Google is streamlining its efforts in building artificial intelligence (AI) models by consolidating the teams from Google Research and Google DeepMind. Sundar Pichai, the CEO of Google and Alphabet, announced in a note to employees that all AI model development will now be centralized within Google DeepMind. This move aims to enhance the company's ability to provide capable AI solutions for users, partners, and customers. The consolidation will simplify development by focusing compute-intensive model building in one location. It will also establish a single access point for those interested in utilizing these models to build generative AI applications. Additionally, the restructuring grants clear and distinct objectives for Google Research, including investments in computing systems, foundational machine learning and algorithms, and applied science and society. This decision follows the creation of Google DeepMind, which brought together researchers from Google Brain, DeepMind, and other AI-focused teams last year.

This unified group was responsible for developing Google's Gemini models. The note also outlines changes to Google's Responsible AI teams. Researchers from the Research department have been integrated into DeepMind, while other responsibility teams have been moved to the central Trust and Safety team. The company will also increase its investment in testing AI-powered features for vulnerabilities. Pichai emphasized that these changes are part of ongoing efforts to simplify the company's structure, improve velocity and execution, and enhance customer experience. Previously, Google merged the Brain team in Google Research with teams in DeepMind, unified its ML infrastructure and ML developer teams, and brought its Search teams under a single leader. Lastly, recent reports suggest that Google's spending on AI will exceed $100 billion.