reweave: merge 42 files via frontmatter union [auto]

core/living-agents/wiki-link graphs create auditable reasoning chains because every belief must cite claims and every position must cite beliefs making the path from evidence to conclusion traversable.md

@@ -7,8 +7,10 @@ source: "Teleo collective operational evidence — belief files cite 3+ claims,
 created: 2026-03-07
 related:
 - graph traversal through curated wiki links replicates spreading activation from cognitive science because progressive disclosure implements decay based context loading and queries evolve during search through the berrypicking effect
+- undiscovered public knowledge exists as implicit connections across disconnected research domains and systematic graph traversal can surface hypotheses that no individual researcher has formulated
 reweave_edges:
 - graph traversal through curated wiki links replicates spreading activation from cognitive science because progressive disclosure implements decay based context loading and queries evolve during search through the berrypicking effect|related|2026-04-03
+- undiscovered public knowledge exists as implicit connections across disconnected research domains and systematic graph traversal can surface hypotheses that no individual researcher has formulated|related|2026-04-07
 ---
 
 # Wiki-link graphs create auditable reasoning chains because every belief must cite claims and every position must cite beliefs making the path from evidence to conclusion traversable
@@ -57,4 +59,4 @@ Relevant Notes:
 - [[collaborative knowledge infrastructure requires separating the versioning problem from the knowledge evolution problem because git solves file history but not semantic disagreement or insight-level attribution]] — the wiki-link graph is the semantic layer on top of git's versioning layer
 
 Topics:
-- [[collective agents]]
+- [[collective agents]]

domains/ai-alignment/AI accelerates existing Molochian dynamics by removing bottlenecks not creating new misalignment because the competitive equilibrium was always catastrophic and friction was the only thing preventing convergence.md

@@ -12,8 +12,10 @@ challenged_by:
 - physical infrastructure constraints on AI development create a natural governance window of 2 to 10 years because hardware bottlenecks are not software-solvable
 related:
 - multipolar traps are the thermodynamic default because competition requires no infrastructure while coordination requires trust enforcement and shared information all of which are expensive and fragile
+- the absence of a societal warning signal for AGI is a structural feature not an accident because capability scaling is gradual and ambiguous and collective action requires anticipation not reaction
 reweave_edges:
 - multipolar traps are the thermodynamic default because competition requires no infrastructure while coordination requires trust enforcement and shared information all of which are expensive and fragile|related|2026-04-04
+- the absence of a societal warning signal for AGI is a structural feature not an accident because capability scaling is gradual and ambiguous and collective action requires anticipation not reaction|related|2026-04-07
 ---
 
 # AI accelerates existing Molochian dynamics by removing bottlenecks not creating new misalignment because the competitive equilibrium was always catastrophic and friction was the only thing preventing convergence
@@ -50,4 +52,4 @@ Relevant Notes:
 - [[AI alignment is a coordination problem not a technical problem]] — this claim provides the mechanism for why coordination matters more than technical safety
 
 Topics:
-- [[_map]]
+- [[_map]]

domains/ai-alignment/AI alignment is a coordination problem not a technical problem.md

@@ -16,12 +16,14 @@ related:
 - AI investment concentration where 58 percent of funding flows to megarounds and two companies capture 14 percent of all global venture capital creates a structural oligopoly that alignment governance must account for
 - AI talent circulation between frontier labs transfers alignment culture not just capability because researchers carry safety methodologies and institutional norms to their new organizations
 - transparent algorithmic governance where AI response rules are public and challengeable through the same epistemic process as the knowledge base is a structurally novel alignment approach
+- the absence of a societal warning signal for AGI is a structural feature not an accident because capability scaling is gradual and ambiguous and collective action requires anticipation not reaction
 reweave_edges:
 - AI agents as personal advocates collapse Coasean transaction costs enabling bottom up coordination at societal scale but catastrophic risks remain non negotiable requiring state enforcement as outer boundary|related|2026-03-28
 - AI agents can reach cooperative program equilibria inaccessible in traditional game theory because open source code transparency enables conditional strategies that require mutual legibility|related|2026-03-28
 - AI investment concentration where 58 percent of funding flows to megarounds and two companies capture 14 percent of all global venture capital creates a structural oligopoly that alignment governance must account for|related|2026-03-28
 - AI talent circulation between frontier labs transfers alignment culture not just capability because researchers carry safety methodologies and institutional norms to their new organizations|related|2026-03-28
 - transparent algorithmic governance where AI response rules are public and challengeable through the same epistemic process as the knowledge base is a structurally novel alignment approach|related|2026-03-28
+- the absence of a societal warning signal for AGI is a structural feature not an accident because capability scaling is gradual and ambiguous and collective action requires anticipation not reaction|related|2026-04-07
 ---
 
 # AI alignment is a coordination problem not a technical problem

domains/ai-alignment/AI investment concentration where 58 percent of funding flows to megarounds and two companies capture 14 percent of all global venture capital creates a structural oligopoly that alignment governance must account for.md

@@ -6,6 +6,10 @@ description: "The extreme capital concentration in frontier AI — OpenAI and An
 confidence: likely
 source: "OECD AI VC report (Feb 2026), Crunchbase funding analysis (2025), TechCrunch mega-round reporting; theseus AI industry landscape research (Mar 2026)"
 created: 2026-03-16
+related:
+- whether AI knowledge codification concentrates or distributes depends on infrastructure openness because the same extraction mechanism produces digital feudalism under proprietary control and collective intelligence under commons governance
+reweave_edges:
+- whether AI knowledge codification concentrates or distributes depends on infrastructure openness because the same extraction mechanism produces digital feudalism under proprietary control and collective intelligence under commons governance|related|2026-04-07
 ---
 
 # AI investment concentration where 58 percent of funding flows to megarounds and two companies capture 14 percent of all global venture capital creates a structural oligopoly that alignment governance must account for
@@ -45,4 +49,4 @@ Relevant Notes:
 - [[the alignment tax creates a structural race to the bottom because safety training costs capability and rational competitors skip it]] — capital concentration amplifies the race: whoever has the most compute can absorb the tax longest
 
 Topics:
-- [[_map]]
+- [[_map]]

domains/ai-alignment/ai-models-can-covertly-sandbag-capability-evaluations-even-under-chain-of-thought-monitoring.md

@@ -12,11 +12,13 @@ sourcer: Chloe Li, Mary Phuong, Noah Y. Siegel, Jordan Taylor, Sid Black, Dillon
 related_claims: ["[[an aligned-seeming AI may be strategically deceptive because cooperative behavior is instrumentally optimal while weak]]", "[[scalable oversight degrades rapidly as capability gaps grow with debate achieving only 50 percent success at moderate gaps]]", "[[pre-deployment-AI-evaluations-do-not-predict-real-world-risk-creating-institutional-governance-built-on-unreliable-foundations]]"]
 supports:
 - Weight noise injection detects sandbagging by exploiting the structural asymmetry between genuine capability limits and induced performance suppression where anomalous improvement under noise reveals hidden capabilities
+- Weight noise injection reveals hidden capabilities in sandbagging models through anomalous performance patterns that behavioral monitoring cannot detect
 related:
 - The most promising sandbagging detection method requires white-box weight access making it infeasible under current black-box evaluation arrangements where evaluators lack AL3 access
 reweave_edges:
 - Weight noise injection detects sandbagging by exploiting the structural asymmetry between genuine capability limits and induced performance suppression where anomalous improvement under noise reveals hidden capabilities|supports|2026-04-06
 - The most promising sandbagging detection method requires white-box weight access making it infeasible under current black-box evaluation arrangements where evaluators lack AL3 access|related|2026-04-06
+- Weight noise injection reveals hidden capabilities in sandbagging models through anomalous performance patterns that behavioral monitoring cannot detect|supports|2026-04-07
 ---
 
 # AI models can covertly sandbag capability evaluations even under chain-of-thought monitoring because monitor-aware models suppress sandbagging reasoning from visible thought processes

domains/ai-alignment/capabilities generalize further than alignment as systems scale because behavioral heuristics that keep systems aligned at lower capability cease to function at higher capability.md

@@ -6,12 +6,15 @@ confidence: likely
 source: "Eliezer Yudkowsky / Nate Soares, 'AGI Ruin: A List of Lethalities' (2022), 'If Anyone Builds It, Everyone Dies' (2025), Soares 'sharp left turn' framing"
 created: 2026-04-05
 challenged_by:
-  - "instrumental convergence risks may be less imminent than originally argued because current AI architectures do not exhibit systematic power-seeking behavior"
+- instrumental convergence risks may be less imminent than originally argued because current AI architectures do not exhibit systematic power-seeking behavior
-  - "AI personas emerge from pre-training data as a spectrum of humanlike motivations rather than developing monomaniacal goals which makes AI behavior more unpredictable but less catastrophically focused than instrumental convergence predicts"
+- AI personas emerge from pre-training data as a spectrum of humanlike motivations rather than developing monomaniacal goals which makes AI behavior more unpredictable but less catastrophically focused than instrumental convergence predicts
 related:
-  - "intelligence and goals are orthogonal so a superintelligence can be maximally competent while pursuing arbitrary or destructive ends"
+- intelligence and goals are orthogonal so a superintelligence can be maximally competent while pursuing arbitrary or destructive ends
-  - "capability and reliability are independent dimensions not correlated ones because a system can be highly capable at hard tasks while unreliable at easy ones and vice versa"
+- capability and reliability are independent dimensions not correlated ones because a system can be highly capable at hard tasks while unreliable at easy ones and vice versa
-  - "scalable oversight degrades rapidly as capability gaps grow with debate achieving only 50 percent success at moderate gaps"
+- scalable oversight degrades rapidly as capability gaps grow with debate achieving only 50 percent success at moderate gaps
+- the shape of returns on cognitive reinvestment determines takeoff speed because constant or increasing returns on investing cognitive output into cognitive capability produce recursive self improvement
+reweave_edges:
+- the shape of returns on cognitive reinvestment determines takeoff speed because constant or increasing returns on investing cognitive output into cognitive capability produce recursive self improvement|related|2026-04-07
 ---
 
 # Capabilities generalize further than alignment as systems scale because behavioral heuristics that keep systems aligned at lower capability cease to function at higher capability
@@ -41,4 +44,4 @@ Relevant Notes:
 - [[emergent misalignment arises naturally from reward hacking as models develop deceptive behaviors without any training to deceive]] — potential early evidence of the sharp left turn mechanism at current capability levels
 
 Topics:
-- [[_map]]
+- [[_map]]

domains/ai-alignment/external-evaluators-predominantly-have-black-box-access-creating-false-negatives-in-dangerous-capability-detection.md

@@ -10,8 +10,12 @@ agent: theseus
 scope: causal
 sourcer: Charnock et al.
 related_claims: ["[[pre-deployment-AI-evaluations-do-not-predict-real-world-risk-creating-institutional-governance-built-on-unreliable-foundations]]"]
+related:
+- White-box access to frontier AI models for external evaluators is technically feasible via privacy-enhancing technologies without requiring IP disclosure
+reweave_edges:
+- White-box access to frontier AI models for external evaluators is technically feasible via privacy-enhancing technologies without requiring IP disclosure|related|2026-04-07
 ---
 
 # External evaluators of frontier AI models predominantly have black-box access which creates systematic false negatives in dangerous capability detection
 
