Drug discovery is one of the most expensive tasks in modern industry. Finding a single viable molecule can take 10 years and cost billions of dollars, and most candidates still don’t succeed. A generation of AI startups promised to solve this problem. Most have already made the problem less painful for researchers who are technologically sophisticated enough to use the tools.
However, SandboxAQ believes that the bottleneck is not the model. It’s an interface.
The company worked with Anthropic to integrate scientific AI models directly into Claude, adding powerful drug discovery and materials science tools behind an interactive interface that requires no special computing infrastructure.
SandboxAQ, founded as an Alphabet spinoff about five years ago, counts former Google CEO Eric Schmidt as its chairman. The company, which has raised more than $950 million from investors, has built a variety of business lines, including a cybersecurity business.
But one of the more unique things SandboxAQ does is create large-scale quantitative models (LQMs). These proprietary models are “based in physics.” This means that it is built based on rules from the real world rather than patterns in text. You can run quantum chemical calculations and simulate molecular dynamics and microdynamics (the study of how chemical reactions unfold at the molecular level). This is important because it tells researchers how a candidate molecule is likely to behave before anyone sets foot in the lab.
“Trained on real lab data and scientific equations, LQM is an AI model designed for the quantitative economy, a $50 trillion-plus sector spanning biopharmaceuticals, financial services, energy, and advanced materials,” the company said in a press release, strongly suggesting that Sandbox AQ is not building yet another chatbot or code assistant, but rather going after the economy that AI is supposed to transform.
Chai Discovery and Isomorphic Labs, which invested a lot of money to develop better models, focused on the science. SandboxAQ focuses on who can actually use it.
“For the first time, we have a frontier (quantitative) model for a frontier LLM that anyone can access in natural language,” Nadia Harhen, general manager of AI simulation at SandboxAQ, told TechCrunch. Previously, SandboxAQ’s LQM users had to provide their own digital infrastructure to run their models.
SandboxAQ’s customers are primarily computational scientists, research scientists, or experimenters. Typically, they work for large pharmaceutical or industrial companies and are looking for new materials that can become marketable products.
“Our customers come to us because they have tried all the other software, but the problem is so complex that when the translation is run in the real world, it doesn’t work or doesn’t produce positive results,” Harhen said.
If you purchase through links in our articles, we may receive a small commission. This does not affect our editorial independence.
