Information Hazards in Capability Research
Capability research can create knowledge that helps science and also lowers barriers to misuse. The practical task is responsible research communication, not broad policy writing.
- Identify information hazards in capability-focused AI-biology work.
- Use publication review without weakening reproducibility.
- Separate scientific detail needed for review from detail that increases misuse risk.
Responsible capability research keeps enough detail for verification while avoiding unnecessary operational detail that raises misuse risk. The standard is not secrecy by default. The standard is deliberate disclosure.
Introduction
AI-biology work now includes protein design, genome modeling, lab automation, and agentic workflows. ARPA-H IGoR’s inclusion of agentic systems and laboratory automation shows why researcher practice needs explicit review gates (ARPA-H IGoR, 2026).
Demonstrated
Demonstrated capability includes models that design proteins, model biomolecular interactions, generate sequences, and help plan experiments in bounded settings. RFdiffusion, AlphaFold 3, Evo, and self-driving laboratory systems supply concrete examples of capability progress (Watson et al., 2023; Abramson et al., 2024; Nguyen et al., 2024; Abolhasani and Kumacheva, 2023).
| Evidence Anchor | What It Supports | Practical Constraint |
|---|---|---|
| RFdiffusion and Evo | Capability gains in design and sequence modeling | Publication detail should match legitimate verification needs |
| ARPA-H IGoR | Agentic and automated research infrastructure | Human authorization and protocol standards matter |
| Bridge2AI | Ethical AI-ready data framing | Data and model release need governance |
Theoretical
Theoretical capability includes publication norms that preserve reproducibility while reducing operational misuse. This requires journal policies, institutional review, and field-specific norms rather than ad hoc redaction.
Beyond Current Capabilities
Beyond current capabilities includes perfect separation of beneficial and harmful uses at publication time. Dual-use judgment remains contextual and imperfect.
Practice Notes
- Perform a disclosure review before releasing code, model weights, protocols, or datasets that materially increase biological capability.
- Keep reproducibility artifacts available to trusted reviewers when public release is not appropriate.
- Avoid step-by-step operational details when high-level scientific reporting is sufficient.
- Document the reason for any redaction or staged release.