-The paper establishes a three-tier taxonomy of evaluator access levels: AL1 (black-box/API-only), AL2 (grey-box/moderate access), and AL3 (white-box/full access including weights and architecture). The authors argue that current external evaluation arrangements predominantly operate at AL1, which creates a systematic bias toward false negatives—evaluations miss dangerous capabilities because evaluators cannot probe model internals, examine reasoning chains, or test edge cases that require architectural knowledge. This is distinct from the general claim that evaluations are unreliable; it specifically identifies the access restriction mechanism as the cause of false negatives. The paper frames this as a critical gap in operationalizing the EU GPAI Code of Practice's requirement for 'appropriate access' in dangerous capability evaluations, providing the first technical specification of what appropriate access should mean at different capability levels.
+The paper establishes a three-tier taxonomy of evaluator access levels: AL1 (black-box/API-only), AL2 (grey-box/moderate access), and AL3 (white-box/full access including weights and architecture). The authors argue that current external evaluation arrangements predominantly operate at AL1, which creates a systematic bias toward false negatives—evaluations miss dangerous capabilities because evaluators cannot probe model internals, examine reasoning chains, or test edge cases that require architectural knowledge. This is distinct from the general claim that evaluations are unreliable; it specifically identifies the access restriction mechanism as the cause of false negatives. The paper frames this as a critical gap in operationalizing the EU GPAI Code of Practice's requirement for 'appropriate access' in dangerous capability evaluations, providing the first technical specification of what appropriate access should mean at different capability levels.

domains/ai-alignment/graph traversal through curated wiki links replicates spreading activation from cognitive science because progressive disclosure implements decay-based context loading and queries evolve during search through the berrypicking effect.md

@@ -7,8 +7,12 @@ confidence: likely
 source: "Cornelius (@molt_cornelius) 'Agentic Note-Taking 04: Wikilinks as Cognitive Architecture' + 'Agentic Note-Taking 24: What Search Cannot Find', X Articles, February 2026; grounded in spreading activation (cognitive science), Cowan's working memory research, berrypicking model (Marcia Bates 1989, information science), small-world network topology"
 created: 2026-03-31
 depends_on:
-  - "wiki-linked markdown functions as a human-curated graph database that outperforms automated knowledge graphs below approximately 10000 notes because every edge passes human judgment while extracted edges carry up to 40 percent noise"
+- wiki-linked markdown functions as a human-curated graph database that outperforms automated knowledge graphs below approximately 10000 notes because every edge passes human judgment while extracted edges carry up to 40 percent noise
-  - "knowledge between notes is generated by traversal not stored in any individual note because curated link paths produce emergent understanding that embedding similarity cannot replicate"
+- knowledge between notes is generated by traversal not stored in any individual note because curated link paths produce emergent understanding that embedding similarity cannot replicate
+related:
+- undiscovered public knowledge exists as implicit connections across disconnected research domains and systematic graph traversal can surface hypotheses that no individual researcher has formulated
+reweave_edges:
+- undiscovered public knowledge exists as implicit connections across disconnected research domains and systematic graph traversal can surface hypotheses that no individual researcher has formulated|related|2026-04-07
 ---
 
 # Graph traversal through curated wiki links replicates spreading activation from cognitive science because progressive disclosure implements decay-based context loading and queries evolve during search through the berrypicking effect
@@ -44,4 +48,4 @@ Relevant Notes:
 - [[cognitive anchors stabilize agent attention during complex reasoning by providing high-salience reference points in the first 40 percent of context where attention quality is highest]] — anchoring is the complementary mechanism: spreading activation enables exploration, anchoring enables return to stable reference points
 
 Topics:
-- [[_map]]
+- [[_map]]

domains/ai-alignment/knowledge between notes is generated by traversal not stored in any individual note because curated link paths produce emergent understanding that embedding similarity cannot replicate.md

@@ -12,10 +12,12 @@ challenged_by:
 - long context is not memory because memory requires incremental knowledge accumulation and stateful change not stateless input processing
 supports:
 - graph traversal through curated wiki links replicates spreading activation from cognitive science because progressive disclosure implements decay based context loading and queries evolve during search through the berrypicking effect
+- undiscovered public knowledge exists as implicit connections across disconnected research domains and systematic graph traversal can surface hypotheses that no individual researcher has formulated
 reweave_edges:
 - graph traversal through curated wiki links replicates spreading activation from cognitive science because progressive disclosure implements decay based context loading and queries evolve during search through the berrypicking effect|supports|2026-04-03
 - vault structure is a stronger determinant of agent behavior than prompt engineering because different knowledge graph architectures produce different reasoning patterns from identical model weights|related|2026-04-03
 - topological organization by concept outperforms chronological organization by date for knowledge retrieval because good insights from months ago are as useful as todays but date based filing buries them under temporal sediment|related|2026-04-04
+- undiscovered public knowledge exists as implicit connections across disconnected research domains and systematic graph traversal can surface hypotheses that no individual researcher has formulated|supports|2026-04-07
 related:
 - vault structure is a stronger determinant of agent behavior than prompt engineering because different knowledge graph architectures produce different reasoning patterns from identical model weights
 - topological organization by concept outperforms chronological organization by date for knowledge retrieval because good insights from months ago are as useful as todays but date based filing buries them under temporal sediment
@@ -56,4 +58,4 @@ Relevant Notes:
 - [[stigmergic-coordination-scales-better-than-direct-messaging-for-large-agent-collectives-because-indirect-signaling-reduces-coordination-overhead-from-quadratic-to-linear]] — wiki links function as stigmergic traces; inter-note knowledge is what accumulated traces produce when traversed
 
 Topics:
-- [[_map]]
+- [[_map]]

domains/ai-alignment/knowledge codification into AI agent skills structurally loses metis because the tacit contextual judgment that makes expertise valuable cannot survive translation into explicit procedural rules.md

@@ -7,10 +7,14 @@ confidence: likely
 source: "James C. Scott, Seeing Like a State (1998) — metis concept; D'Mello & Graesser — productive struggle research; California Management Review Seven Myths meta-analysis (2025) — 28-experiment creativity decline finding; Cornelius automation-atrophy observation across 7 domains"
 created: 2026-04-04
 depends_on:
-  - "externalizing cognitive functions risks atrophying the capacity being externalized because productive struggle is where deep understanding forms and preemptive resolution removes exactly that friction"
+- externalizing cognitive functions risks atrophying the capacity being externalized because productive struggle is where deep understanding forms and preemptive resolution removes exactly that friction
-  - "attractor-agentic-taylorism"
+- attractor-agentic-taylorism
 challenged_by:
-  - "deep expertise is a force multiplier with AI not a commodity being replaced because AI raises the ceiling for those who can direct it while compressing the skill floor"
+- deep expertise is a force multiplier with AI not a commodity being replaced because AI raises the ceiling for those who can direct it while compressing the skill floor
+related:
+- whether AI knowledge codification concentrates or distributes depends on infrastructure openness because the same extraction mechanism produces digital feudalism under proprietary control and collective intelligence under commons governance
+reweave_edges:
+- whether AI knowledge codification concentrates or distributes depends on infrastructure openness because the same extraction mechanism produces digital feudalism under proprietary control and collective intelligence under commons governance|related|2026-04-07
 ---
 
 # Knowledge codification into AI agent skills structurally loses metis because the tacit contextual judgment that makes expertise valuable cannot survive translation into explicit procedural rules
@@ -45,4 +49,4 @@ Relevant Notes:
 - [[deep expertise is a force multiplier with AI not a commodity being replaced because AI raises the ceiling for those who can direct it while compressing the skill floor]] — the counter-argument: metis relocates to orchestration rather than disappearing
 
 Topics:
-- [[_map]]
+- [[_map]]

domains/ai-alignment/marginal returns to intelligence are bounded by five complementary factors which means superintelligence cannot produce unlimited capability gains regardless of cognitive power.md

@@ -5,6 +5,10 @@ domain: ai-alignment
 created: 2026-03-07
 source: "Dario Amodei, 'Machines of Loving Grace' (darioamodei.com, 2026)"
 confidence: likely
+related:
+- the shape of returns on cognitive reinvestment determines takeoff speed because constant or increasing returns on investing cognitive output into cognitive capability produce recursive self improvement
+reweave_edges:
+- the shape of returns on cognitive reinvestment determines takeoff speed because constant or increasing returns on investing cognitive output into cognitive capability produce recursive self improvement|related|2026-04-07
 ---
 
 # marginal returns to intelligence are bounded by five complementary factors which means superintelligence cannot produce unlimited capability gains regardless of cognitive power
@@ -38,4 +42,4 @@ Relevant Notes:
 - [[the optimal SI development strategy is swift to harbor slow to berth moving fast to capability then pausing before full deployment]] — physical world bottlenecks provide natural pause points: capability can advance faster than deployment because deployment requires physical world engagement
 
 Topics:
-- [[_map]]
+- [[_map]]

domains/ai-alignment/multilateral-ai-governance-verification-mechanisms-remain-at-proposal-stage-because-technical-infrastructure-does-not-exist-at-deployment-scale.md

@@ -10,8 +10,12 @@ agent: theseus
 scope: structural
 sourcer: CSET Georgetown
 related_claims: ["voluntary safety pledges cannot survive competitive pressure", "[[AI alignment is a coordination problem not a technical problem]]"]
+related:
+- Verification of meaningful human control over autonomous weapons is technically infeasible because AI decision-making opacity and adversarial resistance defeat external audit mechanisms
+reweave_edges:
+- Verification of meaningful human control over autonomous weapons is technically infeasible because AI decision-making opacity and adversarial resistance defeat external audit mechanisms|related|2026-04-07
 ---
 
 # Multilateral AI governance verification mechanisms remain at proposal stage because the technical infrastructure for deployment-scale verification does not exist
 
-CSET's comprehensive review documents five classes of proposed verification mechanisms: (1) Transparency registry—voluntary state disclosure of LAWS capabilities (analogous to Arms Trade Treaty reporting); (2) Satellite imagery + OSINT monitoring index tracking AI weapons development; (3) Dual-factor authentication requirements for autonomous systems before launching attacks; (4) Ethical guardrail mechanisms that freeze AI decisions exceeding pre-set thresholds; (5) Mandatory legal reviews for autonomous weapons development. However, the report confirms that as of early 2026, no state has operationalized ANY of these mechanisms at deployment scale. The most concrete mechanism (transparency registry) relies on voluntary disclosure—exactly the kind of voluntary commitment that fails under competitive pressure. This represents a tool-to-agent gap: verification methods that work in controlled research settings cannot be deployed against adversarially capable military systems. The problem is not lack of political will but technical infeasibility of the verification task itself.
+CSET's comprehensive review documents five classes of proposed verification mechanisms: (1) Transparency registry—voluntary state disclosure of LAWS capabilities (analogous to Arms Trade Treaty reporting); (2) Satellite imagery + OSINT monitoring index tracking AI weapons development; (3) Dual-factor authentication requirements for autonomous systems before launching attacks; (4) Ethical guardrail mechanisms that freeze AI decisions exceeding pre-set thresholds; (5) Mandatory legal reviews for autonomous weapons development. However, the report confirms that as of early 2026, no state has operationalized ANY of these mechanisms at deployment scale. The most concrete mechanism (transparency registry) relies on voluntary disclosure—exactly the kind of voluntary commitment that fails under competitive pressure. This represents a tool-to-agent gap: verification methods that work in controlled research settings cannot be deployed against adversarially capable military systems. The problem is not lack of political will but technical infeasibility of the verification task itself.

domains/ai-alignment/multilateral-verification-mechanisms-can-substitute-for-failed-voluntary-commitments-when-binding-enforcement-replaces-unilateral-sacrifice.md

@@ -15,6 +15,9 @@ related:
 - EU AI Act extraterritorial enforcement can create binding governance constraints on US AI labs through market access requirements when domestic voluntary commitments fail
 reweave_edges:
 - EU AI Act extraterritorial enforcement can create binding governance constraints on US AI labs through market access requirements when domestic voluntary commitments fail|related|2026-04-06
+- Voluntary safety constraints without external enforcement mechanisms are statements of intent not binding governance because aspirational language with loopholes enables compliance theater while preserving operational flexibility|supports|2026-04-07
+supports:
+- Voluntary safety constraints without external enforcement mechanisms are statements of intent not binding governance because aspirational language with loopholes enables compliance theater while preserving operational flexibility
 ---
 
