SandboxAQ brings its drug discovery models to Claude — no PhD in computing required

Drug discovery is one of the most expensive pursuits in modern industry. Finding a single viable molecule can take ten years and cost billions, and most candidates still don’t make the cut. A generation of AI startups have promised to solve this – most of them have made the problem less painful for researchers, who are already technically advanced enough to use the tools.

But SandboxAQ thinks the bottleneck isn’t the models. It’s the interface.

The company is working with Anthropic to integrate its scientific AI models directly into Claude, putting powerful drug discovery and materials science tools behind a conversational interface that doesn’t require specialized computing infrastructure.

SandboxAQ was founded about five years ago as a spinout from Alphabet and has Eric Schmidt, the former CEO of Google, as chairman. The company, which has increased more than $950 million from investorshas developed a number of different business lines, including a cybersecurity company.

However, one of the more unique things SandboxAQ does is produce large quantitative models, or LQMs. These proprietary models are “physically grounded,” meaning they are built on the rules of the physical world rather than patterns in the text. They can perform quantum chemical calculations and simulate both molecular dynamics and microkinetics, the study of how chemical reactions unfold at the molecular level. That’s important because it tells researchers how candidate molecules are likely to behave before anyone sets foot in a lab.

“Trained on laboratory data and real-world scientific equations, LQMs are AI models designed for the quantitative economy, a more than $50 trillion industry spanning biopharmaceuticals, financial services, energy and advanced materials,” the company said in a press release that strongly suggests that Sandbox AQ isn’t building another chatbot or code assistant — it’s chasing the economy that AI is supposed to transform.

Chai Discovery and Isomorphic Labs – both well-funded bets on better models – have focused on the science. SandboxAQ is aimed at those who can actually use it.

“For the first time we have a border [quantitative] model on a frontier LLM that someone can access in natural language,” Nadia Harhen, general manager of AI simulation at SandboxAQ, told TechCrunch. Previously, users of SandboxAQ’s LQMs had to provide their own digital infrastructure to run the models.

SandboxAQ’s customers are usually computational scientists, research scientists, or experimentalists. Generally, these people work at large pharmaceutical or industrial companies and are looking for new materials that can become marketable products.

“Our customers come to us because they’ve tried all the other software, and the complexity of their problem is such that it didn’t work or didn’t produce positive results for them when that translation took place in the real world,” Harhen said.