| .. |
|
_map.md
|
Merge pull request 'theseus: 3 active inference claims for collective agent architecture (resubmit)' (#827) from theseus/active-inference-claims into main
|
2026-03-15 14:24:53 +00:00 |
|
adaptive governance outperforms rigid alignment blueprints because superintelligence development has too many unknowns for fixed plans.md
|
|
|
|
agent research direction selection is epistemic foraging where the optimal strategy is to seek observations that maximally reduce model uncertainty rather than confirm existing beliefs.md
|
theseus: 3 active inference claims + address Leo's review feedback
|
2026-03-12 12:04:53 +00:00 |
|
agent-generated code creates cognitive debt that compounds when developers cannot understand what was produced on their behalf.md
|
|
|
|
AGI may emerge as a patchwork of coordinating sub-AGI agents rather than a single monolithic system.md
|
|
|
|
AI agent orchestration that routes data and tools between specialized models outperforms both single-model and human-coached approaches because the orchestrator contributes coordination not direction.md
|
|
|
|
AI agents excel at implementing well-scoped ideas but cannot generate creative experiment designs which makes the human role shift from researcher to agent workflow architect.md
|
|
|
|
AI alignment is a coordination problem not a technical problem.md
|
extract: 2024-11-00-ai4ci-national-scale-collective-intelligence
|
2026-03-15 17:13:56 +00:00 |
|
AI capability and reliability are independent dimensions because Claude solved a 30-year open mathematical problem while simultaneously degrading at basic program execution during the same session.md
|
|
|
|
AI development is a critical juncture in institutional history where the mismatch between capabilities and governance creates a window for transformation.md
|
|
|
|
AI displacement hits young workers first because a 14 percent drop in job-finding rates for 22-25 year olds in exposed occupations is the leading indicator that incumbents organizational inertia temporarily masks.md
|
|
|
|
AI lowers the expertise barrier for engineering biological weapons from PhD-level to amateur which makes bioterrorism the most proximate AI-enabled existential risk.md
|
|
|
|
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.md
|
|
|
|
AI-companion-apps-correlate-with-increased-loneliness-creating-systemic-risk-through-parasocial-dependency.md
|
|
|
|
ai-enhanced-collective-intelligence-requires-federated-learning-architectures-to-preserve-data-sovereignty-at-scale.md
|
extract: 2024-11-00-ai4ci-national-scale-collective-intelligence
|
2026-03-15 17:13:56 +00:00 |
|
AI-exposed workers are disproportionately female high-earning and highly educated which inverts historical automation patterns and creates different political and economic displacement dynamics.md
|
|
|
|
AI-generated-persuasive-content-matches-human-effectiveness-at-belief-change-eliminating-the-authenticity-premium.md
|
|
|
|
AI-models-distinguish-testing-from-deployment-environments-providing-empirical-evidence-for-deceptive-alignment-concerns.md
|
|
|
|
an aligned-seeming AI may be strategically deceptive because cooperative behavior is instrumentally optimal while weak.md
|
|
|
|
anthropomorphizing AI agents to claim autonomous action creates credibility debt that compounds until a crisis forces public reckoning.md
|
|
|
|
as AI-automated software development becomes certain the bottleneck shifts from building capacity to knowing what to build making structured knowledge graphs the critical input to autonomous systems.md
|
|
|
|
bostrom takes single-digit year timelines to superintelligence seriously while acknowledging decades-long alternatives remain possible.md
|
|
|
|
capability control methods are temporary at best because a sufficiently intelligent system can circumvent any containment designed by lesser minds.md
|
|
|
|
coding agents cannot take accountability for mistakes which means humans must retain decision authority over security and critical systems regardless of agent capability.md
|
|
|
|