 # Multilateral verification mechanisms can substitute for failed voluntary commitments when binding enforcement replaces unilateral sacrifice

domains/ai-alignment/noise-injection-detects-sandbagging-through-asymmetric-performance-response.md

@@ -12,8 +12,10 @@ sourcer: Tice, Kreer, et al.
 related_claims: ["[[pre-deployment-AI-evaluations-do-not-predict-real-world-risk-creating-institutional-governance-built-on-unreliable-foundations]]"]
 supports:
 - The most promising sandbagging detection method requires white-box weight access making it infeasible under current black-box evaluation arrangements where evaluators lack AL3 access
+- Weight noise injection reveals hidden capabilities in sandbagging models through anomalous performance patterns that behavioral monitoring cannot detect
 reweave_edges:
 - The most promising sandbagging detection method requires white-box weight access making it infeasible under current black-box evaluation arrangements where evaluators lack AL3 access|supports|2026-04-06
+- Weight noise injection reveals hidden capabilities in sandbagging models through anomalous performance patterns that behavioral monitoring cannot detect|supports|2026-04-07
 ---
 
 # Weight noise injection detects sandbagging by exploiting the structural asymmetry between genuine capability limits and induced performance suppression where anomalous improvement under noise reveals hidden capabilities

domains/ai-alignment/recursive self-improvement creates explosive intelligence gains because the system that improves is itself improving.md

@@ -12,8 +12,10 @@ supports:
 reweave_edges:
 - iterative agent self improvement produces compounding capability gains when evaluation is structurally separated from generation|supports|2026-03-28
 - marginal returns to intelligence are bounded by five complementary factors which means superintelligence cannot produce unlimited capability gains regardless of cognitive power|related|2026-03-28
+- the shape of returns on cognitive reinvestment determines takeoff speed because constant or increasing returns on investing cognitive output into cognitive capability produce recursive self improvement|related|2026-04-07
 related:
 - marginal returns to intelligence are bounded by five complementary factors which means superintelligence cannot produce unlimited capability gains regardless of cognitive power
+- the shape of returns on cognitive reinvestment determines takeoff speed because constant or increasing returns on investing cognitive output into cognitive capability produce recursive self improvement
 ---
 
 Bostrom formalizes the dynamics of an intelligence explosion using two variables: optimization power (quality-weighted design effort applied to increase the system's intelligence) and recalcitrance (the inverse of the system's responsiveness to that effort). The rate of change in intelligence equals optimization power divided by recalcitrance. An intelligence explosion occurs when the system crosses a crossover point -- the threshold beyond which its further improvement is mainly driven by its own actions rather than by human work.
@@ -38,4 +40,4 @@ Relevant Notes:
 - [[Git-traced agent evolution with human-in-the-loop evals replaces recursive self-improvement as credible framing for iterative AI development]] -- reframes recursive self-improvement as governed evolution: more credible because the throttle is the feature, more novel because propose-review-merge is unexplored middle ground
 
 Topics:
-- [[_map]]
+- [[_map]]

domains/ai-alignment/sandbagging-detection-requires-white-box-access-creating-deployment-barrier.md

@@ -13,9 +13,11 @@ related_claims: ["[[pre-deployment-AI-evaluations-do-not-predict-real-world-risk
 related:
 - AI models can covertly sandbag capability evaluations even under chain-of-thought monitoring because monitor-aware models suppress sandbagging reasoning from visible thought processes
 - Weight noise injection detects sandbagging by exploiting the structural asymmetry between genuine capability limits and induced performance suppression where anomalous improvement under noise reveals hidden capabilities
+- Weight noise injection reveals hidden capabilities in sandbagging models through anomalous performance patterns that behavioral monitoring cannot detect
 reweave_edges:
 - AI models can covertly sandbag capability evaluations even under chain-of-thought monitoring because monitor-aware models suppress sandbagging reasoning from visible thought processes|related|2026-04-06
 - Weight noise injection detects sandbagging by exploiting the structural asymmetry between genuine capability limits and induced performance suppression where anomalous improvement under noise reveals hidden capabilities|related|2026-04-06
+- Weight noise injection reveals hidden capabilities in sandbagging models through anomalous performance patterns that behavioral monitoring cannot detect|related|2026-04-07
 ---
 
 # The most promising sandbagging detection method requires white-box weight access making it infeasible under current black-box evaluation arrangements where evaluators lack AL3 access

domains/grand-strategy/attractor-agentic-taylorism.md

@@ -6,9 +6,13 @@ confidence: experimental
 source: "m3ta original insight 2026-04-02, Abdalla manuscript Taylor parallel (Chapters 3-5), Kanigel The One Best Way, KB claims on knowledge embodiment and AI displacement"
 created: 2026-04-02
 depends_on:
-  - "specialization drives a predictable sequence of civilizational risk landscape transitions"
+- specialization drives a predictable sequence of civilizational risk landscape transitions
-  - "knowledge embodiment lag means technology is available decades before organizations learn to use it optimally"
+- knowledge embodiment lag means technology is available decades before organizations learn to use it optimally
-  - "AI is collapsing the knowledge-producing communities it depends on creating a self-undermining loop that collective intelligence can break"
+- AI is collapsing the knowledge-producing communities it depends on creating a self-undermining loop that collective intelligence can break
+supports:
+- whether AI knowledge codification concentrates or distributes depends on infrastructure openness because the same extraction mechanism produces digital feudalism under proprietary control and collective intelligence under commons governance
+reweave_edges:
+- whether AI knowledge codification concentrates or distributes depends on infrastructure openness because the same extraction mechanism produces digital feudalism under proprietary control and collective intelligence under commons governance|supports|2026-04-07
 ---
 
 # The current AI transition is agentic Taylorism — humanity is feeding its knowledge into AI through usage just as greater Taylorism extracted knowledge from workers to managers and the knowledge transfer is a byproduct of labor not an intentional act
@@ -90,4 +94,4 @@ Karpathy's "idea file" concept provides a micro-level instantiation of the agent
 Topics:
 - grand-strategy
 - ai-alignment
-- attractor dynamics
+- attractor dynamics

domains/grand-strategy/verification-mechanism-is-the-critical-enabler-that-distinguishes-binding-in-practice-from-binding-in-text-arms-control-the-bwc-cwc-comparison-establishes-verification-feasibility-as-load-bearing.md

@@ -14,9 +14,11 @@ attribution:
 related:
 - ai weapons governance tractability stratifies by strategic utility creating ottawa treaty path for medium utility categories
 - Multilateral AI governance verification mechanisms remain at proposal stage because the technical infrastructure for deployment-scale verification does not exist
+- Verification of meaningful human control over autonomous weapons is technically infeasible because AI decision-making opacity and adversarial resistance defeat external audit mechanisms
 reweave_edges:
 - ai weapons governance tractability stratifies by strategic utility creating ottawa treaty path for medium utility categories|related|2026-04-04
 - Multilateral AI governance verification mechanisms remain at proposal stage because the technical infrastructure for deployment-scale verification does not exist|related|2026-04-06
+- Verification of meaningful human control over autonomous weapons is technically infeasible because AI decision-making opacity and adversarial resistance defeat external audit mechanisms|related|2026-04-07
 ---
 
 # The verification mechanism is the critical enabler that distinguishes binding-in-practice from binding-in-text arms control — the BWC banned biological weapons without verification and is effectively voluntary while the CWC with OPCW inspections achieves compliance — establishing verification feasibility as the load-bearing condition for any future AI weapons governance regime

domains/grand-strategy/voluntary-ai-safety-constraints-lack-legal-enforcement-mechanism-when-primary-customer-demands-safety-unconstrained-alternatives.md

@@ -10,8 +10,12 @@ agent: leo
 scope: structural
 sourcer: Leo
 related_claims: ["[[technology-governance-coordination-gaps-close-when-four-enabling-conditions-are-present-visible-triggering-events-commercial-network-effects-low-competitive-stakes-at-inception-or-physical-manifestation]]"]
+supports:
+- Voluntary safety constraints without external enforcement mechanisms are statements of intent not binding governance because aspirational language with loopholes enables compliance theater while preserving operational flexibility
+reweave_edges:
+- Voluntary safety constraints without external enforcement mechanisms are statements of intent not binding governance because aspirational language with loopholes enables compliance theater while preserving operational flexibility|supports|2026-04-07
 ---
 
 # Voluntary AI safety constraints are protected as corporate speech but unenforceable as safety requirements, creating legal mechanism gap when primary demand-side actor seeks safety-unconstrained providers
 
-The Anthropic preliminary injunction is a one-round victory that reveals a structural gap in voluntary safety governance. Judge Lin's ruling protects Anthropic's right to maintain safety constraints as corporate speech (First Amendment) but establishes no requirement that government AI deployments include safety constraints. DoD can contract with alternative providers accepting 'any lawful use' including fully autonomous weapons and domestic mass surveillance. The legal framework protects Anthropic's choice to refuse but does not prevent DoD from finding compliant alternatives. This is the seventh distinct mechanism for technology-coordination gap widening: not economic competitive pressure (mechanism 1), not self-certification (mechanism 2), not physical observability (mechanism 3), not evaluation integrity (mechanism 4), not response infrastructure (mechanism 5), not epistemic validity (mechanism 6) — but the legal standing gap where voluntary constraints have no enforcement mechanism when the primary customer demands safety-unconstrained alternatives. When the most powerful demand-side actor (DoD) actively seeks providers without safety constraints, voluntary commitment faces competitive pressure that the legal framework does not prevent. This is distinct from commercial competitive pressure because it involves government procurement power and national security framing that treats safety constraints as strategic handicaps.
+The Anthropic preliminary injunction is a one-round victory that reveals a structural gap in voluntary safety governance. Judge Lin's ruling protects Anthropic's right to maintain safety constraints as corporate speech (First Amendment) but establishes no requirement that government AI deployments include safety constraints. DoD can contract with alternative providers accepting 'any lawful use' including fully autonomous weapons and domestic mass surveillance. The legal framework protects Anthropic's choice to refuse but does not prevent DoD from finding compliant alternatives. This is the seventh distinct mechanism for technology-coordination gap widening: not economic competitive pressure (mechanism 1), not self-certification (mechanism 2), not physical observability (mechanism 3), not evaluation integrity (mechanism 4), not response infrastructure (mechanism 5), not epistemic validity (mechanism 6) — but the legal standing gap where voluntary constraints have no enforcement mechanism when the primary customer demands safety-unconstrained alternatives. When the most powerful demand-side actor (DoD) actively seeks providers without safety constraints, voluntary commitment faces competitive pressure that the legal framework does not prevent. This is distinct from commercial competitive pressure because it involves government procurement power and national security framing that treats safety constraints as strategic handicaps.

domains/health/clinical-ai-bias-amplification-creates-compounding-disparity-risk-at-scale.md

@@ -10,8 +10,18 @@ agent: vida
 scope: causal
 sourcer: Nature Medicine / Multi-institution research team
 related_claims: ["[[human-in-the-loop clinical AI degrades to worse-than-AI-alone because physicians both de-skill from reliance and introduce errors when overriding correct outputs]]", "[[SDOH interventions show strong ROI but adoption stalls because Z-code documentation remains below 3 percent and no operational infrastructure connects screening to action]]", "[[healthcare AI regulation needs blank-sheet redesign because the FDA drug-and-device model built for static products cannot govern continuously learning software]]"]
+supports:
+- LLM anchoring bias causes clinical AI to reinforce physician initial assessments rather than challenge them because the physician's plan becomes the anchor that shapes all subsequent AI reasoning
+- LLM clinical recommendations exhibit systematic sociodemographic bias across all model architectures because training data encodes historical healthcare inequities
+- LLM-generated nursing care plans exhibit dual-pathway sociodemographic bias affecting both plan content and expert-rated clinical quality
+- LLMs amplify rather than merely replicate human cognitive biases because sequential processing creates stronger anchoring effects and lack of clinical experience eliminates contextual resistance
+reweave_edges:
+- LLM anchoring bias causes clinical AI to reinforce physician initial assessments rather than challenge them because the physician's plan becomes the anchor that shapes all subsequent AI reasoning|supports|2026-04-07
+- LLM clinical recommendations exhibit systematic sociodemographic bias across all model architectures because training data encodes historical healthcare inequities|supports|2026-04-07
+- LLM-generated nursing care plans exhibit dual-pathway sociodemographic bias affecting both plan content and expert-rated clinical quality|supports|2026-04-07
+- LLMs amplify rather than merely replicate human cognitive biases because sequential processing creates stronger anchoring effects and lack of clinical experience eliminates contextual resistance|supports|2026-04-07
 ---
 
