to create an accurate model, resulting in most world models that they work only for limited and specific contexts.
The DeepMind team seems intent on taking the global model wider. The plan is to build “real-time interactive generation” tools on top of the models and potentially look at how they could integrate their world model into Google’s big Gemini language model.
One likely area that DeepMind will try to tackle is video games. The job description for the new team notes that they will collaborate with the Veo and Genie teams at Google. genius is Google’s Sora video generator and Genie is an existing world model that can simulate 3D environments in real time. The video game industry already is eager to adopt AI toolsdisplacing thousands of workers. A Economy CVL The survey found that more than 86% of all gaming companies have already adopted generative AI tools and nearly 15% of all gaming jobs could be disrupted by 2026.
Perhaps improving this world would be a better use of time than the model.
