SandboxAQ and Anthropic Partner to Bring Scientific AI Models to Claude Assistant

​SandboxAQ has partnered with Anthropic to integrate its scientific AI models into Claude, Anthropic’s AI assistant. This collaboration allows researchers to access advanced drug discovery and materials science tools through a conversational interface. The partnership aims to make scientific modeling more accessible by removing the need for specialized computing infrastructure or technical systems.

Key Highlights

• SandboxAQ partners with Anthropic to integrate scientific AI models into Claude assistant.
• Researchers can access drug discovery and materials science tools through a conversational interface.
• SandboxAQ’s large quantitative models simulate molecular dynamics and quantum chemistry calculations.

AI Integration for Scientific Research

SandboxAQ, chaired by former Google CEO Eric Schmidt, has raised over $950 million from investors. Its partnership with Anthropic enables scientists to use SandboxAQ’s large quantitative models (LQMs) directly within Claude. These LQMs are physics-grounded, meaning they are built using the rules of the physical world rather than just patterns in text.

The models can perform quantum chemistry calculations and simulate molecular dynamics and microkinetics. This helps researchers predict how molecules behave before conducting laboratory experiments. As a result, the technology can reduce the time, cost, and complexity involved in discovering new drugs or materials.

How SandboxAQ Models Work

SandboxAQ’s LQMs are trained using real-world laboratory data and scientific equations. This makes them valuable for scientists in pharmaceutical and industrial companies searching for new materials or products. The models allow users to simulate chemical reactions at the molecular level, providing insights that were previously difficult to obtain without extensive laboratory work.

According to Nadia Harhen, SandboxAQ’s general manager of AI simulation, this is the first time a frontier quantitative model has been integrated with a frontier large language model in a conversational format. Previously, researchers needed to set up and manage their own digital infrastructure to operate these models. Now, they can access the same capabilities through Claude using natural language.

Broader AI Developments

Meta recently unveiled its TRIBE v2 (Trimodal Brain Encoder) foundation model, which predicts how the human brain responds to various sights and sounds. This model aims to help neuroscientists conduct experiments more efficiently and understand brain activity better. These advancements highlight how AI is not only changing the workforce but also supporting scientific and medical research.

Recent announcements from Apple and Google also show continued progress in AI and technology. Apple has announced the dates for WWDC 2026, where updates like a new Siri and iOS 27 are expected. Google I/O 2026 begins today, offering livestreams and previews of upcoming features. Additionally, Telegram has introduced a feature allowing bots to interact with each other without human involvement.