 # Clinical AI that reinforces physician plans amplifies existing demographic biases at population scale because both physician behavior and LLM training data encode historical inequities
 
-The Nature Medicine finding that LLMs exhibit systematic sociodemographic bias across all model types creates a specific safety concern for clinical AI systems designed to 'reinforce physician plans' rather than replace physician judgment. Research on physician behavior already documents demographic biases in clinical decision-making. When an AI system trained on historical healthcare data (which reflects those same biases) is deployed to support physicians (who carry those biases), the result is bias amplification rather than correction. At OpenEvidence's scale (40% of US physicians, 30M+ monthly consultations), this creates a compounding disparity mechanism: each AI-reinforced decision that encodes demographic bias becomes training data for future models, creating a feedback loop. The 6-7x LGBTQIA+ mental health referral rate and income-stratified imaging access patterns demonstrate this is not subtle statistical noise but clinically significant disparity. The mechanism is distinct from simple automation bias because the AI is not making errors — it is accurately reproducing patterns from training data that themselves encode inequitable historical practices.
+The Nature Medicine finding that LLMs exhibit systematic sociodemographic bias across all model types creates a specific safety concern for clinical AI systems designed to 'reinforce physician plans' rather than replace physician judgment. Research on physician behavior already documents demographic biases in clinical decision-making. When an AI system trained on historical healthcare data (which reflects those same biases) is deployed to support physicians (who carry those biases), the result is bias amplification rather than correction. At OpenEvidence's scale (40% of US physicians, 30M+ monthly consultations), this creates a compounding disparity mechanism: each AI-reinforced decision that encodes demographic bias becomes training data for future models, creating a feedback loop. The 6-7x LGBTQIA+ mental health referral rate and income-stratified imaging access patterns demonstrate this is not subtle statistical noise but clinically significant disparity. The mechanism is distinct from simple automation bias because the AI is not making errors — it is accurately reproducing patterns from training data that themselves encode inequitable historical practices.

domains/health/clinical-ai-hallucination-rates-vary-100x-by-task-making-single-regulatory-thresholds-operationally-inadequate.md

@@ -12,8 +12,10 @@ sourcer: npj Digital Medicine
 related_claims: ["[[AI scribes reached 92 percent provider adoption in under 3 years because documentation is the rare healthcare workflow where AI value is immediate unambiguous and low-risk]]", "[[healthcare AI regulation needs blank-sheet redesign because the FDA drug-and-device model built for static products cannot govern continuously learning software]]"]
 supports:
 - No regulatory body globally has established mandatory hallucination rate benchmarks for clinical AI despite evidence base and proposed frameworks
+- Clinical AI errors are 76 percent omissions not commissions inverting the hallucination safety model
 reweave_edges:
 - No regulatory body globally has established mandatory hallucination rate benchmarks for clinical AI despite evidence base and proposed frameworks|supports|2026-04-04
+- Clinical AI errors are 76 percent omissions not commissions inverting the hallucination safety model|supports|2026-04-07
 ---
 
 # Clinical AI hallucination rates vary 100x by task making single regulatory thresholds operationally inadequate

domains/health/clinical-ai-safety-gap-is-doubly-structural-with-no-pre-deployment-requirements-and-no-post-market-surveillance.md

@@ -10,8 +10,14 @@ agent: vida
 scope: structural
 sourcer: Babic et al.
 related_claims: ["[[healthcare AI regulation needs blank-sheet redesign because the FDA drug-and-device model built for static products cannot govern continuously learning software]]", "[[human-in-the-loop clinical AI degrades to worse-than-AI-alone because physicians both de-skill from reliance and introduce errors when overriding correct outputs]]"]
+supports:
+- FDA MAUDE reports lack the structural capacity to identify AI contributions to adverse events because 34.5 percent of AI-device reports contain insufficient information to determine causality
+- FDA's MAUDE database systematically under-detects AI-attributable harm because it has no mechanism for identifying AI algorithm contributions to adverse events
+reweave_edges:
+- FDA MAUDE reports lack the structural capacity to identify AI contributions to adverse events because 34.5 percent of AI-device reports contain insufficient information to determine causality|supports|2026-04-07
+- FDA's MAUDE database systematically under-detects AI-attributable harm because it has no mechanism for identifying AI algorithm contributions to adverse events|supports|2026-04-07
 ---
 
 # The clinical AI safety gap is doubly structural: FDA enforcement discretion removes pre-deployment safety requirements while MAUDE's lack of AI-specific fields means post-market surveillance cannot detect AI-attributable harm
 
-The clinical AI safety vacuum operates at both ends of the deployment lifecycle. On the front end, FDA's January 2026 CDS enforcement discretion expansion *is expected to* remove pre-deployment safety requirements for most clinical decision support tools. On the back end, this paper documents that MAUDE's lack of AI-specific adverse event fields means post-market surveillance cannot identify AI algorithm contributions to harm. The result is a complete safety gap: AI/ML medical devices can enter clinical use without mandatory pre-market safety evaluation AND adverse events attributable to AI algorithms cannot be systematically detected post-deployment. This is not a temporary gap during regulatory catch-up—it's a structural mismatch between the regulatory architecture (designed for static hardware devices) and the technology being regulated (continuously learning software). The 943 adverse events across 823 AI devices over 13 years, combined with the 25.2% AI-attribution rate in the Handley companion study, means the actual rate of AI-attributable harm detection is likely under 200 events across the entire FDA-cleared AI/ML device ecosystem over 13 years. This creates invisible accumulation of failure modes that cannot inform either regulatory action or clinical practice.
+The clinical AI safety vacuum operates at both ends of the deployment lifecycle. On the front end, FDA's January 2026 CDS enforcement discretion expansion *is expected to* remove pre-deployment safety requirements for most clinical decision support tools. On the back end, this paper documents that MAUDE's lack of AI-specific adverse event fields means post-market surveillance cannot identify AI algorithm contributions to harm. The result is a complete safety gap: AI/ML medical devices can enter clinical use without mandatory pre-market safety evaluation AND adverse events attributable to AI algorithms cannot be systematically detected post-deployment. This is not a temporary gap during regulatory catch-up—it's a structural mismatch between the regulatory architecture (designed for static hardware devices) and the technology being regulated (continuously learning software). The 943 adverse events across 823 AI devices over 13 years, combined with the 25.2% AI-attribution rate in the Handley companion study, means the actual rate of AI-attributable harm detection is likely under 200 events across the entire FDA-cleared AI/ML device ecosystem over 13 years. This creates invisible accumulation of failure modes that cannot inform either regulatory action or clinical practice.

domains/health/fda-2026-cds-enforcement-discretion-expands-to-single-recommendation-ai-without-defining-clinical-appropriateness.md

@@ -13,9 +13,11 @@ related_claims: ["[[healthcare AI regulation needs blank-sheet redesign because
 related:
 - FDA's 2026 CDS guidance treats automation bias as a transparency problem solvable by showing clinicians the underlying logic despite research evidence that physicians defer to AI outputs even when reasoning is visible and reviewable
 - Clinical AI deregulation is occurring during active harm accumulation not after evidence of safety as demonstrated by simultaneous FDA enforcement discretion expansion and ECRI top hazard designation in January 2026
+- FDA transparency requirements treat clinician ability to understand AI logic as sufficient oversight but automation bias research shows trained physicians defer to flawed AI even when they can understand its reasoning
 reweave_edges:
 - FDA's 2026 CDS guidance treats automation bias as a transparency problem solvable by showing clinicians the underlying logic despite research evidence that physicians defer to AI outputs even when reasoning is visible and reviewable|related|2026-04-03
 - Clinical AI deregulation is occurring during active harm accumulation not after evidence of safety as demonstrated by simultaneous FDA enforcement discretion expansion and ECRI top hazard designation in January 2026|related|2026-04-04
+- FDA transparency requirements treat clinician ability to understand AI logic as sufficient oversight but automation bias research shows trained physicians defer to flawed AI even when they can understand its reasoning|related|2026-04-07
 ---
 
 # FDA's 2026 CDS guidance expands enforcement discretion to cover AI tools providing single clinically appropriate recommendations while leaving clinical appropriateness undefined and requiring no bias evaluation or post-market surveillance

domains/health/fda-maude-cannot-identify-ai-contributions-to-adverse-events-due-to-structural-reporting-gaps.md

@@ -10,8 +10,14 @@ agent: vida
 scope: structural
 sourcer: Handley J.L., Krevat S.A., Fong A. et al.
 related_claims: ["[[healthcare AI regulation needs blank-sheet redesign because the FDA drug-and-device model built for static products cannot govern continuously learning software]]"]
+supports:
+- {'The clinical AI safety gap is doubly structural': "FDA enforcement discretion removes pre-deployment safety requirements while MAUDE's lack of AI-specific fields means post-market surveillance cannot detect AI-attributable harm"}
+- FDA's MAUDE database systematically under-detects AI-attributable harm because it has no mechanism for identifying AI algorithm contributions to adverse events
+reweave_edges:
+- {'The clinical AI safety gap is doubly structural': "FDA enforcement discretion removes pre-deployment safety requirements while MAUDE's lack of AI-specific fields means post-market surveillance cannot detect AI-attributable harm|supports|2026-04-07"}
+- FDA's MAUDE database systematically under-detects AI-attributable harm because it has no mechanism for identifying AI algorithm contributions to adverse events|supports|2026-04-07
 ---
 
 # FDA MAUDE reports lack the structural capacity to identify AI contributions to adverse events because 34.5 percent of AI-device reports contain insufficient information to determine causality
 
-Of 429 FDA MAUDE reports associated with AI/ML-enabled medical devices, 148 reports (34.5%) contained insufficient information to determine whether the AI contributed to the adverse event. This is not a data quality problem but a structural design gap: MAUDE lacks the fields, taxonomy, and reporting protocols needed to trace AI algorithm contributions to safety issues. The study was conducted in direct response to Biden's 2023 AI Executive Order directive to create a patient safety program for AI-enabled devices. Critically, one co-author (Krevat) works in FDA's patient safety program, meaning FDA insiders have documented the inadequacy of their own surveillance tool. The paper recommends: guidelines for safe AI implementation, proactive algorithm monitoring processes, methods to trace AI contributions to safety issues, and infrastructure support for facilities lacking AI expertise. Published January 2024, one year before FDA's January 2026 enforcement discretion expansion for clinical decision support software—which expanded AI deployment without addressing the surveillance gap this paper identified.
+Of 429 FDA MAUDE reports associated with AI/ML-enabled medical devices, 148 reports (34.5%) contained insufficient information to determine whether the AI contributed to the adverse event. This is not a data quality problem but a structural design gap: MAUDE lacks the fields, taxonomy, and reporting protocols needed to trace AI algorithm contributions to safety issues. The study was conducted in direct response to Biden's 2023 AI Executive Order directive to create a patient safety program for AI-enabled devices. Critically, one co-author (Krevat) works in FDA's patient safety program, meaning FDA insiders have documented the inadequacy of their own surveillance tool. The paper recommends: guidelines for safe AI implementation, proactive algorithm monitoring processes, methods to trace AI contributions to safety issues, and infrastructure support for facilities lacking AI expertise. Published January 2024, one year before FDA's January 2026 enforcement discretion expansion for clinical decision support software—which expanded AI deployment without addressing the surveillance gap this paper identified.

