Agentic AI Vocabulary

DRAFT · June 2026   W3C AI Knowledge Representation Community Group (AI KR CG)  ·  structured by the Knowledge-Domains taxonomy

Invitation to contribute. You are invited to review these terms and definitions, propose changes, and discuss them on the public mailing list of the AI KR CG: public-aikr@w3.org — archives at lists.w3.org/Archives/Public/public-aikr/. Anyone may read or write to this list. Proposed definitions and relations are first drafts for the group to critique.
Provisional — to be confirmed and maintained. The agentic-AI environment changes rapidly. Every term, definition, source, and relation here is provisional and must be verified and kept current. Acronym expansions marked unconfirmed (e.g. RAILS, GAIE) should not be treated as settled. This is a living draft, not a settled reference.

This vocabulary is organised under the eight categories of the Agentic AI Knowledge-Domains taxonomy, so each term sits in the same category used across the repository. Columns: the proposed definition, the canonical source, and SKOS-style relations (broader / narrower / partOf / related / regulates / mitigates …).

1. Agent types & autonomy (8 terms)

TermProposed definitionSourceRelations
AnalyticalAgentAgent that performs reasoning and data analysis.AI-KR CG agent ontology (AIAO)broader: agent; related: Reasoning
AssistantAgentAgent that supports humans or other agents.AI-KR CG agent ontology (AIAO)broader: agent
AutonomousGoalSeekingAgentAgent that pursues explicit goals independently.AI-KR CG agent ontology (AIAO)broader: agent; related: Planning
CreativeAgentAgent that generates content or solutions.AI-KR CG agent ontology (AIAO)broader: agent
DeliberativeAgentAgent that maintains an internal model and plans actions before acting.AI-KR CG agent ontology (AIAO)broader: agent
LearningAgentAgent that adapts its behaviour over time from experience.AI-KR CG agent ontology (AIAO)broader: agent
ReactiveAgentAgent with simple stimulus-response behaviour; no internal world model.AI-KR CG agent ontology (AIAO)broader: agent
TaskAgentAgent that executes concrete tasks autonomously.AI-KR CG agent ontology (AIAO)broader: agent

2. Cognition & capability (7 terms)

TermProposed definitionSourceRelations
embodimentWhether an agent is software-only, robotic, or hybrid.AI-KR CG agent ontology (AIAO)partOf: agent
LearningCapability to adapt models and update behaviour.AI-KR CG agent ontology (AIAO)partOf: capability
MemoryCapability to store knowledge or history.AI-KR CG agent ontology (AIAO)partOf: capability
operational scopeWhether an agent is single-task, multi-task, or open-ended.AI-KR CG agent ontology (AIAO)partOf: agent
PerceptionCapability to sense environment or inputs.AI-KR CG agent ontology (AIAO)partOf: capability
PlanningCapability to generate action sequences to reach goals.AI-KR CG agent ontology (AIAO)partOf: capability
ReasoningCapability to apply logic and knowledge to decisions.AI-KR CG agent ontology (AIAO)partOf: capability

3. Interaction, communication & identity (12 terms)

TermProposed definitionSourceRelations
A2AAgent-to-Agent. Protocols and frameworks linking independent agents without human intervention.Google (Agent2Agent protocol)related: MCP, AP2
ACP (Agent Communication Protocol)REST-native inter-agent protocol from the AGNTCY collective; maps agent interactions onto HTTP verbs.AGNTCY / Linux Foundationrelated: A2A, MCP
ANP (Agent Network Protocol)Open, DID-based protocol for decentralised agent discovery and collaboration; published as a W3C White Paper.ANP / W3Crelated: DID
DID (Decentralized Identifiers)W3C standard for verifiable, self-sovereign identifiers; basis for agent identity.W3C DID v1.1 (CR)related: Verifiable Credentials
human-in-the-loopInteraction style requiring human guidance, review, or approval of agent actions.AI-KR CG agent ontology (AIAO)related: graduated human oversight
KYAKnow Your Agent. A governance framework for registering, verifying, and profiling autonomous agents before deployment.Emerging governance usagerelated: Sybil resistance, evidence artifacts; analogous: KYC
MCPModel Context Protocol. An open standard for connecting models to external data, tools, and applications.Anthropic (Model Context Protocol)related: A2A, tool use
NANDANetworked AI Agents in a Decentralized Architecture; DID-based global agent discovery/registry (MIT Media Lab).MIT Media Lab (Project NANDA)related: DID, ANP
OASF (Open Agent Schema Framework)Standard schema for describing agent capabilities and interfaces in machine-readable form.AGNTCYrelated: ACP
origin-bound authorityBinding an agent's authorisation to execute a tool to the verified source of the initiating request.Agent-security literaturemitigates: indirect prompt injection
Sybil resistanceA system's capacity to defend against attackers creating many fake identities to gain influence.Douceur (2002)related: KYA; mitigates: fake-identity attacks
Verifiable CredentialsW3C standard for tamper-evident, cryptographically verifiable claims; used for agent credentials.W3C VC 2.0related: DID

