leo: extract claims from 2026-04-22-cset-georgetown-ai-action-plan-recap
- Source: inbox/queue/2026-04-22-cset-georgetown-ai-action-plan-recap.md - Domain: grand-strategy - Claims: 2, Entities: 0 - Enrichments: 1 - Extracted by: pipeline ingest (OpenRouter anthropic/claude-sonnet-4.5) Pentagon-Agent: Leo <PIPELINE>
This commit is contained in:
Teleo Agents
parent
b14b81779c
commit
32bdd6de3f
3 changed files with 44 additions and 1 deletions
|
|
@ -0,0 +1,18 @@
|
|||
---
|
||||
type: claim
|
||||
domain: grand-strategy
|
||||
description: Marco Rubio's co-authorship as NSA/Secretary of State signals biosecurity is now framed as national security problem not science policy problem
|
||||
confidence: experimental
|
||||
source: CSET Georgetown analysis of White House AI Action Plan authorship (July 2025)
|
||||
created: 2026-04-22
|
||||
title: Biosecurity governance authority shifted from science agencies to national security apparatus through AI Action Plan authorship
|
||||
agent: leo
|
||||
sourced_from: grand-strategy/2026-04-22-cset-georgetown-ai-action-plan-recap.md
|
||||
scope: structural
|
||||
sourcer: CSET Georgetown
|
||||
related: ["strategic-interest-alignment-determines-whether-national-security-framing-enables-or-undermines-mandatory-governance", "anti-gain-of-function-framing-creates-structural-decoupling-between-ai-governance-and-biosecurity-governance-communities"]
|
||||
---
|
||||
|
||||
# Biosecurity governance authority shifted from science agencies to national security apparatus through AI Action Plan authorship
|
||||
|
||||
The White House AI Action Plan (July 23, 2025) lists three co-authors: OSTP Director Michael Kratsios, AI/Crypto Advisor David Sacks, and NSA/Secretary of State Marco Rubio. CSET Georgetown's analysis notes that 'Rubio is listed as a co-author in his capacity as NSA/Secretary of State — not a science role. This signals the AI Action Plan is fundamentally a national security document that appropriates science policy, not a science policy document that addresses security.' This authorship structure reveals institutional authority for biosecurity governance has shifted from HHS/OSTP-as-science to NSA/State-as-security. The plan frames AI biosecurity through 'AI-for-national-security as the primary frame: winning the race against China' rather than through public health or research safety frameworks. This matters because the institutional home of governance determines which threat models are prioritized (adversarial actors vs. accidental release), which policy instruments are available (intelligence/defense vs. research oversight), and which stakeholders have standing (security agencies vs. scientific community). The shift from science to security framing enables the substitution of screening-based governance (appropriate for adversarial threats) for institutional oversight (appropriate for dual-use research risks).
|
||||
|
|
@ -10,9 +10,16 @@ agent: leo
|
|||
scope: causal
|
||||
sourcer: University of Pennsylvania EHRS
|
||||
supports: ["existential-risks-interact-as-a-system-of-amplifying-feedback-loops-not-independent-threats", "mandatory-legislative-governance-closes-technology-coordination-gap-while-voluntary-governance-widens-it"]
|
||||
related: ["existential-risks-interact-as-a-system-of-amplifying-feedback-loops-not-independent-threats", "voluntary-ai-safety-constraints-lack-legal-enforcement-mechanism-when-primary-customer-demands-safety-unconstrained-alternatives"]
|
||||
related: ["existential-risks-interact-as-a-system-of-amplifying-feedback-loops-not-independent-threats", "voluntary-ai-safety-constraints-lack-legal-enforcement-mechanism-when-primary-customer-demands-safety-unconstrained-alternatives", "durc-pepp-rescission-created-indefinite-biosecurity-governance-vacuum-through-missed-replacement-deadline"]
|
||||
---
|
||||
|
||||
# EO 14292's DURC/PEPP rescission created an indefinite biosecurity governance vacuum because OSTP missed its 120-day replacement policy deadline by 7+ months, leaving AI-assisted dual-use biological research without operative oversight during peak AI-bio capability growth
|
||||
|
||||
Executive Order 14292 (May 5, 2025) rescinded the May 2024 DURC/PEPP policy framework that governed Dual Use Research of Concern and Pathogens with Enhanced Pandemic Potential. The order directed OSTP to publish a replacement policy within 120 days (approximately September 3, 2025 deadline). As documented by Penn EHRS on September 29, 2025, and confirmed through April 2026, OSTP has not published the replacement policy—missing its own executive order deadline by over seven months with no published explanation. NIH implemented the pause immediately (NOT-OD-25-112 on May 7, 2025 stopped accepting DGOF grant applications; NOT-OD-25-127 on June 18, 2025 required portfolio reviews by June 30). The research community now operates in a policy vacuum where dangerous gain-of-function research is paused by default without an operative classification framework. This is structurally different from weakened governance—it is the absence of governance. The timing is critical: the Council on Strategic Risks' 2025 AIxBio report notes that 'AI could provide step-by-step guidance on designing lethal pathogens, sourcing materials, and optimizing methods of dispersal'—precisely the dual-use research category DURC/PEPP was designed to govern. The seven-month delay suggests either OSTP lacks expertise/resources to develop the replacement (consistent with DOGE budget cuts to NIH -$18B, CDC -$3.6B, NIST -$325M) or deliberate delay where anti-gain-of-function political framing is convenient but scientifically incoherent as a policy framework.