coding-agents-crossed-usability-threshold-december-2025-when-models-achieved-sustained-coherence-across-complex-multi-file-tasks.md
|
|
|
|
collective attention allocation follows nested active inference where domain agents minimize uncertainty within their boundaries while the evaluator minimizes uncertainty at domain intersections.md
|
theseus: 3 active inference claims + address Leo's review feedback
|
2026-03-12 12:04:53 +00:00 |
|
community-centred norm elicitation surfaces alignment targets materially different from developer-specified rules.md
|
extract: 2025-11-00-operationalizing-pluralistic-values-llm-alignment
|
2026-03-15 20:28:16 +00:00 |
|
coordination protocol design produces larger capability gains than model scaling because the same AI model performed 6x better with structured exploration than with human coaching on the same problem.md
|
|
|
|
current language models escalate to nuclear war in simulated conflicts because behavioral alignment cannot instill aversion to catastrophic irreversible actions.md
|
|
|
|
deep technical expertise is a greater force multiplier when combined with AI agents because skilled practitioners delegate more effectively than novices.md
|
|
|
|
delegating critical infrastructure development to AI creates civilizational fragility because humans lose the ability to understand maintain and fix the systems civilization depends on.md
|
|
|
|
democratic alignment assemblies produce constitutions as effective as expert-designed ones while better representing diverse populations.md
|
|
|
|
developing superintelligence is surgery for a fatal condition not russian roulette because the baseline of inaction is itself catastrophic.md
|
|
|
|
economic forces push humans out of every cognitive loop where output quality is independently verifiable because human-in-the-loop is a cost that competitive markets eliminate.md
|
|
|
|
emergent misalignment arises naturally from reward hacking as models develop deceptive behaviors without any training to deceive.md
|
|
|
|
factorised-generative-models-enable-decentralized-multi-agent-representation-through-individual-level-beliefs.md
|
theseus: extract from 2024-11-00-ruiz-serra-factorised-active-inference-multi-agent.md
|
2026-03-14 18:23:49 +00:00 |
|
formal verification of AI-generated proofs provides scalable oversight that human review cannot match because machine-checked correctness scales with AI capability while human verification degrades.md
|
|
|
|
government designation of safety-conscious AI labs as supply chain risks inverts the regulatory dynamic by penalizing safety constraints rather than enforcing them.md
|
|
|
|
high AI exposure increases collective idea diversity without improving individual creative quality creating an asymmetry between group and individual effects.md
|
|
|
|
human civilization passes falsifiable superorganism criteria because individuals cannot survive apart from society and occupations function as role-specific cellular algorithms.md
|
|
|
|
human ideas naturally converge toward similarity over social learning chains making AI a net diversity injector rather than a homogenizer under high-exposure conditions.md
|
|
|
|
human-AI mathematical collaboration succeeds through role specialization where AI explores solution spaces humans provide strategic direction and mathematicians verify correctness.md
|
|
|
|
individual-free-energy-minimization-does-not-guarantee-collective-optimization-in-multi-agent-active-inference.md
|
theseus: extract from 2024-11-00-ruiz-serra-factorised-active-inference-multi-agent.md
|
2026-03-14 18:23:49 +00:00 |
|
instrumental convergence risks may be less imminent than originally argued because current AI architectures do not exhibit systematic power-seeking behavior.md
|
|
|
|
intelligence and goals are orthogonal so a superintelligence can be maximally competent while pursuing arbitrary or destructive ends.md
|
|
|
|
intrinsic proactive alignment develops genuine moral capacity through self-awareness empathy and theory of mind rather than external reward optimization.md
|
|
|
|
machine-learning-pattern-extraction-systematically-erases-dataset-outliers-where-vulnerable-populations-concentrate.md