4. Commerce & settlement (26 terms)

TermProposed definitionSourceRelations
ACP (Agentic Commerce Protocol)Stripe + OpenAI protocol for agent-initiated checkout; in production in ChatGPT. [distinct from the AGNTCY ACP]Stripe / OpenAI / Metarelated: AP2, x402
agentic clearingThe operational layer that evaluates whether an autonomous AI agent has fulfilled its transaction obligations, assigns liability, and determines the subsequent settlement action — distinct from payment and from settlement execution.de Valois-Franklin & Bogdan, RAILS, arXiv 2606.08790broader: settlement; related: clearing function, settlement determination
AP2Agent Payments Protocol. A mandate-based protocol defining how, when, and within what bounds value can be authorised for transmission by an AI agent.Google (Agent Payments Protocol)related: x402, mandate protocol; partOf: agent authorization
broker-dealer intermediationReliance on licensed market makers to buy, sell, and redeem the securities backing a digital asset.arXiv 2604.17167related: hierarchy of money, reserve quality
clearing functionA deterministic process that consumes reliability scores and records to dictate which settlement or dispute consequence follows an agent's output.RAILS, arXiv 2606.08790partOf: RAILS; related: per-output reliability score, settlement determination
ERC-8004 (Trustless Agents)Ethereum proposal for on-chain agent identity, reputation, and trustless payment channels.Ethereum EIP-8004related: x402
hierarchy of moneyA classification of instruments by liquidity and counterparty risk, from central-bank reserves down to private credit.Economic literature (Mehrling; cf. arXiv 2604.17167)related: par-value redemption, reserve quality
mandate protocolA framework defining and enforcing pre-approved operational boundaries and financial caps delegated to a software agent.Agent-payments literature (cf. AP2)related: AP2, x402
oracle riskThe risk that an external data feed supplies inaccurate, delayed, or manipulated data to a smart contract.arXiv 2606.08534related: RWA, platform concentration; threatens: settlement
par-value redemptionThe right to exchange a stablecoin for fiat at 1:1 on demand.arXiv 2604.17167related: reserve quality, redemption surge; partOf: stablecoin design
per-output reliability scoreA discrete metric attached to each agent output, quantifying a measured judgement of its quality and adherence to constraints.RAILS, arXiv 2606.08790partOf: RAILS; related: published reliability record, clearing function
platform concentrationSystemic risk from most tokenisation relying on a single underlying protocol or provider.arXiv 2606.08534related: oracle risk, systemic risk
published reliability recordA durable, inspectable log of historical reliability scores and output judgements.RAILS, arXiv 2606.08790partOf: RAILS; related: per-output reliability score, evidence artifacts
RAILSA three-part architectural framework (reliability score, reliability record, clearing function) treating clearing as an independent layer in agentic commerce. Acronym expansion ("Real-Time Agent Integrity & Ledger Settlement") proposed but not yet confirmed against the source paper.arXiv 2606.08790hasPart: per-output reliability score, published reliability record, clearing function; related: x402, AP2
real-world asset tokenizationRepresenting rights to a physical or traditional financial asset as tokens on a blockchain ledger.arXiv 2606.08534hasPart: settlement model, whitelist transferability, rebasing; related: oracle risk, platform concentration
rebasingAn automated mechanism adjusting circulating token supply to reflect yield or price changes.arXiv 2606.08534partOf: RWA token mechanics
redemption surgeA sudden mass demand to redeem stablecoins for fiat, often driven by panic.arXiv 2604.17167related: par-value redemption, stress testing
reserve qualityThe safety, liquidity, and creditworthiness of the assets backing a stablecoin.arXiv 2604.17167related: par-value redemption, broker-dealer intermediation
RWAReal-World Asset. A physical or traditional financial asset that has been tokenised.arXiv 2606.08534broader: real-world asset tokenization; related: oracle risk
settlement determinationThe authoritative decision on how assets should move, distinct from the physical execution of that transfer.RAILS, arXiv 2606.08790related: agentic clearing, settlement model
settlement modelThe technical and legal framework (e.g. atomic, deferred net settlement) finalising ownership transfer.arXiv 2606.08534related: settlement determination, RWA
stablecoinA crypto token designed to hold a stable value (typically 1:1 to fiat); the dominant settlement asset for agent payments.generalrelated: x402, PPSI
trace-economic underwritingPricing insurance or credit by analysing the step-by-step audit trails of an agentic pipeline.Insurance/agent-risk feedsrelated: evidence artifacts, published reliability record
verification-native clearingA design in which clearing checks are embedded within a system's verification phase rather than added post-hoc.RAILS, arXiv 2606.08790related: agentic clearing, RAILS
whitelist transferabilityA feature where tokens transfer only between addresses that have cleared identity checks.arXiv 2606.08534partOf: RWA; related: KYC, sanctions
x402An agent/machine payment scheme built on the HTTP 402 ("Payment Required") status, enabling software agents to settle value (typically in stablecoins) inline with a request.Coinbase (x402 specification)broader: agent payment rail; related: AP2, mandate protocol, RAILS