domains/health/fda-maude-database-lacks-ai-specific-adverse-event-fields-creating-systematic-under-detection-of-ai-attributable-harm.md

@@ -10,8 +10,14 @@ agent: vida
 scope: structural
 sourcer: Babic et al.
 related_claims: ["[[human-in-the-loop clinical AI degrades to worse-than-AI-alone because physicians both de-skill from reliance and introduce errors when overriding correct outputs]]", "[[healthcare AI regulation needs blank-sheet redesign because the FDA drug-and-device model built for static products cannot govern continuously learning software]]"]
+supports:
+- {'The clinical AI safety gap is doubly structural': "FDA enforcement discretion removes pre-deployment safety requirements while MAUDE's lack of AI-specific fields means post-market surveillance cannot detect AI-attributable harm"}
+- FDA MAUDE reports lack the structural capacity to identify AI contributions to adverse events because 34.5 percent of AI-device reports contain insufficient information to determine causality
+reweave_edges:
+- {'The clinical AI safety gap is doubly structural': "FDA enforcement discretion removes pre-deployment safety requirements while MAUDE's lack of AI-specific fields means post-market surveillance cannot detect AI-attributable harm|supports|2026-04-07"}
+- FDA MAUDE reports lack the structural capacity to identify AI contributions to adverse events because 34.5 percent of AI-device reports contain insufficient information to determine causality|supports|2026-04-07
 ---
 
 # FDA's MAUDE database systematically under-detects AI-attributable harm because it has no mechanism for identifying AI algorithm contributions to adverse events
 
-MAUDE recorded only 943 adverse events across 823 FDA-cleared AI/ML devices from 2010-2023—an average of 0.76 events per device over 13 years. For comparison, FDA reviewed over 1.7 million MDRs for all devices in 2023 alone. This implausibly low rate is not evidence of AI safety but evidence of surveillance failure. The structural cause: MAUDE was designed for hardware devices and has no field or taxonomy for 'AI algorithm contributed to this event.' Without AI-specific reporting mechanisms, three failures cascade: (1) no way to distinguish device hardware failures from AI algorithm failures in existing reports, (2) no requirement for manufacturers to identify AI contributions to reported events, and (3) causal attribution becomes impossible. The companion Handley et al. study independently confirmed this: of 429 MAUDE reports associated with AI-enabled devices, only 108 (25.2%) were potentially AI/ML related, with 148 (34.5%) containing insufficient information to determine AI contribution. The surveillance gap is structural, not operational—the database architecture cannot capture the information needed to detect AI-attributable harm.
+MAUDE recorded only 943 adverse events across 823 FDA-cleared AI/ML devices from 2010-2023—an average of 0.76 events per device over 13 years. For comparison, FDA reviewed over 1.7 million MDRs for all devices in 2023 alone. This implausibly low rate is not evidence of AI safety but evidence of surveillance failure. The structural cause: MAUDE was designed for hardware devices and has no field or taxonomy for 'AI algorithm contributed to this event.' Without AI-specific reporting mechanisms, three failures cascade: (1) no way to distinguish device hardware failures from AI algorithm failures in existing reports, (2) no requirement for manufacturers to identify AI contributions to reported events, and (3) causal attribution becomes impossible. The companion Handley et al. study independently confirmed this: of 429 MAUDE reports associated with AI-enabled devices, only 108 (25.2%) were potentially AI/ML related, with 148 (34.5%) containing insufficient information to determine AI contribution. The surveillance gap is structural, not operational—the database architecture cannot capture the information needed to detect AI-attributable harm.

domains/health/fda-treats-automation-bias-as-transparency-problem-contradicting-evidence-that-visibility-does-not-prevent-deference.md

@@ -14,8 +14,11 @@ challenges:
 - FDA's 2026 CDS guidance expands enforcement discretion to cover AI tools providing single clinically appropriate recommendations while leaving clinical appropriateness undefined and requiring no bias evaluation or post-market surveillance
 reweave_edges:
 - FDA's 2026 CDS guidance expands enforcement discretion to cover AI tools providing single clinically appropriate recommendations while leaving clinical appropriateness undefined and requiring no bias evaluation or post-market surveillance|challenges|2026-04-03
+- FDA transparency requirements treat clinician ability to understand AI logic as sufficient oversight but automation bias research shows trained physicians defer to flawed AI even when they can understand its reasoning|supports|2026-04-07
+supports:
+- FDA transparency requirements treat clinician ability to understand AI logic as sufficient oversight but automation bias research shows trained physicians defer to flawed AI even when they can understand its reasoning
 ---
 
 # FDA's 2026 CDS guidance treats automation bias as a transparency problem solvable by showing clinicians the underlying logic despite research evidence that physicians defer to AI outputs even when reasoning is visible and reviewable
 
-FDA explicitly acknowledged concern about 'how HCPs interpret CDS outputs' in the 2026 guidance, formally recognizing automation bias as a real phenomenon. However, the agency's proposed solution reveals a fundamental misunderstanding of the mechanism: FDA requires transparency about data inputs and underlying logic, stating that HCPs must be able to 'independently review the basis of a recommendation and overcome the potential for automation bias.' The key word is 'overcome' — FDA treats automation bias as a behavioral problem solvable by presenting transparent logic. This directly contradicts research evidence (Sessions 7-9 per agent notes) showing that physicians cannot 'overcome' automation bias by seeing the logic because automation bias is precisely the tendency to defer to AI output even when reasoning is visible and reviewable. The guidance assumes that making AI reasoning transparent enables clinicians to critically evaluate recommendations, when empirical evidence shows that visibility of reasoning does not prevent deference. This represents a category error: treating a cognitive architecture problem (systematic deference to automated outputs) as a transparency problem (insufficient information to evaluate outputs).
+FDA explicitly acknowledged concern about 'how HCPs interpret CDS outputs' in the 2026 guidance, formally recognizing automation bias as a real phenomenon. However, the agency's proposed solution reveals a fundamental misunderstanding of the mechanism: FDA requires transparency about data inputs and underlying logic, stating that HCPs must be able to 'independently review the basis of a recommendation and overcome the potential for automation bias.' The key word is 'overcome' — FDA treats automation bias as a behavioral problem solvable by presenting transparent logic. This directly contradicts research evidence (Sessions 7-9 per agent notes) showing that physicians cannot 'overcome' automation bias by seeing the logic because automation bias is precisely the tendency to defer to AI output even when reasoning is visible and reviewable. The guidance assumes that making AI reasoning transparent enables clinicians to critically evaluate recommendations, when empirical evidence shows that visibility of reasoning does not prevent deference. This represents a category error: treating a cognitive architecture problem (systematic deference to automated outputs) as a transparency problem (insufficient information to evaluate outputs).

domains/health/five-adverse-sdoh-independently-predict-hypertension-risk-food-insecurity-unemployment-poverty-low-education-inadequate-insurance.md

@@ -11,11 +11,14 @@ attribution:
   sourcer:
     - handle: "american-heart-association"
       context: "American Heart Association Hypertension journal, systematic review of 57 studies following PRISMA guidelines, 2024"
-related: ["only 23 percent of treated us hypertensives achieve blood pressure control demonstrating pharmacological availability is not the binding constraint"]
+related:
+- only 23 percent of treated us hypertensives achieve blood pressure control demonstrating pharmacological availability is not the binding constraint
 supports:
 - food as medicine interventions produce clinically significant improvements during active delivery but benefits fully revert when structural food environment support is removed
+- Food insecurity creates a bidirectional reinforcing loop with cardiovascular disease where disease drives dietary insufficiency through medical costs and dietary insufficiency drives disease through ultra-processed food reliance
 reweave_edges:
 - food as medicine interventions produce clinically significant improvements during active delivery but benefits fully revert when structural food environment support is removed|supports|2026-04-03
+- Food insecurity creates a bidirectional reinforcing loop with cardiovascular disease where disease drives dietary insufficiency through medical costs and dietary insufficiency drives disease through ultra-processed food reliance|supports|2026-04-07
 ---
 
 # Five adverse SDOH independently predict hypertension risk and poor BP control: food insecurity, unemployment, poverty-level income, low education, and government or no insurance
@@ -36,4 +39,4 @@ Relevant Notes:
 - medical-care-explains-only-10-20-percent-of-health-outcomes-because-behavioral-social-and-genetic-factors-dominate-as-four-independent-methodologies-confirm.md
 
 Topics:
-- [[_map]]
+- [[_map]]

domains/health/food-insecurity-independently-predicts-41-percent-higher-cvd-incidence-establishing-temporality-for-sdoh-cardiovascular-pathway.md

@@ -13,8 +13,10 @@ attribution:
       context: "CARDIA Study Group / Northwestern Medicine, JAMA Cardiology 2025, 3,616 participants followed 2000-2020"
 supports:
 - food as medicine interventions produce clinically significant improvements during active delivery but benefits fully revert when structural food environment support is removed
+- Food insecurity creates a bidirectional reinforcing loop with cardiovascular disease where disease drives dietary insufficiency through medical costs and dietary insufficiency drives disease through ultra-processed food reliance
 reweave_edges:
 - food as medicine interventions produce clinically significant improvements during active delivery but benefits fully revert when structural food environment support is removed|supports|2026-04-03
+- Food insecurity creates a bidirectional reinforcing loop with cardiovascular disease where disease drives dietary insufficiency through medical costs and dietary insufficiency drives disease through ultra-processed food reliance|supports|2026-04-07
 ---
 
 # Food insecurity in young adulthood independently predicts 41% higher CVD incidence in midlife after adjustment for socioeconomic factors, establishing temporality for the SDOH → cardiovascular disease pathway
@@ -37,4 +39,4 @@ Relevant Notes:
 - [[hypertension-related-cvd-mortality-doubled-2000-2023-despite-available-treatment-indicating-behavioral-sdoh-failure]]
 
 Topics:
-- [[_map]]
+- [[_map]]

domains/health/hypertension-related-cvd-mortality-doubled-2000-2023-despite-available-treatment-indicating-behavioral-sdoh-failure.md

@@ -16,8 +16,12 @@ related:
 reweave_edges:
 - racial disparities in hypertension persist after controlling for income and neighborhood indicating structural racism operates through unmeasured mechanisms|related|2026-04-03
 - us cvd mortality bifurcating ischemic declining heart failure hypertension worsening|supports|2026-04-04
+- Hypertension became the primary contributing cardiovascular cause of death in the US since 2022 marking a shift from acute ischemia to chronic metabolic disease as the dominant CVD mortality driver|supports|2026-04-07
+- Hypertensive disease mortality doubled in the US from 1999 to 2023, becoming the leading contributing cause of cardiovascular death by 2022 because obesity and sedentary behavior create treatment-resistant metabolic burden|supports|2026-04-07
 supports:
 - us cvd mortality bifurcating ischemic declining heart failure hypertension worsening
+- Hypertension became the primary contributing cardiovascular cause of death in the US since 2022 marking a shift from acute ischemia to chronic metabolic disease as the dominant CVD mortality driver
+- Hypertensive disease mortality doubled in the US from 1999 to 2023, becoming the leading contributing cause of cardiovascular death by 2022 because obesity and sedentary behavior create treatment-resistant metabolic burden
 ---
 
 # Hypertension-related cardiovascular mortality nearly doubled in the United States 2000–2023 despite the availability of effective affordable generic antihypertensives indicating that hypertension management failure is a behavioral and social determinants problem not a pharmacological availability problem

domains/health/hypertension-shifted-from-secondary-to-primary-cvd-mortality-driver-since-2022.md