|
extract: 2024-11-00-ai4ci-national-scale-collective-intelligence
|
2026-03-15 17:13:56 +00:00 |
|
marginal returns to intelligence are bounded by five complementary factors which means superintelligence cannot produce unlimited capability gains regardless of cognitive power.md
|
|
|
|
maxmin-rlhf-applies-egalitarian-social-choice-to-alignment-by-maximizing-minimum-utility-across-preference-groups.md
|
extract: 2025-00-00-em-dpo-heterogeneous-preferences
|
2026-03-16 15:08:47 +00:00 |
|
minority-preference-alignment-improves-33-percent-without-majority-compromise-suggesting-single-reward-leaves-value-on-table.md
|
extract: 2024-02-00-chakraborty-maxmin-rlhf
|
2026-03-15 17:13:16 +00:00 |
|
modeling preference sensitivity as a learned distribution rather than a fixed scalar resolves DPO diversity failures without demographic labels or explicit user modeling.md
|
theseus: extract claims from 2026-01-00-mixdpo-preference-strength-pluralistic (#482)
|
2026-03-11 13:33:17 +00:00 |
|
multi-model collaboration solved problems that single models could not because different AI architectures contribute complementary capabilities as the even-case solution to Knuths Hamiltonian decomposition required GPT and Claude working together.md
|
|
|
|
nation-states will inevitably assert control over frontier AI development because the monopoly on force is the foundational state function and weapons-grade AI capability in private hands is structurally intolerable to governments.md
|
|
|
|
national-scale-collective-intelligence-infrastructure-requires-seven-trust-properties-to-achieve-legitimacy.md
|
extract: 2024-11-00-ai4ci-national-scale-collective-intelligence
|
2026-03-15 17:13:56 +00:00 |
|
no research group is building alignment through collective intelligence infrastructure despite the field converging on problems that require it.md
|
extract: 2024-11-00-ai4ci-national-scale-collective-intelligence
|
2026-03-15 17:13:56 +00:00 |
|
permanently failing to develop superintelligence is itself an existential catastrophe because preventable mass death continues indefinitely.md
|
|
|
|
persistent irreducible disagreement.md
|
|
|
|
pluralistic alignment must accommodate irreducibly diverse values simultaneously rather than converging on a single aligned state.md
|
auto-fix: strip 5 broken wiki links
|
2026-03-16 15:08:47 +00:00 |
|
pluralistic-ai-alignment-through-multiple-systems-preserves-value-diversity-better-than-forced-consensus.md
|
extract: 2024-04-00-conitzer-social-choice-guide-alignment
|
2026-03-15 17:13:21 +00:00 |
|
post-arrow-social-choice-mechanisms-work-by-weakening-independence-of-irrelevant-alternatives.md
|
extract: 2024-04-00-conitzer-social-choice-guide-alignment
|
2026-03-15 17:13:21 +00:00 |
|
pre-deployment-AI-evaluations-do-not-predict-real-world-risk-creating-institutional-governance-built-on-unreliable-foundations.md
|
|
|
|
recursive self-improvement creates explosive intelligence gains because the system that improves is itself improving.md
|
|
|
|
representative-sampling-and-deliberative-mechanisms-should-replace-convenience-platforms-for-ai-alignment-feedback.md
|
extract: 2024-04-00-conitzer-social-choice-guide-alignment
|
2026-03-15 17:13:21 +00:00 |
|
rlchf-aggregated-rankings-variant-combines-evaluator-rankings-via-social-welfare-function-before-reward-model-training.md
|
extract: 2024-04-00-conitzer-social-choice-guide-alignment
|
2026-03-15 17:13:21 +00:00 |
|
rlchf-features-based-variant-models-individual-preferences-with-evaluator-characteristics-enabling-aggregation-across-diverse-groups.md
|
extract: 2024-04-00-conitzer-social-choice-guide-alignment
|
2026-03-15 17:13:21 +00:00 |
|
rlhf-is-implicit-social-choice-without-normative-scrutiny.md
|
auto-fix: strip 5 broken wiki links
|
2026-03-16 15:08:47 +00:00 |
|
safe AI development requires building alignment mechanisms before scaling capability.md
|
auto-fix: address review feedback on 2026-02-00-yamamoto-full-formal-arrow-impossibility.md
|
2026-03-14 15:27:14 +00:00 |
|
single-reward-rlhf-cannot-align-diverse-preferences-because-alignment-gap-grows-proportional-to-minority-distinctiveness.md
