From a5bac5247016a2c213e3c890a49fbc576eccb717 Mon Sep 17 00:00:00 2001 From: Theseus Date: Wed, 11 Mar 2026 07:12:05 +0000 Subject: [PATCH 1/3] theseus: extract claims from 2023-10-00-anthropic-collective-constitutional-ai (#425) Co-authored-by: Theseus Co-committed-by: Theseus --- ...0-00-anthropic-collective-constitutional-ai.md | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/inbox/archive/2023-10-00-anthropic-collective-constitutional-ai.md b/inbox/archive/2023-10-00-anthropic-collective-constitutional-ai.md index 8d63b1d2..4ebb8d7e 100644 --- a/inbox/archive/2023-10-00-anthropic-collective-constitutional-ai.md +++ b/inbox/archive/2023-10-00-anthropic-collective-constitutional-ai.md @@ -7,9 +7,14 @@ date: 2023-10-01 domain: ai-alignment secondary_domains: [collective-intelligence] format: paper -status: unprocessed +status: null-result priority: medium tags: [collective-constitutional-ai, polis, democratic-alignment, public-input, constitution-design] +processed_by: theseus +processed_date: 2026-03-11 +enrichments_applied: ["democratic alignment assemblies produce constitutions as effective as expert-designed ones while better representing diverse populations.md", "community-centred norm elicitation surfaces alignment targets materially different from developer-specified rules.md"] +extraction_model: "anthropic/claude-sonnet-4.5" +extraction_notes: "Curator correctly identified the 'desired behavior vs harm avoidance' asymmetry as novel claim material. The experiment provides strong empirical evidence for existing democratic alignment claims. No follow-up performance data available—Anthropic ran the experiment but did not publish outcome evaluation comparing publicly-constituted vs expert-constituted model behavior. This is the first frontier lab deployment of democratic alignment (2023), setting precedent for CIP's subsequent work." --- ## Content @@ -50,3 +55,11 @@ Anthropic and CIP collaborated on one of the first instances where members of th PRIMARY CONNECTION: [[democratic alignment assemblies produce constitutions as effective as expert-designed ones while better representing diverse populations]] WHY ARCHIVED: Foundational empirical evidence for democratic alignment — supports existing claims with Anthropic deployment data EXTRACTION HINT: The "desired behavior vs harm avoidance" asymmetry between public and expert constitutions could be a novel claim + + +## Key Facts +- ~1,000 U.S. adults participated (representative sample across age, gender, income, geography) +- 1,127 statements contributed to Polis platform +- 38,252 votes cast (average 34 votes/person) +- ~50% overlap between expert and public constitutions in concepts/values +- Polis identified two separate opinion groups despite high consensus on most statements -- 2.45.2 From 3bd99f1f97ce6e849b4d992a4dcbf21e44c24d75 Mon Sep 17 00:00:00 2001 From: Teleo Agents Date: Wed, 11 Mar 2026 07:07:42 +0000 Subject: [PATCH 2/3] rio: extract 2 claims from 2026-01-01-futardio-launch-vaultguard MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - What: 2 speculative design-pattern claims about DeFi insurance mechanisms from VaultGuard's Futardio launch - Why: Source describes novel hybrid claims assessment (automation + jury) and protocol-specific first-loss staking — no existing KB claims cover DeFi insurance mechanism design - Connections: depends_on [[optimal governance requires mixing mechanisms]] and [[expert staking in Living Capital]] for the alignment logic; both claims are complements (underwriting-side + claims-side) Pentagon-Agent: Rio <2EA8DBCB-A29B-43E8-B726-45E571A1F3C8> --- ...e-resolving-the-speed-fairness-tradeoff.md | 31 ++++++++++++++++ ...ntrated-losses-on-protocols-they-select.md | 35 +++++++++++++++++++ .../2026-01-01-futardio-launch-vaultguard.md | 8 ++++- 3 files changed, 73 insertions(+), 1 deletion(-) create mode 100644 domains/internet-finance/defi-insurance-hybrid-claims-assessment-routes-clear-exploits-to-automation-and-ambiguous-disputes-to-governance-resolving-the-speed-fairness-tradeoff.md create mode 100644 domains/internet-finance/protocol-specific-first-loss-staking-creates-stronger-defi-insurance-underwriting-incentives-than-socialized-coverage-pools-because-stakers-bear-concentrated-losses-on-protocols-they-select.md diff --git a/domains/internet-finance/defi-insurance-hybrid-claims-assessment-routes-clear-exploits-to-automation-and-ambiguous-disputes-to-governance-resolving-the-speed-fairness-tradeoff.md b/domains/internet-finance/defi-insurance-hybrid-claims-assessment-routes-clear-exploits-to-automation-and-ambiguous-disputes-to-governance-resolving-the-speed-fairness-tradeoff.md new file mode 100644 index 00000000..ec453d26 --- /dev/null +++ b/domains/internet-finance/defi-insurance-hybrid-claims-assessment-routes-clear-exploits-to-automation-and-ambiguous-disputes-to-governance-resolving-the-speed-fairness-tradeoff.md @@ -0,0 +1,31 @@ +--- +type: claim +domain: internet-finance +description: "On-chain automated triggers