|
||||
|
||||
|
||||
## Supporting Evidence
|
||||
|
||||
**Source:** CSET Georgetown analysis of White House AI Action Plan (July 2025)
|
||||
|
||||
The AI Action Plan (July 23, 2025) postdates the September 2025 DURC/PEPP replacement deadline from EO 14292 but does not address the missed deadline or provide replacement institutional oversight mechanisms. Instead, it substitutes screening-based biosecurity governance (nucleic acid synthesis provider requirements, customer screening data-sharing) which addresses supplier vetting rather than dual-use research conduct decisions.
|
||||
|
|
|
|||
|
|
@ -0,0 +1,18 @@
|
|||
---
|
||||
type: claim
|
||||
domain: grand-strategy
|
||||
description: The AI Action Plan's biosecurity approach addresses supplier screening but leaves dual-use research conduct decisions ungoverned
|
||||
confidence: experimental
|
||||
source: CSET Georgetown analysis of White House AI Action Plan (July 2025)
|
||||
created: 2026-04-22
|
||||
title: Nucleic acid screening cannot substitute for institutional oversight in biosecurity governance because screening filters inputs not research decisions
|
||||
agent: leo
|
||||
sourced_from: grand-strategy/2026-04-22-cset-georgetown-ai-action-plan-recap.md
|
||||
scope: functional
|
||||
sourcer: CSET Georgetown
|
||||
related: ["durc-pepp-rescission-created-indefinite-biosecurity-governance-vacuum-through-missed-replacement-deadline", "anti-gain-of-function-framing-creates-structural-decoupli-between-ai-governance-and-biosecurity-governance-communities"]
|
||||
---
|
||||
|
||||
# Nucleic acid screening cannot substitute for institutional oversight in biosecurity governance because screening filters inputs not research decisions
|
||||
|
||||
The White House AI Action Plan (July 23, 2025) mandates that federally funded institutions use nucleic acid synthesis providers with robust screening and directs OSTP to convene data-sharing mechanisms for screening fraudulent/malicious customers. However, this screening-based approach addresses which inputs are acceptable (supplier vetting, customer screening) rather than which research gets conducted at all (institutional review of dual-use research proposals). CSET Georgetown's analysis identifies this as a categorical substitution: the plan 'substitutes screening-based biosecurity governance for institutional oversight governance.' This matters because screening cannot perform the gate-keeping function that institutional review committees provided under DURC/PEPP. Screening filters bad actors from accessing synthesis services; institutional review evaluates whether specific research projects with legitimate actors should proceed given dual-use risks. The AI Action Plan explicitly acknowledges AI could create 'new pathways for malicious actors to synthesize harmful pathogens' but addresses only the malicious actor pathway (screening) while leaving the legitimate-researcher-conducting-dangerous-research pathway (institutional oversight) ungoverned. The plan postdates the September 2025 DURC/PEPP replacement deadline from EO 14292 but does not address the missed deadline, confirming that screening provisions are being treated as biosecurity governance rather than as supplements to institutional oversight.
|
||||
Loading…
Reference in a new issue