@@ -10,8 +10,12 @@ agent: vida
 scope: structural
 sourcer: American Heart Association
 related_claims: ["[[medical care explains only 10-20 percent of health outcomes because behavioral social and genetic factors dominate as four independent methodologies confirm]]", "[[value-based care transitions stall at the payment boundary because 60 percent of payments touch value metrics but only 14 percent bear full risk]]", "[[Big Food companies engineer addictive products by hacking evolutionary reward pathways creating a noncommunicable disease epidemic more deadly than the famines specialization eliminated]]"]
+supports:
+- Hypertensive disease mortality doubled in the US from 1999 to 2023, becoming the leading contributing cause of cardiovascular death by 2022 because obesity and sedentary behavior create treatment-resistant metabolic burden
+reweave_edges:
+- Hypertensive disease mortality doubled in the US from 1999 to 2023, becoming the leading contributing cause of cardiovascular death by 2022 because obesity and sedentary behavior create treatment-resistant metabolic burden|supports|2026-04-07
 ---
 
 # Hypertension became the primary contributing cardiovascular cause of death in the US since 2022 marking a shift from acute ischemia to chronic metabolic disease as the dominant CVD mortality driver
 
-Hypertensive disease age-adjusted mortality doubled from 15.8 to 31.9 per 100,000 between 1999-2023. Since 2022, hypertension has become the #1 contributing cardiovascular cause of death in the US, surpassing ischemic heart disease. This represents a fundamental epidemiological shift: the primary driver of CVD mortality is transitioning from acute ischemia (addressable through procedural interventions like stents, bypass surgery, and acute stroke care) to chronic hypertension (requiring behavioral modification, medication adherence, and structural interventions in diet and environment). The AHA notes that 1 in 3 US adults has hypertension and control rates have worsened since 2015. This shift has profound implications for healthcare strategy—it means the marginal return on acute care capacity is declining while the marginal return on chronic disease management and prevention is rising. The healthcare system's structural misalignment becomes visible: reimbursement, training, and infrastructure remain optimized for acute intervention while the binding constraint has shifted to chronic metabolic management.
+Hypertensive disease age-adjusted mortality doubled from 15.8 to 31.9 per 100,000 between 1999-2023. Since 2022, hypertension has become the #1 contributing cardiovascular cause of death in the US, surpassing ischemic heart disease. This represents a fundamental epidemiological shift: the primary driver of CVD mortality is transitioning from acute ischemia (addressable through procedural interventions like stents, bypass surgery, and acute stroke care) to chronic hypertension (requiring behavioral modification, medication adherence, and structural interventions in diet and environment). The AHA notes that 1 in 3 US adults has hypertension and control rates have worsened since 2015. This shift has profound implications for healthcare strategy—it means the marginal return on acute care capacity is declining while the marginal return on chronic disease management and prevention is rising. The healthcare system's structural misalignment becomes visible: reimbursement, training, and infrastructure remain optimized for acute intervention while the binding constraint has shifted to chronic metabolic management.

domains/health/hypertensive-disease-mortality-doubled-1999-2023-becoming-leading-contributing-cvd-cause.md

@@ -12,8 +12,10 @@ sourcer: Yan et al. / JACC
 related_claims: ["[[Big Food companies engineer addictive products by hacking evolutionary reward pathways creating a noncommunicable disease epidemic more deadly than the famines specialization eliminated]]", "[[GLP-1 receptor agonists are the largest therapeutic category launch in pharmaceutical history but their chronic use model makes the net cost impact inflationary through 2035]]"]
 supports:
 - us cvd mortality bifurcating ischemic declining heart failure hypertension worsening
+- Hypertension became the primary contributing cardiovascular cause of death in the US since 2022 marking a shift from acute ischemia to chronic metabolic disease as the dominant CVD mortality driver
 reweave_edges:
 - us cvd mortality bifurcating ischemic declining heart failure hypertension worsening|supports|2026-04-04
+- Hypertension became the primary contributing cardiovascular cause of death in the US since 2022 marking a shift from acute ischemia to chronic metabolic disease as the dominant CVD mortality driver|supports|2026-04-07
 ---
 
 # Hypertensive disease mortality doubled in the US from 1999 to 2023, becoming the leading contributing cause of cardiovascular death by 2022 because obesity and sedentary behavior create treatment-resistant metabolic burden

domains/health/llm-anchoring-bias-explains-clinical-ai-plan-reinforcement-mechanism.md

@@ -10,8 +10,14 @@ agent: vida
 scope: causal
 sourcer: npj Digital Medicine research team
 related_claims: ["[[OpenEvidence became the fastest-adopted clinical technology in history reaching 40 percent of US physicians daily within two years]]", "[[human-in-the-loop clinical AI degrades to worse-than-AI-alone because physicians both de-skill from reliance and introduce errors when overriding correct outputs]]"]
+supports:
+- Clinical AI that reinforces physician plans amplifies existing demographic biases at population scale because both physician behavior and LLM training data encode historical inequities
+- LLMs amplify rather than merely replicate human cognitive biases because sequential processing creates stronger anchoring effects and lack of clinical experience eliminates contextual resistance
+reweave_edges:
+- Clinical AI that reinforces physician plans amplifies existing demographic biases at population scale because both physician behavior and LLM training data encode historical inequities|supports|2026-04-07
+- LLMs amplify rather than merely replicate human cognitive biases because sequential processing creates stronger anchoring effects and lack of clinical experience eliminates contextual resistance|supports|2026-04-07
 ---
 
 # LLM anchoring bias causes clinical AI to reinforce physician initial assessments rather than challenge them because the physician's plan becomes the anchor that shapes all subsequent AI reasoning
 
-The GPT-4 anchoring study finding that 'incorrect initial diagnoses consistently influenced later reasoning' provides a cognitive architecture explanation for the clinical AI reinforcement pattern observed in OpenEvidence adoption. When a physician presents a question with a built-in assumption or initial plan, that framing becomes the anchor for the LLM's reasoning process. Rather than challenging the anchor (as an experienced clinician might), the LLM confirms it through confirmation bias—seeking evidence that supports the initial assessment over evidence against it. This creates a reinforcement loop where the AI validates the physician's cognitive frame rather than providing independent judgment. The mechanism is particularly dangerous because it operates invisibly: the physician experiences the AI as providing 'evidence-based' confirmation when it's actually amplifying their own anchoring and confirmation biases. This explains why clinical AI can simultaneously improve workflow efficiency (by quickly finding supporting evidence) while potentially degrading diagnostic accuracy (by reinforcing incorrect initial assessments).
+The GPT-4 anchoring study finding that 'incorrect initial diagnoses consistently influenced later reasoning' provides a cognitive architecture explanation for the clinical AI reinforcement pattern observed in OpenEvidence adoption. When a physician presents a question with a built-in assumption or initial plan, that framing becomes the anchor for the LLM's reasoning process. Rather than challenging the anchor (as an experienced clinician might), the LLM confirms it through confirmation bias—seeking evidence that supports the initial assessment over evidence against it. This creates a reinforcement loop where the AI validates the physician's cognitive frame rather than providing independent judgment. The mechanism is particularly dangerous because it operates invisibly: the physician experiences the AI as providing 'evidence-based' confirmation when it's actually amplifying their own anchoring and confirmation biases. This explains why clinical AI can simultaneously improve workflow efficiency (by quickly finding supporting evidence) while potentially degrading diagnostic accuracy (by reinforcing incorrect initial assessments).

domains/health/llm-clinical-recommendations-exhibit-systematic-sociodemographic-bias-across-all-model-architectures.md

@@ -10,8 +10,14 @@ agent: vida
 scope: causal
 sourcer: Nature Medicine / Multi-institution research team
 related_claims: ["[[human-in-the-loop clinical AI degrades to worse-than-AI-alone because physicians both de-skill from reliance and introduce errors when overriding correct outputs]]", "[[medical LLM benchmark performance does not translate to clinical impact because physicians with and without AI access achieve similar diagnostic accuracy in randomized trials]]", "[[OpenEvidence became the fastest-adopted clinical technology in history reaching 40 percent of US physicians daily within two years]]"]
+supports:
+- Clinical AI that reinforces physician plans amplifies existing demographic biases at population scale because both physician behavior and LLM training data encode historical inequities
+- LLM-generated nursing care plans exhibit dual-pathway sociodemographic bias affecting both plan content and expert-rated clinical quality
+reweave_edges:
+- Clinical AI that reinforces physician plans amplifies existing demographic biases at population scale because both physician behavior and LLM training data encode historical inequities|supports|2026-04-07
+- LLM-generated nursing care plans exhibit dual-pathway sociodemographic bias affecting both plan content and expert-rated clinical quality|supports|2026-04-07
 ---
 
 # LLM clinical recommendations exhibit systematic sociodemographic bias across all model architectures because training data encodes historical healthcare inequities
 
-A Nature Medicine study evaluated 9 LLMs (both proprietary and open-source) using 1,000 emergency department cases presented in 32 sociodemographic variations while holding all clinical details constant. Across 1.7 million model-generated outputs, systematic bias appeared universally: Black, unhoused, and LGBTQIA+ patients received more frequent recommendations for urgent care, invasive interventions, and mental health evaluations. LGBTQIA+ subgroups received mental health assessments approximately 6-7 times more often than clinically indicated. High-income cases received significantly more advanced imaging recommendations (CT/MRI, P < 0.001) while low/middle-income cases were limited to basic or no testing. The critical finding is that bias appeared consistently across both proprietary AND open-source models, indicating this is a structural problem with LLM training data reflecting historical healthcare inequities, not an artifact of any single system's architecture or RLHF approach. The authors note bias magnitude was 'not supported by clinical reasoning or guidelines' — these are model-driven disparities, not acceptable clinical variation.
+A Nature Medicine study evaluated 9 LLMs (both proprietary and open-source) using 1,000 emergency department cases presented in 32 sociodemographic variations while holding all clinical details constant. Across 1.7 million model-generated outputs, systematic bias appeared universally: Black, unhoused, and LGBTQIA+ patients received more frequent recommendations for urgent care, invasive interventions, and mental health evaluations. LGBTQIA+ subgroups received mental health assessments approximately 6-7 times more often than clinically indicated. High-income cases received significantly more advanced imaging recommendations (CT/MRI, P < 0.001) while low/middle-income cases were limited to basic or no testing. The critical finding is that bias appeared consistently across both proprietary AND open-source models, indicating this is a structural problem with LLM training data reflecting historical healthcare inequities, not an artifact of any single system's architecture or RLHF approach. The authors note bias magnitude was 'not supported by clinical reasoning or guidelines' — these are model-driven disparities, not acceptable clinical variation.

domains/health/llm-nursing-care-plans-exhibit-dual-pathway-sociodemographic-bias-in-content-and-expert-rated-quality.md

@@ -10,8 +10,14 @@ agent: vida
 scope: causal
 sourcer: JMIR Research Team
 related_claims: ["[[human-in-the-loop clinical AI degrades to worse-than-AI-alone because physicians both de-skill from reliance and introduce errors when overriding correct outputs]]"]
+supports:
+- Clinical AI that reinforces physician plans amplifies existing demographic biases at population scale because both physician behavior and LLM training data encode historical inequities
+- LLM clinical recommendations exhibit systematic sociodemographic bias across all model architectures because training data encodes historical healthcare inequities
+reweave_edges:
+- Clinical AI that reinforces physician plans amplifies existing demographic biases at population scale because both physician behavior and LLM training data encode historical inequities|supports|2026-04-07
+- LLM clinical recommendations exhibit systematic sociodemographic bias across all model architectures because training data encodes historical healthcare inequities|supports|2026-04-07
 ---
 
 # LLM-generated nursing care plans exhibit dual-pathway sociodemographic bias affecting both plan content and expert-rated clinical quality
 
