Google CEO Sundar Pichai just announced major structural changes consolidating the tech giant’s AI teams and capabilities under DeepMind.
Google and Alphabet CEO Sundar Pichai has announced sweeping structural changes aimed at improving execution velocity across the tech giant. The reorganisation consolidates AI research and product teams in an effort to bring breakthrough technologies to market faster.
The centrepiece of the shakeup is the merger of AI-model building efforts at Google Research and DeepMind under the DeepMind umbrella. Pichai, in a blog post published on Thursday, announced that this consolidation will concentrate compute-intensive model training into a single organisation, establishing clear pipelines for product teams to access the latest large language models and other foundational AI technologies.
“This will simplify development by concentrating compute-intensive model building in one place and establishing single access points for PAs looking to take these models and build generative AI applications,” wrote Pichai.
While DeepMind absorbs model building, Google Research will pivot to focus solely on foundational and applied computer science breakthroughs in areas like quantum computing, core machine learning, and the societal impacts of new technologies.
The reorganisation also aims to improve responsible AI practices across Google. Ethical AI teams previously embedded in Google Research will join DeepMind to instil safety considerations earlier in the model development process. Additional trust and safety teams will move to centralised groups to standardise AI evaluation and “red team” testing protocols.
“We need to be the best in class at deploying accurate, trustworthy, and transparent AI products for users and customers. To help do this, we’re making changes to the way our Responsible AI teams work at crucial points in the development cycle,” stated Pichai.
Another major pillar is the formation of a new “Platforms & Devices” product area unifying the Android, Chrome OS, Search, Photos and other developer platforms with Google’s device engineering teams. The aim is to accelerate AI-driven platform innovations by tightly coupling hardware, software and machine learning under one organisation.
Additionally, computational photography and on-device AI experts from Google Research will be part of this new group. Rick Osterloh will oversee the entire Platforms & Devices area, with longtime Android chief Hiroshi Lockheimer transitioning to an advisory role before taking on new projects.
The overarching objective, according to Pichai’s memo, is to “simplify our structure and improve velocity and execution” when it comes to developing AI systems in a responsible way.
Source:indianexpress.com