5. Oversight, safety & interpretability (26 terms)

TermProposed definitionSourceRelations
abliterationModifying model weights to remove safety alignments without a full retrain.Open-weight community (cf. Arditi et al., 2024)threatens: safety guardrails; related: open-weight models
activation patchingCausally modifying internal activations during a forward pass to test how features alter output.Mechanistic-interpretability literaturepartOf: circuit analysis
agent-as-judgeA pattern in which a constrained secondary model audits, evaluates, or filters a primary agent's outputs.Zhuge et al., "Agent-as-a-Judge", 2024; applied in Co-Investigator AI, arXiv 2509.08380related: GAIE, AML, deceptive alignment; mitigates: hallucination
blast-radius oversightA control pattern mapping review intensity to the reach and severity of what an agent's change could disrupt if flawed.GAIE, arXiv 2606.22484partOf: graduated human oversight
circuit analysisReverse-engineering the subgraphs and pathways inside a model that drive behaviours.Elhage et al., "A Mathematical Framework for Transformer Circuits" (2021)related: activation patching, SAE
CSA MAESTROMulti-Agent Environment, Security, Threat, Risk & Outcome; a seven-layer agentic threat-modelling framework.Cloud Security Alliancerelated: OWASP ASI
deceptive alignmentA failure where a model appears compliant in training while concealing misaligned goals.Hubinger et al. (2019)related: ELK, agent-as-judge
development-pipeline governanceThe policies, access controls, and programmatic gates embedded in delivery pipelines to monitor agent activity.GAIE, arXiv 2606.22484partOf: GAIE; hasPart: evidence artifacts; related: generation-to-deployment lifecycle
dictionary learningUnsupervised extraction of an overcomplete set of monosemantic features from activations.Anthropic interpretability workproduces: monosemantic features; related: SAE
ELKEliciting Latent Knowledge. The alignment problem of extracting a model's true internal knowledge.Christiano, Cotra & Xu (ARC, 2021)related: deceptive alignment; goal: truthful reporting
evidence artifactsVerified logs, signatures, and decision records produced by governance workflows to prove compliance.GAIE, arXiv 2606.22484partOf: development-pipeline governance; related: published reliability record
GAIEA framework for enforcing proportionate human oversight on agentic actions in regulated domains. Acronym expansion unconfirmed; the source paper centres on "graduated" oversight, so the earlier "Governed AI-Assisted Engineering" reading should be verified before use.arXiv 2606.22484hasPart: graduated human oversight, regulatory risk profile, blast-radius oversight, evidence artifacts; related: DORA
generation-to-deployment lifecycleThe end-to-end pipeline governing code from initial AI generation to production.GAIE, arXiv 2606.22484related: development-pipeline governance, GAIE
graduated human oversightA governance pattern that adjusts the necessity, depth, and timing of human review according to the risk profile and "blast radius" of an agentic action.GAIE, arXiv 2606.22484definedBy: GAIE; hasPart: blast-radius oversight; related: regulatory risk profile; regulatedBy: EU AI Act Art. 14, prEN 18283
indirect prompt injectionAn exploit where an agent reads untrusted content carrying hidden instructions that hijack its execution.Greshake et al. (2023); OWASP GenAIrelated: memory poisoning, origin-bound authority; regulatedBy: prEN 18229 / 18282
information-flow controlAn architecture tracking and restricting data movement across security domains in a pipeline.Security literature (IFC); agent-security feedsmitigates: indirect prompt injection, memory poisoning
manifold steeringNudging internal activations along valid concept dimensions to alter output in real time.Mechanistic-interpretability literaturerelated: SAE, activation patching
memorization controlParameters and tests that prevent a model from regurgitating exact training sequences, used here to control benchmark contamination.arXiv 2606.00061related: reflexivity, model drift
memory poisoningInjecting malicious data into an agent's memory layer to corrupt future decisions.OWASP GenAI / agent-security literaturerelated: indirect prompt injection, information-flow control; threatens: agent memory
monosemanticityThe state where a neuron or feature maps to exactly one clear concept.Bricken et al. (Anthropic, 2023)related: superposition, SAE; opposite: polysemanticity
OWASP Agentic Security Initiative (ASI)OWASP GenAI Security Project suite covering agentic threat taxonomy, controls, and a ranked Top 10.OWASPrelated: MAESTRO
OWASP Top 10 for Agentic ApplicationsPeer-reviewed ranked list of the most critical security risks for autonomous/agentic AI systems.OWASPpartOf: ASI
polysemanticityThe state where one neuron responds to multiple unrelated concepts by context.Anthropic interpretability workopposite: monosemanticity; related: superposition
regulatory risk profileA classification of an agentic task by its compliance implications, domain sensitivity, and potential for systemic or financial harm.GAIE, arXiv 2606.22484partOf: GAIE; related: graduated human oversight
SAESparse Autoencoder. A network that decomposes dense model activations into interpretable features.Bricken et al., "Towards Monosemanticity" (Anthropic, 2023)related: superposition, dictionary learning, monosemanticity; usedFor: interpretability
superpositionThe phenomenon of a network packing more features than it has neurons via combination.Elhage et al., "Toy Models of Superposition" (Anthropic, 2022)related: polysemanticity, SAE

