Google Cloud will offer specialist AI models from SandboxAQ, expanding enterprise and research access to technology designed to accelerate drug discovery, materials science, and semiconductor manufacturing. The partnership adds SandboxAQ's large quantitative models to Google's cloud marketplace, enabling researchers to combine Google's Gemini language models with AI systems trained on scientific equations and laboratory data. Unlike large language models that generate text, these quantitative models are trained on numerical data to solve problems in chemistry, biology, and physics. SandboxAQ, an Alphabet spinout, joins a growing field of companies using AI for thorny research questions, including Isomorphic Labs and OpenAI.
The first available model, AQCat, helps researchers identify promising catalyst and materials candidates for applications from semiconductor manufacturing to battery development. A second model, AQPotency, focuses on drug discovery by identifying molecules likely to bind to disease targets. SandboxAQ CEO Jack Hidary emphasized the technology saves years of research time while exploring a larger candidate space, enabling novel solutions. The partnership follows SandboxAQ's $500 million award through the US CHIPS program to develop AI models for semiconductor manufacturing, including new battery chemistries and methods for breaking down PFAS.
SandboxAQ originally started as a quantum tech lab in Alphabet's X division and spun out as an independent company in 2022. Its backers include former Google CEO Eric Schmidt and Salesforce CEO Marc Benioff. The company relies on quantum techniques embedded in its AI models, which currently run on conventional accelerated computing hardware like GPUs and TPUs, with future quantum computers expected to expand capabilities. Hidary noted biopharmaceutical research represents the largest near-term opportunity, with customers studying cancers including glioblastoma and prostate cancer, as well as Alzheimer's, Parkinson's, and cardiovascular conditions.
What to watch next: Whether SandboxAQ's models achieve breakthrough results in drug discovery and materials science that justify the investment, and how quickly quantum computing matures to further amplify their capabilities.
Key Takeaways
- Google Cloud is integrating specialized quantitative AI models for scientific research, moving beyond general-purpose language models.
- SandboxAQ's AQCat and AQPotency models target semiconductor manufacturing, battery development, and drug discovery.
- The partnership builds on SandboxAQ's $500 million US CHIPS program award for AI-driven semiconductor innovation.
- Biopharmaceutical research, including cancer and neurodegenerative disease studies, is the largest near-term opportunity for these models.
Insights & Analysis
- This move signals a strategic shift for cloud providers to offer domain-specific AI tools, potentially creating a new revenue stream beyond general-purpose AI services.
- The integration of quantum techniques into conventional hardware suggests a hybrid approach that could accelerate scientific discovery before full-scale quantum computers are viable.