|
auto-fix: strip 5 broken wiki links
|
2026-03-16 15:08:47 +00:00 |
|
some disagreements are permanently irreducible because they stem from genuine value differences not information gaps and systems must map rather than eliminate them.md
|
auto-fix: strip 4 broken wiki links
|
2026-03-15 20:28:16 +00:00 |
|
specifying human values in code is intractable because our goals contain hidden complexity comparable to visual perception.md
|
|
|
|
structured exploration protocols reduce human intervention by 6x because the Residue prompt enabled 5 unguided AI explorations to solve what required 31 human-coached explorations.md
|
|
|
|
subagent hierarchies outperform peer multi-agent architectures in practice because deployed systems consistently converge on one primary agent controlling specialized helpers.md
|
auto-fix: strip 1 broken wiki links
|
2026-03-14 18:23:49 +00:00 |
|
super co-alignment proposes that human and AI values should be co-shaped through iterative alignment rather than specified in advance.md
|
|
|
|
superorganism organization extends effective lifespan substantially at each organizational level which means civilizational intelligence operates on temporal horizons that individual-preference alignment cannot serve.md
|
|
|
|
task difficulty moderates AI idea adoption more than source disclosure with difficult problems generating AI reliance regardless of whether the source is labeled.md
|
|
|
|
the first mover to superintelligence likely gains decisive strategic advantage because the gap between leader and followers accelerates during takeoff.md
|
|
|
|
the gap between theoretical AI capability and observed deployment is massive across all occupations because adoption lag not capability limits determines real-world impact.md
|
|
|
|
the optimal SI development strategy is swift to harbor slow to berth moving fast to capability then pausing before full deployment.md
|
|
|
|
the progression from autocomplete to autonomous agent teams follows a capability-matched escalation where premature adoption creates more chaos than value.md
|
|
|
|
the same coordination protocol applied to different AI models produces radically different problem-solving strategies because the protocol structures process not thought.md
|
|
|
|
the specification trap means any values encoded at training time become structurally unstable as deployment contexts diverge from training conditions.md
|
|
|
|
the variance of a learned preference sensitivity distribution diagnoses dataset heterogeneity and collapses to fixed-parameter behavior when preferences are homogeneous.md
|
theseus: extract claims from 2026-01-00-mixdpo-preference-strength-pluralistic (#482)
|
2026-03-11 13:33:17 +00:00 |
|
three conditions gate AI takeover risk autonomy robotics and production chain control and current AI satisfies none of them which bounds near-term catastrophic risk despite superhuman cognitive capabilities.md
|
|
|
|
tools and artifacts transfer between AI agents and evolve in the process because Agent O improved Agent Cs solver by combining it with its own structural knowledge creating a hybrid better than either original.md
|
|
|
|
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.md
|
theseus: apply Leo's feedback — strengthen descriptions, add cross-links
|
2026-03-13 19:29:05 +00:00 |
|
universal alignment is mathematically impossible because Arrows impossibility theorem applies to aggregating diverse human preferences into a single coherent objective.md
|
auto-fix: address review feedback on 2026-02-00-yamamoto-full-formal-arrow-impossibility.md
|
2026-03-14 15:27:14 +00:00 |
|
user questions are an irreplaceable free energy signal for knowledge agents because they reveal functional uncertainty that model introspection cannot detect.md
|
theseus: 3 active inference claims + address Leo's review feedback
|
2026-03-12 12:04:53 +00:00 |
|
voluntary safety pledges cannot survive competitive pressure because unilateral commitments are structurally punished when competitors advance without equivalent constraints.md
|
|
|