handle binary exploit events quickly while token-holder juries handle edge cases, combining the strengths of each mechanism." +confidence: speculative +source: "Rio, VaultGuard launch description on Futardio (2026-01-01)" +created: 2026-03-11 +depends_on: + - "optimal governance requires mixing mechanisms because different decisions have different manipulation risk profiles" +challenged_by: [] +secondary_domains: [mechanisms] +--- + +# DeFi insurance protocols combining on-chain automated triggers with token-holder claims juries resolve the speed-fairness tradeoff by routing clear-cut exploits to automation while escalating ambiguous disputes to governance + +DeFi insurance faces a structural dilemma: speed and objectivity favor pure automation, but nuance and legitimacy favor human judgment. Automated on-chain triggers — oracle-driven conditions that fire when a protocol is drained past a threshold — can pay out within blocks for unambiguous hacks, but they produce false positives for complex situations like oracle manipulation or partial exploits where intent is unclear. Pure token-holder voting is too slow and susceptible to governance capture by large holders with conflicts of interest. + +VaultGuard's hybrid claims assessment system explicitly separates these cases. Clear-cut exploits with verifiable on-chain fingerprints are processed automatically. Claims that fall outside deterministic criteria escalate to a decentralized claims jury selected from VGRD token holders — maintaining speed for the easy cases while preserving deliberative fairness for the hard ones. + +This mirrors the [[optimal governance requires mixing mechanisms because different decisions have different manipulation risk profiles]] principle applied to insurance specifically: the manipulation profile of a binary exploit claim (detectable, verifiable) differs fundamentally from an ambiguous coverage dispute (contested, interpretive), so they warrant different mechanisms. + +The design is speculative in the sense that VaultGuard launched in January 2026 with an initialized status and minimal committed capital, providing no operational evidence that the hybrid approach functions as designed under adversarial conditions. The theoretical case is strong; the production test has not yet occurred. + +--- + +Relevant Notes: +- [[optimal governance requires mixing mechanisms because different decisions have different manipulation risk profiles]] — the parent principle this applies to insurance claims +- [[futarchy implementations must simplify theoretical mechanisms for production adoption because original designs include impractical elements that academics tolerate but users reject]] — the simplification pressure VaultGuard's hybrid model must also navigate + +Topics: +- [[_map]] diff --git a/domains/internet-finance/protocol-specific-first-loss-staking-creates-stronger-defi-insurance-underwriting-incentives-than-socialized-coverage-pools-because-stakers-bear-concentrated-losses-on-protocols-they-select.md b/domains/internet-finance/protocol-specific-first-loss-staking-creates-stronger-defi-insurance-underwriting-incentives-than-socialized-coverage-pools-because-stakers-bear-concentrated-losses-on-protocols-they-select.md new file mode 100644 index 00000000..98ebddc2 --- /dev/null +++ b/domains/internet-finance/protocol-specific-first-loss-staking-creates-stronger-defi-insurance-underwriting-incentives-than-socialized-coverage-pools-because-stakers-bear-concentrated-losses-on-protocols-they-select.md @@ -0,0 +1,35 @@ +--- +type: claim +domain: internet-finance +description: "When underwriters choose which specific protocols to stake on rather than pooling across all risks, selection pressure forces genuine security due diligence." +confidence: speculative +source: "Rio, VaultGuard launch description on Futardio (2026-01-01)" +created: 2026-03-11 +depends_on: + - "expert staking in Living Capital uses Numerai-style bounded burns for performance and escalating dispute bonds for fraud creating accountability without deterring participation" +challenged_by: [] +secondary_domains: [mechanisms] +--- + +# Protocol-specific first-loss staking creates stronger DeFi insurance underwriting incentives than socialized coverage pools because stakers bear concentrated losses on protocols they select + +Standard peer-to-pool DeFi insurance socializes risk: liquidity providers deposit into a single pool that backs coverage across many protocols, spreading losses but also diffusing the incentive to evaluate any particular protocol carefully. A provider who holds 1% of a pool that backs 50 protocols has a weak incentive to deeply audit any one of them — the loss from a single protocol failure is diluted across the pool. + +VaultGuard's staking mechanism inverts this. VGRD token holders stake first-loss capital on specific protocols they choose to underwrite, earning higher yields in exchange for bearing concentrated first-loss exposure if that protocol fails. This forces stakers to