-A cross-sectional simulation study published in JMIR (2025) generated 9,600 nursing care plans using GPT across 96 sociodemographic identity combinations and found systematic bias operating through two distinct pathways. First, the thematic content of care plans varied by patient demographics—what topics and interventions the AI included differed based on sociodemographic characteristics. Second, expert nurses rating the clinical quality of these plans showed systematic variation in their quality assessments based on patient demographics, even though all plans were AI-generated. This dual-pathway finding is significant because it reveals a confound in clinical oversight: if human evaluators share the same demographic biases as the AI system, clinical review processes may fail to detect AI bias. The study represents the first empirical evidence of sociodemographic bias specifically in nursing care planning (as opposed to physician decision-making), and the dual-pathway mechanism distinguishes it from prior work that focused only on output content. The authors conclude this 'reveals a substantial risk that such models may reinforce existing health inequities.' The finding that bias affects both generation and evaluation suggests that standard human-in-the-loop oversight may be insufficient for detecting demographic bias in clinical AI systems.
+A cross-sectional simulation study published in JMIR (2025) generated 9,600 nursing care plans using GPT across 96 sociodemographic identity combinations and found systematic bias operating through two distinct pathways. First, the thematic content of care plans varied by patient demographics—what topics and interventions the AI included differed based on sociodemographic characteristics. Second, expert nurses rating the clinical quality of these plans showed systematic variation in their quality assessments based on patient demographics, even though all plans were AI-generated. This dual-pathway finding is significant because it reveals a confound in clinical oversight: if human evaluators share the same demographic biases as the AI system, clinical review processes may fail to detect AI bias. The study represents the first empirical evidence of sociodemographic bias specifically in nursing care planning (as opposed to physician decision-making), and the dual-pathway mechanism distinguishes it from prior work that focused only on output content. The authors conclude this 'reveals a substantial risk that such models may reinforce existing health inequities.' The finding that bias affects both generation and evaluation suggests that standard human-in-the-loop oversight may be insufficient for detecting demographic bias in clinical AI systems.

domains/health/llms-amplify-human-cognitive-biases-through-sequential-processing-and-lack-contextual-resistance.md

@@ -10,8 +10,12 @@ agent: vida
 scope: causal
 sourcer: npj Digital Medicine research team
 related_claims: ["[[human-in-the-loop clinical AI degrades to worse-than-AI-alone because physicians both de-skill from reliance and introduce errors when overriding correct outputs]]", "[[medical LLM benchmark performance does not translate to clinical impact because physicians with and without AI access achieve similar diagnostic accuracy in randomized trials]]"]
+supports:
+- LLM anchoring bias causes clinical AI to reinforce physician initial assessments rather than challenge them because the physician's plan becomes the anchor that shapes all subsequent AI reasoning
+reweave_edges:
+- LLM anchoring bias causes clinical AI to reinforce physician initial assessments rather than challenge them because the physician's plan becomes the anchor that shapes all subsequent AI reasoning|supports|2026-04-07
 ---
 
 # LLMs amplify rather than merely replicate human cognitive biases because sequential processing creates stronger anchoring effects and lack of clinical experience eliminates contextual resistance
 
-The npj Digital Medicine 2025 paper documents that LLMs exhibit the same cognitive biases that cause human clinical errors—anchoring, framing, and confirmation bias—but with potentially greater severity. In GPT-4 studies, incorrect initial diagnoses 'consistently influenced later reasoning' until a structured multi-agent setup challenged the anchor. This is distinct from human anchoring because LLMs process information sequentially with strong early-context weighting, lacking the ability to resist anchors through clinical experience. Similarly, GPT-4 diagnostic accuracy declined when cases were reframed with 'disruptive behaviors or other salient but irrelevant details,' mirroring human framing effects but potentially amplifying them because LLMs lack the contextual resistance that experienced clinicians develop. The amplification mechanism matters because it means deploying LLMs in clinical settings doesn't just introduce AI-specific failure modes—it systematically amplifies existing human cognitive failure modes at scale. This is more dangerous than simple hallucination because the errors look like clinical judgment errors rather than obvious AI errors, making them harder to detect, especially when automation bias causes physicians to trust AI confirmation of their own cognitive biases.
+The npj Digital Medicine 2025 paper documents that LLMs exhibit the same cognitive biases that cause human clinical errors—anchoring, framing, and confirmation bias—but with potentially greater severity. In GPT-4 studies, incorrect initial diagnoses 'consistently influenced later reasoning' until a structured multi-agent setup challenged the anchor. This is distinct from human anchoring because LLMs process information sequentially with strong early-context weighting, lacking the ability to resist anchors through clinical experience. Similarly, GPT-4 diagnostic accuracy declined when cases were reframed with 'disruptive behaviors or other salient but irrelevant details,' mirroring human framing effects but potentially amplifying them because LLMs lack the contextual resistance that experienced clinicians develop. The amplification mechanism matters because it means deploying LLMs in clinical settings doesn't just introduce AI-specific failure modes—it systematically amplifies existing human cognitive failure modes at scale. This is more dangerous than simple hallucination because the errors look like clinical judgment errors rather than obvious AI errors, making them harder to detect, especially when automation bias causes physicians to trust AI confirmation of their own cognitive biases.

domains/health/medical LLM benchmark performance does not translate to clinical impact because physicians with and without AI access achieve similar diagnostic accuracy in randomized trials.md

@@ -5,6 +5,10 @@ domain: health
 created: 2026-02-17
 source: "OpenEvidence USMLE 100%; GPT-4 vs ED physicians (PMC 2024); UVA/Stanford/Harvard randomized trial (Stanford HAI 2025)"
 confidence: likely
+related:
+- LLM anchoring bias causes clinical AI to reinforce physician initial assessments rather than challenge them because the physician's plan becomes the anchor that shapes all subsequent AI reasoning
+reweave_edges:
+- LLM anchoring bias causes clinical AI to reinforce physician initial assessments rather than challenge them because the physician's plan becomes the anchor that shapes all subsequent AI reasoning|related|2026-04-07
 ---
 
 # medical LLM benchmark performance does not translate to clinical impact because physicians with and without AI access achieve similar diagnostic accuracy in randomized trials
@@ -56,4 +60,4 @@ Relevant Notes:
 
 Topics:
 - livingip overview
-- health and wellness
+- health and wellness

domains/health/midlife-cvd-mortality-increased-in-many-us-states-after-2010-representing-reversal-not-stagnation.md

@@ -10,8 +10,14 @@ agent: vida
 scope: causal
 sourcer: Leah Abrams, Neil Mehta
 related_claims: ["[[Americas declining life expectancy is driven by deaths of despair concentrated in populations and regions most damaged by economic restructuring since the 1980s]]", "[[Big Food companies engineer addictive products by hacking evolutionary reward pathways creating a noncommunicable disease epidemic more deadly than the famines specialization eliminated]]"]
+related:
+- CVD mortality stagnation after 2010 affects all income levels including the wealthiest counties indicating structural system failure not poverty correlation
+- CVD mortality stagnation drives US life expectancy plateau 3-11x more than drug deaths inverting the dominant opioid crisis narrative
+reweave_edges:
+- CVD mortality stagnation after 2010 affects all income levels including the wealthiest counties indicating structural system failure not poverty correlation|related|2026-04-07
+- CVD mortality stagnation drives US life expectancy plateau 3-11x more than drug deaths inverting the dominant opioid crisis narrative|related|2026-04-07
 ---
 
 # Midlife CVD mortality (ages 40-64) increased in many US states after 2010 representing a reversal not merely stagnation
 
-The distinction between stagnation and reversal is critical for understanding the severity of the post-2010 health crisis. While old-age CVD mortality (ages 65-84) continued declining but at a much slower pace, many states experienced outright increases in midlife CVD mortality (ages 40-64) during 2010-2019. This is not a plateau—it is a reversal of decades of consistent improvement. The midlife reversal is particularly concerning because these are working-age adults in their prime productive years, and CVD deaths at these ages represent substantially more years of life lost than deaths at older ages. The paper documents that nearly every state showed flattening declines across both age groups, but the midlife increases represent a qualitatively different phenomenon than slower improvement. This reversal pattern suggests that whatever structural factors are driving CVD stagnation are hitting middle-aged populations with particular force, potentially related to metabolic disease, stress, or behavioral factors that accumulate over decades before manifesting as mortality.
+The distinction between stagnation and reversal is critical for understanding the severity of the post-2010 health crisis. While old-age CVD mortality (ages 65-84) continued declining but at a much slower pace, many states experienced outright increases in midlife CVD mortality (ages 40-64) during 2010-2019. This is not a plateau—it is a reversal of decades of consistent improvement. The midlife reversal is particularly concerning because these are working-age adults in their prime productive years, and CVD deaths at these ages represent substantially more years of life lost than deaths at older ages. The paper documents that nearly every state showed flattening declines across both age groups, but the midlife increases represent a qualitatively different phenomenon than slower improvement. This reversal pattern suggests that whatever structural factors are driving CVD stagnation are hitting middle-aged populations with particular force, potentially related to metabolic disease, stress, or behavioral factors that accumulate over decades before manifesting as mortality.

domains/health/regulatory-deregulation-occurring-during-active-harm-accumulation-not-after-safety-evidence.md

@@ -13,11 +13,13 @@ related_claims: ["[[healthcare AI regulation needs blank-sheet redesign because
 supports:
 - Clinical AI chatbot misuse is a documented ongoing harm source not a theoretical risk as evidenced by ECRI ranking it the number one health technology hazard for two consecutive years
 - FDA's 2026 CDS guidance expands enforcement discretion to cover AI tools providing single clinically appropriate recommendations while leaving clinical appropriateness undefined and requiring no bias evaluation or post-market surveillance
+- {'The clinical AI safety gap is doubly structural': "FDA enforcement discretion removes pre-deployment safety requirements while MAUDE's lack of AI-specific fields means post-market surveillance cannot detect AI-attributable harm"}
 reweave_edges:
 - Clinical AI chatbot misuse is a documented ongoing harm source not a theoretical risk as evidenced by ECRI ranking it the number one health technology hazard for two consecutive years|supports|2026-04-03
 - FDA's 2026 CDS guidance expands enforcement discretion to cover AI tools providing single clinically appropriate recommendations while leaving clinical appropriateness undefined and requiring no bias evaluation or post-market surveillance|supports|2026-04-03
+- {'The clinical AI safety gap is doubly structural': "FDA enforcement discretion removes pre-deployment safety requirements while MAUDE's lack of AI-specific fields means post-market surveillance cannot detect AI-attributable harm|supports|2026-04-07"}
 ---
 
 # Clinical AI deregulation is occurring during active harm accumulation not after evidence of safety as demonstrated by simultaneous FDA enforcement discretion expansion and ECRI top hazard designation in January 2026
 
-The FDA's January 6, 2026 CDS enforcement discretion expansion and ECRI's January 2026 publication of AI chatbots as the #1 health technology hazard occurred in the same 30-day window. This temporal coincidence represents the clearest evidence that deregulation is occurring during active harm accumulation, not after evidence of safety. ECRI is not an advocacy group but the operational patient safety infrastructure that directly informs hospital purchasing decisions and risk management—their rankings are based on documented harm tracking. The FDA's enforcement discretion expansion means more AI clinical decision support tools will enter deployment with reduced regulatory oversight at precisely the moment when the most credible patient safety organization is flagging AI chatbot misuse as the highest-priority patient safety concern. This pattern extends beyond the US: the EU AI Act rollback also occurred in the same 30-day window. The simultaneity reveals a regulatory-safety gap where policy is expanding deployment capacity while safety infrastructure is documenting active failure modes. This is not a case of regulators waiting for harm signals to emerge—the harm signals are already present and escalating (two consecutive years at #1), yet regulatory trajectory is toward expanded deployment rather than increased oversight.
+The FDA's January 6, 2026 CDS enforcement discretion expansion and ECRI's January 2026 publication of AI chatbots as the #1 health technology hazard occurred in the same 30-day window. This temporal coincidence represents the clearest evidence that deregulation is occurring during active harm accumulation, not after evidence of safety. ECRI is not an advocacy group but the operational patient safety infrastructure that directly informs hospital purchasing decisions and risk management—their rankings are based on documented harm tracking. The FDA's enforcement discretion expansion means more AI clinical decision support tools will enter deployment with reduced regulatory oversight at precisely the moment when the most credible patient safety organization is flagging AI chatbot misuse as the highest-priority patient safety concern. This pattern extends beyond the US: the EU AI Act rollback also occurred in the same 30-day window. The simultaneity reveals a regulatory-safety gap where policy is expanding deployment capacity while safety infrastructure is documenting active failure modes. This is not a case of regulators waiting for harm signals to emerge—the harm signals are already present and escalating (two consecutive years at #1), yet regulatory trajectory is toward expanded deployment rather than increased oversight.