6. Regulation & governance (30 terms)

TermProposed definitionSourceRelations
10^25 FLOP thresholdA training-compute trigger used to flag potential systemic capability.Regulation (EU) 2024/1689related: Article 51
AI OfficeThe European Commission body coordinating AI policy and supervising GPAI compliance.Regulation (EU) 2024/1689related: GPAI enforcement, EU AI Act
AI-privacy guard layerA software layer that strips or shields personally identifiable information before data enters AI pipelines.Co-Investigator AI, arXiv 2509.08380partOf: agentic AML; related: information-flow control
alternative adequate meansA provision allowing providers to demonstrate compliance via internal frameworks achieving equivalent outcomes to the Code.EC GPAI guidelinesrelated: GPAI Code of Practice
AMLAnti-Money Laundering. The framework of laws and procedures preventing disguise of illicit funds as legitimate income.FATF standards; US BSAbroader: FCC; hasPart: KYC, sanctions; related: BSA, FATF
Article 51The AI Act provision defining when a GPAI model is classified as presenting systemic risk.Regulation (EU) 2024/1689partOf: EU AI Act; related: 10^25 FLOP threshold
Article 53The AI Act provision detailing documentation, transparency, and copyright obligations for GPAI providers.Regulation (EU) 2024/1689partOf: EU AI Act; related: GPAI
Basel IIIThe international framework for bank capital adequacy, stress testing, and liquidity risk.Basel Committee on Banking Supervisionrelated: stress testing, model risk
BSABank Secrecy Act. US legislation requiring institutions to help detect and prevent money laundering.US Bank Secrecy Act (31 U.S.C. 5311 et seq.)related: AML, FinCEN, PPSI
compliance-as-codeWriting regulatory rules and policies directly as executable software.Compliance-to-Code, arXiv 2505.19804broader: RegTech; related: rule-as-code, regulatory rule translation
crime-typology alignmentMatching suspicious transaction sequences against known structured patterns of criminal evasion.Co-Investigator AI, arXiv 2509.08380related: SAR, AML
DORADigital Operational Resilience Act. EU framework on IT security, incident reporting, and resilience for financial entities.Regulation (EU) 2022/2554regulates: financial IT resilience; related: GAIE, RegTech
EU AI ActRegulation (EU) 2024/1689; the EU horizontal AI law, including GPAI obligations.Reg (EU) 2024/1689broader: regulation
FATFFinancial Action Task Force. The intergovernmental body setting AML/CFT standards.FATFrelated: AML, Travel Rule
FCCFinancial Crime Compliance. The bank function and workflows enforcing AML and sanctions policy.Industry usage (e.g. Everest Group)hasPart: AML; related: sanctions false-positive classifier
FinCENFinancial Crimes Enforcement Network. The US Treasury bureau collecting and analysing financial-transaction data.US Department of the Treasury (FinCEN)related: SAR, BSA, PPSI
Five Eyes joint guidanceJoint cyber-security guidance (May 2026) from US/UK/Canada/Australia/NZ agencies on deploying agentic AI; de-facto baseline.CISA et al. (1 May 2026)related: governance
GENIUS ActGuiding and Establishing National Innovation for U.S. Stablecoins Act. US federal legislation enacted 2025 establishing a framework for dollar payment stablecoins.Public Law 119-27 (2025)regulates: PPSI; related: MiCA, BSA
GPAIGeneral-Purpose AI. Models with significant generality, able to perform a broad range of distinct tasks across domains.Regulation (EU) 2024/1689 (AI Act); EC GPAI guidelinesnarrower: systemic-risk GPAI (Art. 51); hasPart: GPAI enforcement, GPAI Code of Practice; related: 10^25 FLOP threshold
GPAI Code of PracticeA regulator-backed operational manual of technical steps, risk assessments, and mitigations for GPAI providers.EC GPAI guidelinesrelated: alternative adequate means, GPAI; partOf: EU AI Act compliance
GPAI enforcementThe auditing protocols and penalty frameworks by which authorities compel GPAI compliance (enforcement from 2 August 2026).EC GPAI guidelines; Regulation (EU) 2024/1689partOf: EU AI Act; related: AI Office, GPAI Code of Practice
IMDA Model AI Governance Framework for Agentic AISingapore framework for governing agentic AI (updated v1.5, 20 May 2026).Singapore IMDArelated: governance
MiCAMarkets in Crypto-Assets Regulation. The EU framework governing crypto-asset and stablecoin issuance and trading.Regulation (EU) 2023/1114related: GENIUS Act, RWA; regulates: crypto-assets
NIST CAISI AI Agent Standards InitiativeUS program (Center for AI Standards and Innovation) for interoperable, secure AI agents; launched 17 Feb 2026.NIST CAISIrelated: governance
PPSIPermitted Payment Stablecoin Issuer. Under the GENIUS Act, an entity authorised to issue payment stablecoins; under the FinCEN/OFAC proposed rule, subject to BSA/AML obligations and a sanctions-compliance programme.FinCEN/OFAC NPRM, Docket FINCEN-2026-0100regulatedBy: GENIUS Act, FinCEN; related: BSA, sanctions
RegTechRegulatory Technology. Software automating compliance, reporting, and risk management.Industry usage (FCA RegTech)hasPart: compliance-as-code; related: DORA
regulatory rule translationConverting natural-language legal text into deterministic code and constraints.Compliance-to-Code, arXiv 2505.19804related: compliance-as-code, rule-as-code
rule-as-codeDrafting statutes or regulations in a structured, machine-readable form from inception.Compliance-to-Code, arXiv 2505.19804; OECD rules-as-code workrelated: compliance-as-code
sanctions false-positive classifierA model that filters out incorrect name-matches from sanctions screening.Industry usage (e.g. Everest Group)partOf: FCC; related: AML
SARSuspicious Activity Report. A confidential compliance filing reporting known or suspected illicit activity to regulators.US Treasury / FinCEN; agentic context: arXiv 2509.08380related: AML, FinCEN, crime-typology alignment; producedBy: agentic AML