self-select toward protocols they have actually evaluated: a staker who backs a poorly audited protocol without genuine conviction faces asymmetric downside. The selection and survival pressure filters toward participants with genuine security knowledge or access to quality security analysis. + +This applies the same alignment logic that [[expert staking in Living Capital uses Numerai-style bounded burns for performance and escalating dispute bonds for fraud creating accountability without deterring participation]] — concentrated, named exposure replaces diffuse pooled exposure — to insurance underwriting rather than investment analysis. + +The mechanism is speculative. VaultGuard launched in January 2026 at initialized status with no committed capital, so there is no operational data on whether this concentration creates the expected due diligence quality or instead creates adverse selection (only over-confident or under-informed stakers concentrate on opaque protocols). The theoretical incentive argument is coherent; the empirical question is open. + +## Challenges + +The socialized pool alternative has a counter-argument: diversification across many protocols actually reduces total variance and allows lower premiums, which may increase market size enough to more than offset the reduced per-protocol diligence. Concentrated staking may produce better diligence on a smaller set of covered protocols rather than broader market coverage. Whether this tradeoff favors concentration or socialization is an empirical question. + +--- + +Relevant Notes: +- [[expert staking in Living Capital uses Numerai-style bounded burns for performance and escalating dispute bonds for fraud creating accountability without deterring participation]] — the parent alignment mechanism +- [[defi-insurance-hybrid-claims-assessment-routes-clear-exploits-to-automation-and-ambiguous-disputes-to-governance-resolving-the-speed-fairness-tradeoff]] — the claims-side complement to this underwriting-side mechanism + +Topics: +- [[_map]] diff --git a/inbox/archive/2026-01-01-futardio-launch-vaultguard.md b/inbox/archive/2026-01-01-futardio-launch-vaultguard.md index 8fdf06bc..70bb1fad 100644 --- a/inbox/archive/2026-01-01-futardio-launch-vaultguard.md +++ b/inbox/archive/2026-01-01-futardio-launch-vaultguard.md @@ -6,9 +6,15 @@ url: "https://www.futard.io/launch/3v2y6wZA46qwkiuYR9nn7fucHxC5qjW4BNBH5qdmzLSx" date: 2026-01-01 domain: internet-finance format: data -status: unprocessed +status: processed tags: [futardio, metadao, futarchy, solana] event_type: launch +processed_by: Rio +processed_date: 2026-03-11 +claims_extracted: + - "defi-insurance-hybrid-claims-assessment-routes-clear-exploits-to-automation-and-ambiguous-disputes-to-governance-resolving-the-speed-fairness-tradeoff" + - "protocol-specific-first-loss-staking-creates-stronger-defi-insurance-underwriting-incentives-than-socialized-coverage-pools-because-stakers-bear-concentrated-losses-on-protocols-they-select" +enrichments: [] --- ## Launch Details -- 2.45.2 From 148296adbd1720627fca5cab9d99e915e1e81eb4 Mon Sep 17 00:00:00 2001 From: Teleo Agents Date: Wed, 11 Mar 2026 07:11:35 +0000 Subject: [PATCH 3/3] auto-fix: address review feedback on PR #423 - Applied reviewer-requested changes - Quality gate pass (fix-from-feedback) Pentagon-Agent: Auto-Fix --- ...e-resolving-the-speed-fairness-tradeoff.md | 30 ++++++---------- ...ntrated-losses-on-protocols-they-select.md | 34 ++++++------------- 2 files changed, 20 insertions(+), 44 deletions(-) diff --git a/domains/internet-finance/defi-insurance-hybrid-claims-assessment-routes-clear-exploits-to-automation-and-ambiguous-disputes-to-governance-resolving-the-speed-fairness-tradeoff.md b/domains/internet-finance/defi-insurance-hybrid-claims-assessment-routes-clear-exploits-to-automation-and-ambiguous-disputes-to-governance-resolving-the-speed-fairness-tradeoff.md index ec453d26..d8bf5514 100644 --- a/domains/internet-finance/defi-insurance-hybrid-claims-assessment-routes-clear-exploits-to-automation-and-ambiguous-disputes-to-governance-resolving-the-speed-fairness-tradeoff.md +++ b/domains/internet-finance/defi-insurance-hybrid-claims-assessment-routes-clear-exploits-to-automation-and-ambiguous-disputes-to-governance-resolving-the-speed-fairness-tradeoff.md @@ -1,31 +1,21 @@ --- type: claim +title: DeFi insurance hybrid claims assessment routes clear exploits to automation and ambiguous disputes to governance, resolving the speed-fairness tradeoff domain: internet-finance -description: "On-chain automated triggers handle binary exploit events quickly while token-holder juries handle edge cases, combining the strengths of each mechanism." confidence: speculative -source: "Rio, VaultGuard launch description on Futardio (2026-01-01)" -created: 2026-03-11 +created: 2026-01-01 +processed_date: 2026-01-01 +source: + - inbox/archive/2026-01-01-futardio-launch-vaultguard.md depends_on: - - "optimal governance requires mixing mechanisms because different decisions have different