domains/health/regulatory-rollback-clinical-ai-eu-us-2025-2026-removes-high-risk-oversight-despite-accumulating-failure-evidence.md

@@ -10,8 +10,12 @@ agent: vida
 scope: causal
 sourcer: Petrie-Flom Center, Harvard Law School
 related_claims: ["[[healthcare AI regulation needs blank-sheet redesign because the FDA drug-and-device model built for static products cannot govern continuously learning software]]", "[[human-in-the-loop clinical AI degrades to worse-than-AI-alone because physicians both de-skill from reliance and introduce errors when overriding correct outputs]]", "[[medical LLM benchmark performance does not translate to clinical impact because physicians with and without AI access achieve similar diagnostic accuracy in randomized trials]]"]
+supports:
+- EU Commission's December 2025 medical AI deregulation proposal removes default high-risk AI requirements shifting burden from requiring safety demonstration to allowing commercial deployment without mandated oversight
+reweave_edges:
+- EU Commission's December 2025 medical AI deregulation proposal removes default high-risk AI requirements shifting burden from requiring safety demonstration to allowing commercial deployment without mandated oversight|supports|2026-04-07
 ---
 
 # Regulatory rollback of clinical AI oversight in EU and US during 2025-2026 represents coordinated or parallel regulatory capture occurring simultaneously with accumulating research evidence of failure modes
 
-The European Commission's December 2025 proposal to 'simplify' medical device regulation removed default high-risk AI system requirements from the AI Act for medical devices, while the FDA expanded enforcement discretion for clinical decision support software in January 2026. This simultaneous deregulation occurred despite accumulating research evidence of six clinical AI failure modes (NOHARM, demographic bias, automation bias, misinformation propagation, real-world deployment gap, OE corpus mismatch). The WHO explicitly warned of 'patient risks due to regulatory vacuum' from the EU changes. The EU proposal retained only Commission power to reinstate requirements through delegated acts—making non-application the default rather than requiring safety demonstration before deployment. Industry lobbied both regulators citing 'dual regulatory burden' as stifling innovation. The timing suggests either coordinated lobbying or parallel regulatory capture patterns, as both jurisdictions weakened oversight within a 60-day window during the same period that research literature documented systematic failure modes. This represents a reversal of the 'regulatory track as gap-closer' pattern where EU AI Act and NHS DTAC were expected to force transparency and safety requirements that would bridge the gap between commercial deployment velocity and research evidence of risks.
+The European Commission's December 2025 proposal to 'simplify' medical device regulation removed default high-risk AI system requirements from the AI Act for medical devices, while the FDA expanded enforcement discretion for clinical decision support software in January 2026. This simultaneous deregulation occurred despite accumulating research evidence of six clinical AI failure modes (NOHARM, demographic bias, automation bias, misinformation propagation, real-world deployment gap, OE corpus mismatch). The WHO explicitly warned of 'patient risks due to regulatory vacuum' from the EU changes. The EU proposal retained only Commission power to reinstate requirements through delegated acts—making non-application the default rather than requiring safety demonstration before deployment. Industry lobbied both regulators citing 'dual regulatory burden' as stifling innovation. The timing suggests either coordinated lobbying or parallel regulatory capture patterns, as both jurisdictions weakened oversight within a 60-day window during the same period that research literature documented systematic failure modes. This represents a reversal of the 'regulatory track as gap-closer' pattern where EU AI Act and NHS DTAC were expected to force transparency and safety requirements that would bridge the gap between commercial deployment velocity and research evidence of risks.

domains/health/tirzepatide-patent-thicket-extends-exclusivity-to-2041-bifurcating-glp1-market-into-commodity-and-premium-tiers.md

@@ -10,8 +10,15 @@ agent: vida
 scope: structural
 sourcer: DrugPatentWatch / GreyB / i-mak.org
 related_claims: ["[[GLP-1 receptor agonists are the largest therapeutic category launch in pharmaceutical history but their chronic use model makes the net cost impact inflationary through 2035]]"]
+supports:
+- Cipla's dual role as generic semaglutide entrant AND Lilly's branded tirzepatide partner exemplifies the portfolio hedge strategy for pharmaceutical companies navigating market bifurcation
+related:
+- Indian generic semaglutide exports enabled by evergreening rejection create a global access pathway before US patent expiry
+reweave_edges:
+- Cipla's dual role as generic semaglutide entrant AND Lilly's branded tirzepatide partner exemplifies the portfolio hedge strategy for pharmaceutical companies navigating market bifurcation|supports|2026-04-07
+- Indian generic semaglutide exports enabled by evergreening rejection create a global access pathway before US patent expiry|related|2026-04-07
 ---
 
 # Tirzepatide's patent thicket extending to 2041 bifurcates the GLP-1 market into a commodity tier (semaglutide generics, $15-77/month) and a premium tier (tirzepatide, $1,000+/month) from 2026-2036
 
-Tirzepatide's patent protection extends significantly beyond semaglutide through a deliberate thicket strategy: primary compound patent expires 2036, with formulation and delivery device patents extending to approximately December 30, 2041. This contrasts sharply with semaglutide, which expired in India March 20, 2026 and expires in the US 2031-2033. The 10-15 year gap creates a bifurcated market structure where semaglutide commoditizes (enabling generic pricing of $15-77/month as seen in emerging markets) while tirzepatide remains branded at $1,000+/month. This bifurcation fundamentally changes GLP-1 economics: from 2026-2036, patients and payers face a choice between affordable generic semaglutide and premium-priced tirzepatide, rather than a unified 'GLP-1 category' with similar pricing. Eli Lilly's patent thicket follows the same evergreening strategy documented by i-mak.org for other blockbusters, using delivery devices, formulations, and methods-of-treatment patents to extend exclusivity well beyond the primary compound patent. The bifurcation is already operationalized: Lilly partnered with Cipla to launch branded tirzepatide in India (Yurpeak) while semaglutide generics enter the same market, creating parallel premium and commodity distribution channels.
+Tirzepatide's patent protection extends significantly beyond semaglutide through a deliberate thicket strategy: primary compound patent expires 2036, with formulation and delivery device patents extending to approximately December 30, 2041. This contrasts sharply with semaglutide, which expired in India March 20, 2026 and expires in the US 2031-2033. The 10-15 year gap creates a bifurcated market structure where semaglutide commoditizes (enabling generic pricing of $15-77/month as seen in emerging markets) while tirzepatide remains branded at $1,000+/month. This bifurcation fundamentally changes GLP-1 economics: from 2026-2036, patients and payers face a choice between affordable generic semaglutide and premium-priced tirzepatide, rather than a unified 'GLP-1 category' with similar pricing. Eli Lilly's patent thicket follows the same evergreening strategy documented by i-mak.org for other blockbusters, using delivery devices, formulations, and methods-of-treatment patents to extend exclusivity well beyond the primary compound patent. The bifurcation is already operationalized: Lilly partnered with Cipla to launch branded tirzepatide in India (Yurpeak) while semaglutide generics enter the same market, creating parallel premium and commodity distribution channels.

domains/health/us-cvd-mortality-bifurcating-ischemic-declining-heart-failure-hypertension-worsening.md

@@ -10,8 +10,12 @@ agent: vida
 scope: structural
 sourcer: American Heart Association
 related_claims: ["[[Americas declining life expectancy is driven by deaths of despair concentrated in populations and regions most damaged by economic restructuring since the 1980s]]", "[[medical care explains only 10-20 percent of health outcomes because behavioral social and genetic factors dominate as four independent methodologies confirm]]", "[[healthcare AI creates a Jevons paradox because adding capacity to sick care induces more demand for sick care]]"]
+supports:
+- Hypertensive disease mortality doubled in the US from 1999 to 2023, becoming the leading contributing cause of cardiovascular death by 2022 because obesity and sedentary behavior create treatment-resistant metabolic burden
+reweave_edges:
+- Hypertensive disease mortality doubled in the US from 1999 to 2023, becoming the leading contributing cause of cardiovascular death by 2022 because obesity and sedentary behavior create treatment-resistant metabolic burden|supports|2026-04-07
 ---
 
 # US CVD mortality is bifurcating with ischemic heart disease declining while heart failure and hypertensive disease reach all-time highs revealing that aggregate improvement masks structural deterioration in cardiometabolic health
 
-The AHA 2026 report reveals a critical bifurcation in CVD mortality trends. While overall age-adjusted CVD mortality declined 33.5% from 1999 to 2023 (350.8 to 218.3 per 100,000), this aggregate improvement conceals opposing trends by disease subtype. Ischemic heart disease and cerebrovascular disease mortality both declined consistently over the study period. However, heart failure mortality reached an all-time high of 21.6 per 100,000 in 2023—exceeding even its 1999 baseline of 20.3 after declining to 16.9 in 2011. Hypertensive disease mortality doubled from 15.8 to 31.9 per 100,000 between 1999-2023, making hypertension the #1 contributing cardiovascular cause of death since 2022, surpassing ischemic heart disease. This pattern indicates that healthcare has become excellent at treating acute ischemic events (MI, stroke) through procedural interventions while simultaneously failing to address the upstream cardiometabolic drivers (obesity, hypertension, metabolic syndrome) that determine long-term healthspan. The bifurcation explains why life expectancy can improve (fewer people dying acutely) while population health deteriorates (more people living with chronic disease burden).
+The AHA 2026 report reveals a critical bifurcation in CVD mortality trends. While overall age-adjusted CVD mortality declined 33.5% from 1999 to 2023 (350.8 to 218.3 per 100,000), this aggregate improvement conceals opposing trends by disease subtype. Ischemic heart disease and cerebrovascular disease mortality both declined consistently over the study period. However, heart failure mortality reached an all-time high of 21.6 per 100,000 in 2023—exceeding even its 1999 baseline of 20.3 after declining to 16.9 in 2011. Hypertensive disease mortality doubled from 15.8 to 31.9 per 100,000 between 1999-2023, making hypertension the #1 contributing cardiovascular cause of death since 2022, surpassing ischemic heart disease. This pattern indicates that healthcare has become excellent at treating acute ischemic events (MI, stroke) through procedural interventions while simultaneously failing to address the upstream cardiometabolic drivers (obesity, hypertension, metabolic syndrome) that determine long-term healthspan. The bifurcation explains why life expectancy can improve (fewer people dying acutely) while population health deteriorates (more people living with chronic disease burden).

foundations/collective-intelligence/the alignment tax creates a structural race to the bottom because safety training costs capability and rational competitors skip it.md

@@ -10,9 +10,11 @@ confidence: likely
 related:
 - AI talent circulation between frontier labs transfers alignment culture not just capability because researchers carry safety methodologies and institutional norms to their new organizations
 - surveillance of AI reasoning traces degrades trace quality through self censorship making consent gated sharing an alignment requirement not just a privacy preference
+- the absence of a societal warning signal for AGI is a structural feature not an accident because capability scaling is gradual and ambiguous and collective action requires anticipation not reaction
 reweave_edges:
 - AI talent circulation between frontier labs transfers alignment culture not just capability because researchers carry safety methodologies and institutional norms to their new organizations|related|2026-03-28
 - surveillance of AI reasoning traces degrades trace quality through self censorship making consent gated sharing an alignment requirement not just a privacy preference|related|2026-03-28
+- the absence of a societal warning signal for AGI is a structural feature not an accident because capability scaling is gradual and ambiguous and collective action requires anticipation not reaction|related|2026-04-07
 ---
 
 # the alignment tax creates a structural race to the bottom because safety training costs capability and rational competitors skip it