SandboxAQ puts drug discovery AI on Claude, no coding PhD needed
▼ Summary
– Drug discovery is slow and costly, often taking a decade and billions of dollars, with most candidates failing.
– SandboxAQ partnered with Anthropic to put its scientific AI models into Claude, creating a conversational interface that requires no specialized computing.
– SandboxAQ, an Alphabet spinout with over $950 million in funding, produces Large Quantitative Models (LQMs) grounded in physics.
– LQMs simulate molecular behavior using real-world lab data and scientific equations, predicting how candidate molecules act before lab testing.
– SandboxAQ focuses on making advanced AI accessible to computational scientists and experimentalists at large industrial companies.
Drug discovery remains one of the most costly and failure-prone endeavors in modern industry. It can take a full decade and billions of dollars to find a single viable drug candidate, and most never reach patients. A wave of AI startups has promised to accelerate this process, but their tools have mostly benefited researchers who already possess the technical expertise to operate them.
SandboxAQ believes the real problem isn’t the science behind the models , it’s the user interface.
The company has partnered with Anthropic to embed its scientific AI models directly into Claude. This integration puts powerful drug discovery and materials science capabilities behind a conversational interface, eliminating the need for specialized computing infrastructure.
SandboxAQ was launched roughly five years ago as a spinout from Alphabet, with former Google CEO Eric Schmidt serving as its chairman. The company has raised over $950 million from investors and operates multiple business lines, including cybersecurity.
What truly sets SandboxAQ apart, however, is its development of large quantitative models (LQMs). These proprietary models are “physics-grounded,” meaning they rely on the immutable rules of the physical world rather than patterns in text. They can perform quantum chemistry calculations and simulate molecular dynamics and microkinetics , the step-by-step study of chemical reactions at the molecular level. This capability allows researchers to predict how candidate molecules will behave long before any lab work begins.
“Trained on real-world lab data and scientific equations, LQMs are AI models engineered for the quantitative economy, a $50+ trillion sector spanning biopharma, financial services, energy, and advanced materials,” the company stated in a press release. This makes clear that SandboxAQ is not building another chatbot or code assistant. It is targeting the massive economic sectors that AI is supposed to revolutionize.
While competitors like Chai Discovery and Isomorphic Labs have poured funding into improving model accuracy, SandboxAQ is focusing on accessibility.
“For the first time, we have a frontier [quantitative] model on a frontier LLM that someone can access in natural language,” said Nadia Harhen, SandboxAQ’s general manager of AI simulation, in an interview with TechCrunch. Previously, users had to provide their own digital infrastructure to run the LQMs.
SandboxAQ’s typical customers are computational scientists, research scientists, and experimentalists working at large pharmaceutical or industrial companies. Their goal is to discover new materials that can become marketable products.
“Our customers come to us because they’ve tried all the other software out there, and the complexity of their problem is such that it didn’t work or didn’t yield positive results for them when that translation went to take place in the real world,” Harhen added.
(Source: TechCrunch)
