Introducing SecureBio’s ‘Trends in Biology’ AI Benchmarks Dashboard

A comprehensive view of biology capabilities of AI models

SecureBio today releases a public dashboard with all of its AI model evaluation scores. This dashboard shows how frontier AI models perform on biosecurity-relevant tasks, ranging from our flagship tacit knowledge benchmark, Virology Capabilities Test, to our next-generation agentic benchmarks like ABC-Bench.

This dashboard is the most complete display of biosecurity-relevant AI capabilities, spanning more than three years, nine model companies, and over a dozen evaluation metrics. Two results stand out: frontier models have surpassed human-expert baselines, and capabilities continue to rise in biology and biosecurity-relevant domains.

We will not only monitor evaluation improvements over time, but also be able to detect when evaluations saturate.

The dashboard also includes a “Bio Capabilities Index,” an aggregate score underpinned by the same methodology as the Epoch Capabilities Index (ECI).

In addition to capabilities, we also measure and report the effectiveness of biosecurity-related safeguards, such as BioTIER.

All scores are computed using a standardized pipeline to ensure fair and valid comparisons. For example, we make sure evaluations are run using identical configurations.

The dashboard is live and updated as new models, benchmarks, and analyses are released. Check it out here.