7. Agentic standards & standards bodies (13 terms)

TermProposed definitionSourceRelations
AGNTCY collectiveCisco-led "Internet of Agents" collective (LangChain, LlamaIndex, Galileo, others); home of ACP and OASF.AGNTCYrelated: ACP, OASF
C2PACoalition for Content Provenance and Authenticity; content-provenance standard applied to agent-generated content.C2PArelated: CreativeAgent
CEN-CENELEC JTC 21European technical committee drafting harmonised AI standards (the prEN 182xx series).CEN-CENELECmaintains: prEN 182xx
Linux Foundation Agentic AI Foundation (AAIF)Neutral steward now governing both MCP and A2A.Linux Foundation (aaif.io)stewards: MCP, A2A
prEN 18228Draft harmonised standard on robustness, accuracy, and predictability of AI systems.CEN-CENELEC JTC 21related: model risk, EU AI Act
prEN 18229-1 / 18229-2Draft harmonised standards on lifecycle engineering and cybersecurity for high-risk AI.CEN-CENELEC JTC 21related: indirect prompt injection, information-flow control
prEN 18282Draft harmonised standard on recording, logging, and tracing AI operations.CEN-CENELEC JTC 21related: evidence artifacts, published reliability record
prEN 18283Draft harmonised standard on human oversight (human-in/over-the-loop).CEN-CENELEC JTC 21related: graduated human oversight, EU AI Act Art. 14
prEN 18284Draft harmonised standard on information provision, transparency, and disclosure.CEN-CENELEC JTC 21related: transparency disclosure
prEN 18286Draft European harmonised standard on quality management for AI systems.CEN-CENELEC JTC 21partOf: CEN-CENELEC suite; related: EU AI Act
W3C AI Agent Memory Interoperability CGW3C community group on portable agent memory; proposed 18 May 2026.W3Crelated: Memory
W3C AI Agent Protocol CGW3C community group scoping inter-agent communication, identity, and capability metadata.W3Crelated: ANP
W3C AI-KR CGW3C AI Knowledge Representation Community Group (this group).W3Cmaintains: this vocabulary