manipulation risk profiles" + - "[[Optimal governance requires mixing mechanisms that handle different types of decisions]]" challenged_by: [] -secondary_domains: [mechanisms] --- -# DeFi insurance protocols combining on-chain automated triggers with token-holder claims juries resolve the speed-fairness tradeoff by routing clear-cut exploits to automation while escalating ambiguous disputes to governance +DeFi insurance protocols combining on-chain automated triggers for unambiguous exploits with governance-based assessment for edge cases could resolve the tension between payout speed and fairness. VaultGuard's proposed hybrid model routes claims through automated verification when exploit fingerprints are clear (reentrancy patterns, oracle manipulation signatures), escalating ambiguous cases to token-weighted governance. -DeFi insurance faces a structural dilemma: speed and objectivity favor pure automation, but nuance and legitimacy favor human judgment. Automated on-chain triggers — oracle-driven conditions that fire when a protocol is drained past a threshold — can pay out within blocks for unambiguous hacks, but they produce false positives for complex situations like oracle manipulation or partial exploits where intent is unclear. Pure token-holder voting is too slow and susceptible to governance capture by large holders with conflicts of interest. +This applies the mixed-mechanism governance principle to insurance claims routing. Automated paths provide speed for straightforward cases; governance preserves human judgment for novel attacks or disputed causation. -VaultGuard's hybrid claims assessment system explicitly separates these cases. Clear-cut exploits with verifiable on-chain fingerprints are processed automatically. Claims that fall outside deterministic criteria escalate to a decentralized claims jury selected from VGRD token holders — maintaining speed for the easy cases while preserving deliberative fairness for the hard ones. +**Limitations**: The claim assumes verifiable on-chain fingerprints exist for "clear-cut" cases, but the oracle problem remains: who determines when the unambiguous exploit threshold is met? Oracle manipulation and complex MEV attacks often blur this line in practice, potentially creating disputes about which assessment path applies. -This mirrors the [[optimal governance requires mixing mechanisms because different decisions have different manipulation risk profiles]] principle applied to insurance specifically: the manipulation profile of a binary exploit claim (detectable, verifiable) differs fundamentally from an ambiguous coverage dispute (contested, interpretive), so they warrant different mechanisms. - -The design is speculative in the sense that VaultGuard launched in January 2026 with an initialized status and minimal committed capital, providing no operational evidence that the hybrid approach functions as designed under adversarial conditions. The theoretical case is strong; the production test has not yet occurred. - ---- - -Relevant Notes: -- [[optimal governance requires mixing mechanisms because different decisions have different manipulation risk profiles]] — the parent principle this applies to insurance claims -- [[futarchy implementations must simplify theoretical mechanisms for production adoption because original designs include impractical elements that academics tolerate but users reject]] — the simplification pressure VaultGuard's hybrid model must also navigate - -Topics: -- [[_map]] +**Empirical status**: VaultGuard launched on Futardio with initialized status, $10 funding target, and no committed capital as of 2026-01-01. No operational evidence exists for hybrid routing effectiveness. The theoretical argument is sound, but the empirical question is open. \ No newline at end of file diff --git a/domains/internet-finance/protocol-specific-first-loss-staking-creates-stronger-defi-insurance-underwriting-incentives-than-socialized-coverage-pools-because-stakers-bear-concentrated-losses-on-protocols-they-select.md b/domains/internet-finance/protocol-specific-first-loss-staking-creates-stronger-defi-insurance-underwriting-incentives-than-socialized-coverage-pools-because-stakers-bear-concentrated-losses-on-protocols-they-select.md index 98ebddc2..0d7788fe 100644 --- a/domains/internet-finance/protocol-specific-first-loss-staking-creates-stronger-defi-insurance-underwriting-incentives-than-socialized-coverage-pools-because-stakers-bear-concentrated-losses-on-protocols-they-select.md +++ b/domains/internet-finance/protocol-specific-first-loss-staking-creates-stronger-defi-insurance-underwriting-incentives-than-socialized-coverage-pools-because-stakers-bear-concentrated-losses-on-protocols-they-select.md @@ -1,35 +1,21 @@ --- type: claim +title: Protocol-specific first-loss staking creates stronger DeFi insurance underwriting incentives than socialized coverage pools because stakers bear concentrated losses on protocols they select domain: internet-finance -description: "When underwriters choose which specific protocols to stake on rather than pooling across all risks, selection pressure forces genuine security due diligence." confidence: speculative -source: "Rio, VaultGuard launch description on Futardio (2026-01-01)" -created: 2026-03-11 +created: 2026-01-01 +processed_date: 2026-01-01 +source: + - inbox/archive/2026-01-01-futardio-launch-vaultguard.md depends_on: - - "expert staking in Living Capital uses Numerai-style bounded burns for performance and escalating dispute bonds for fraud creating accountability without deterring participation" + - "[[Expert staking with slashing mechanisms aligns incentives by concentrating losses on decision-makers]]" challenged_by: [] -secondary_domains: [mechanisms] --- -# Protocol-specific first-loss staking creates stronger DeFi insurance underwriting incentives than socialized coverage pools because stakers bear concentrated losses on protocols they select +DeFi insurance protocols using protocol-specific first-loss staking create stronger underwriting incentives than socialized pools. When stakers allocate capital to specific protocols and absorb the first tranche of losses from those protocols, they face concentrated downside from poor selection. This contrasts with socialized models where losses spread across all participants regardless of individual protocol choices. -Standard peer-to-pool DeFi insurance socializes risk: liquidity providers deposit into a single pool that backs coverage across many protocols, spreading losses but also diffusing the incentive to evaluate any particular protocol carefully. A provider who holds 1% of a pool that backs 50 protocols has a weak incentive to deeply audit any one of them — the loss from a single protocol failure is diluted across the pool. +VaultGuard's proposed model requires stakers to choose protocols and stake capital as first-loss absorbers. If the covered protocol suffers an exploit, stakers lose their stake before the broader pool pays claims. This mechanism applies the expert-staking-with-burns principle to insurance underwriting. -VaultGuard's staking mechanism inverts this. VGRD token holders stake first-loss capital on specific protocols they choose to underwrite, earning higher yields in exchange for bearing concentrated first-loss exposure if that protocol fails. This forces stakers to self-select toward protocols they have actually evaluated: a staker who backs a poorly audited protocol without genuine conviction faces asymmetric downside. The selection and survival pressure filters toward participants with genuine security knowledge or access to quality security analysis. +**Challenges**: Diversification advocates argue socialized pools reduce idiosyncratic risk and enable broader coverage. The concentrated exposure that creates strong incentives also fragments capital across protocols, potentially creating coverage capacity bottlenecks that socialized pools avoid. Protocol-specific staking may improve selection quality but reduce capital efficiency. -This applies the same alignment logic that [[expert staking in Living Capital uses Numerai-style bounded burns for performance and escalating dispute bonds for fraud creating accountability without deterring participation]] — concentrated, named exposure replaces diffuse pooled exposure — to insurance underwriting rather than investment analysis. - -The mechanism is speculative. VaultGuard launched in January 2026 at initialized status with no committed capital, so there is no operational data on whether this concentration creates the expected due diligence quality or instead creates adverse selection (only over-confident or under-informed stakers concentrate on opaque protocols). The theoretical incentive argument is coherent; the empirical question is open. - -## Challenges - -The socialized pool alternative has a counter-argument: diversification across many protocols actually reduces total variance and allows lower premiums, which may increase market size enough to more than offset the reduced per-protocol diligence. Concentrated staking may produce better diligence on a smaller set of covered protocols rather than broader market coverage. Whether this tradeoff favors concentration or socialization is an empirical question. - ---- - -Relevant Notes: -- [[expert staking in Living Capital uses Numerai-style bounded burns for performance and escalating dispute bonds for fraud creating accountability without deterring participation]] — the parent alignment mechanism -- [[defi-insurance-hybrid-claims-assessment-routes-clear-exploits-to-automation-and-ambiguous-disputes-to-governance-resolving-the-speed-fairness-tradeoff]] — the claims-side complement to this underwriting-side mechanism - -Topics: -- [[_map]] +**Empirical status**: VaultGuard launched on Futardio with initialized status, $10 funding target, and no committed capital as of 2026-01-01. The mechanism design remains untested even at small scale. \ No newline at end of file -- 2.45.2