Scenario PlanningBioR · Health Security
All scenarios

Biotechnology & Biosafety

AI–Bio Convergence — Governance Under a Capability Shift

Advances at the AI–biology interface outpace the oversight meant to govern them.

High severity

Duration

120 min

Injects

5

Audience

National biosecurity policy, research funders & regulators, AI governance, ethics

Situation

A step-change in AI tools relevant to the life sciences prompts national concern that existing biosecurity oversight — designed for a slower, more concentrated research landscape — no longer matches the risk or pace of the field. No incident has occurred. This exercise is deliberately anticipatory: it tests whether governance, screening and coordination can adapt before, rather than after, a capability shift is exploited. The focus throughout is policy, oversight and coordination — not technical detail.

Exercise objectives

  • Stress-test whether existing biosecurity governance keeps pace with a fast-moving enabling technology.
  • Exercise coordination across three communities that rarely share a table: biosecurity, AI governance and research funding.
  • Examine practical safeguards — nucleic-acid synthesis screening, access controls, know-your-customer, publication norms — and who owns them.
  • Rehearse a proportionate response that manages risk without foreclosing beneficial science.

Capability stress

How hard this scenario tests each of the 10 benchmark dimensions (1–5).

Foresight lens

Readinessprecedent-free

Being able to cope with a threat that is precedent-free — a pathogen, dynamic or context we have NOT seen, where experience can actively mislead. Readiness is an adaptive capacity, not a plan for a known pattern.

TUNA profile

Turbulence
Strong
Uncertainty
Dominant
Novelty
Dominant
Ambiguity
Dominant

Assumptions this scenario windtunnelsfull register →

A2The next serious pathogen will resemble what we have seen (respiratory).

Sensitivity: high · TUNA: N

A6Existing biosecurity/biosafety oversight matches the pace of the science.

Sensitivity: high · TUNA: U · A · N

Scenario parameters

Illustrative planning figures for discussion — not operational data.

Trigger

Anticipated capability shift at the AI–bio interface (no incident)

Posture

Anticipatory governance, not incident response

Oversight maturity

Designed for a slower, more concentrated field

Safeguards in scope

Synthesis screening, access controls, publication norms

Stakeholders

Biosecurity, AI governance, funders, industry, academia

Time pressure

Policy cycles in years; technology in months

Roles at the table

National Biosecurity Policy Lead

Owns the biological-risk governance response.

AI Governance Lead

Owns oversight of the enabling technology.

Research Funder / Regulator

Owns the conditions attached to funding and approvals.

Synthesis-Industry Liaison

Represents providers implementing screening and access controls.

Ethics & Openness Adviser

Balances beneficial science, openness and misuse risk.

Inject timeline

  1. T+0Capability concern raised

    An expert body warns that oversight may be lagging the field.

  2. T+2wScreening gap

    A review finds nucleic-acid synthesis screening is voluntary and incomplete.

  3. T+4wJurisdiction gap

    The fast-moving capability falls between biosecurity and AI mandates; neither clearly owns it.

  4. T+6wPublic attention

    Media coverage frames it as a governance failure-in-waiting; pressure to act rises.

  5. T+8wProposal on the table

    A draft safeguards package is proposed; industry and academia raise feasibility and openness concerns.

Decision points

D1Where should authority sit for a risk that straddles biosecurity and AI governance — and who convenes it?

  • The cost of a risk that falls between two mandates and is owned by neither.
  • Whether a standing coordination body exists or must be created.
  • How to give one owner authority without stalling in inter-agency negotiation.

D2Which safeguards are proportionate now, given they impose real costs on legitimate science?

  • Risk reduction per unit of burden on beneficial research.
  • Which safeguards are voluntary today and could be made standard.
  • Feasibility for small labs and academic groups, not just large industry.

D3How do you strengthen oversight without driving research to less-governed jurisdictions?

  • The displacement risk of unilateral national controls.
  • The value of international alignment and shared screening standards.
  • Keeping the beneficial mainstream of the science open and productive.

Response playbook

Assess

  • Map the governance gap
  • Convene the three communities
  • Inventory existing safeguards
  • Define the risk in plain terms

Coordinate

  • Assign clear ownership
  • Align biosecurity and AI mandates
  • Engage industry and academia early
  • Agree decision rights

Safeguard

  • Strengthen synthesis screening and access controls
  • Update funding conditions
  • Revisit publication and responsible-disclosure norms
  • Support smaller labs to comply

Sustain

  • Build an adaptive review cadence tied to capability signals
  • Pursue international alignment
  • Monitor for displacement effects
  • Protect beneficial research

After-action questions

  • Did any single body actually own this cross-cutting risk, or did it fall between mandates?
  • Which safeguard gives the most risk reduction for the least burden on legitimate science?
  • What signal would tell you oversight has fallen dangerously behind — and are you watching it?

National benchmark references

Real national strategies from the Global Pandemic Preparedness Benchmark that inform this scenario.