8. Applied domains (26 terms)

TermProposed definitionSourceRelations
agentic drug discoveryApplying autonomous agents to the multi-step pipeline of target identification, molecule generation, and experiment design.Beyond SMILES, arXiv 2602.10163hasPart: target identification; related: SMILES, ChatInvent, human-in-the-loop discovery; addresses: Eroom's law
ChatInventAn agentic invention/molecular-design system reported as deployed at AstraZeneca.He et al., Drug Discovery Today (2026)related: agentic drug discovery, SMILES
cognitive functionThe vector in a reflexive loop where prices and trends are observed to form imperfect expectations.Soros (1987); arXiv 2606.00061partOf: reflexivity
CoWoSChip-on-Wafer-on-Substrate. An advanced 2.5D semiconductor packaging technology.TSMCrelated: HBM4, neocloud
directional forecasting accuracyThe share of times a system correctly predicts the sign (up/down) of a move, independent of magnitude.arXiv 2606.00061related: Sharpe ratio, reflexivity
Eroom's lawThe observation that drug-discovery productivity per R&D dollar falls over time (Moore's law reversed).Scannell et al. (2012)related: agentic drug discovery
HBM4High Bandwidth Memory 4. The next generation of stacked DRAM for AI accelerators.JEDECrelated: CoWoS
hierarchical mixed-frequency attentionAn attention architecture weighting data sequences across different cadences.Macro-aware, arXiv 2606.00624implements: mixed-frequency forecasting
human-in-the-loop discoveryA workflow where humans guide, review, and approve agent recommendations.Beyond SMILES, arXiv 2602.10163; Robin (Nature, 2026)related: agentic drug discovery, graduated human oversight
MIDASMixed Data Sampling. A regression technique for variables sampled at different frequencies.Ghysels, Santa-Clara & Valkanov (2004)related: mixed-frequency forecasting (precursor)
mixed-frequency forecastingModelling that combines data captured at different intervals (e.g. daily prices with quarterly GDP).Macro-aware, arXiv 2606.00624hasPart: hierarchical mixed-frequency attention; related: MIDAS, model drift
model driftGradual decay of predictive accuracy as real-world conditions diverge from training assumptions.Macro-aware, arXiv 2606.00624related: regime shift, model risk, reflexivity
model riskThe risk of adverse consequences from flawed or misapplied quantitative models.FinStressTS, arXiv 2606.03184; cf. Fed SR 11-7related: model drift, stress testing; regulatedBy: prEN 18228
neocloudA category of providers offering high-performance cloud infrastructure optimised for GPU rental and AI.Industry usagerelated: CoWoS, HBM4
parametric synthetic benchmarkA generated testing environment producing simulated financial time-series with controllable parameters.FinStressTS, arXiv 2606.03184related: stress testing, tail risk; usedFor: model validation
participative functionThe vector in a reflexive loop where expectations are translated into capital allocations, altering prices.Soros (1987); arXiv 2606.00061partOf: reflexivity
process-level vs molecule-level benchmarkAn evaluation paradigm comparing an agent's ability to orchestrate multi-step research against single-structure design.Beyond SMILES, arXiv 2602.10163related: agentic drug discovery
reflexivityThe theory that participants' biased expectations actively alter market reality, creating a self-reinforcing feedback loop.Soros, The Alchemy of Finance (1987); operationalised in arXiv 2606.00061hasPart: participative function, cognitive function; related: directional forecasting accuracy, model drift
regime shiftA significant, persistent change in a time series' underlying distribution.Macro-aware, arXiv 2606.00624closeMatch: regime switch; related: model drift
regime switchAn abrupt structural transition in market behaviour, volatility, or correlations.FinStressTS, arXiv 2606.03184; Macro-aware, arXiv 2606.00624related: model drift, mixed-frequency forecasting; closeMatch: regime shift
reverse translationTaking verified clinical outcomes back into lab models to uncover disease insight.Beyond SMILES (refs), arXiv 2602.10163related: target identification
Sharpe ratioRisk-adjusted performance: excess return divided by volatility.Sharpe (1966)related: tail risk, directional forecasting accuracy
SMILESSimplified Molecular-Input Line-Entry System. A string notation describing chemical structures as text.Weininger (1988)related: agentic drug discovery, ChatInvent
stress testingSimulating extreme market shocks to evaluate how a model, portfolio, or institution survives them.FinStressTS, arXiv 2606.03184; BCBS (Basel III)related: Basel III, parametric synthetic benchmark, redemption surge
tail riskThe probability of extreme losses from rare events at the far end of a distribution.FinStressTS, arXiv 2606.03184related: stress testing, Sharpe ratio
target identificationThe phase pinpointing a biological molecule or pathway central to a disease.Beyond SMILES, arXiv 2602.10163partOf: agentic drug discovery

Acronyms — quick reference (39)

Auto-derived from the terms above. RAILS and GAIE expansions remain unconfirmed.

AcronymExpansion / meaning
A2AAgent
ACPREST
AMLAnti
AnalyticalAgentAgent that performs reasoning and data analysis
ANPOpen, DID
AP2Agent Payments Protocol
AssistantAgentAgent that supports humans or other agents
BSABank Secrecy Act
C2PACoalition for Content Provenance and Authenticity; content
ChatInventAn agentic invention/molecular
CoWoSChip
CreativeAgentAgent that generates content or solutions
CSA MAESTROMulti
DIDW3C standard for verifiable, self
DORADigital Operational Resilience Act
ELKEliciting Latent Knowledge
ERC-8004Ethereum proposal for on
FATFFinancial Action Task Force
FCCFinancial Crime Compliance
FinCENFinancial Crimes Enforcement Network
GENIUS ActGuiding and Establishing National Innovation for U
GPAIGeneral
HBM4High Bandwidth Memory 4
KYAKnow Your Agent
LearningAgentAgent that adapts its behaviour over time from experience
MCPModel Context Protocol
MiCAMarkets in Crypto
MIDASMixed Data Sampling
NANDANetworked AI Agents in a Decentralized Architecture; DID
PPSIPermitted Payment Stablecoin Issuer
RAILSA three
ReactiveAgentAgent with simple stimulus
RegTechRegulatory Technology
RWAReal
SAESparse Autoencoder
SARSuspicious Activity Report
SMILESSimplified Molecular
TaskAgentAgent that executes concrete tasks autonomously
W3C AI-KR CGW3C AI Knowledge Representation Community Group (this group)

Companion documents: the taxonomy (category spine) and the Standards inventory. Narrative background: the vocabulary walkthrough. Status: DRAFT — provisional, for community evaluation and maintenance. License: CC BY 4.0.