Counterparty Profile

Counterparty risk knowledge lives entirely in the heads of individual relationship managers and credit analysts. There is no written counterparty profile for any entity the firm does business with. When a trader asks 'what is our credit exposure to Goldman Sachs?' nobody can answer because there is no consolidated view of counterparty relationships, exposure, or risk characteristics. When a new trade is proposed with a counterparty, the credit analyst relies on personal memory and informal conversations with colleagues to assess the counterparty's creditworthiness. Different parts of the firm may be transacting with the same legal entity under different names or through different subsidiaries without realizing they are accumulating exposure to a single economic group. When a counterparty defaults or is downgraded, the firm scrambles to determine which desks have exposure, what instruments are involved, and what the total potential loss is — a process that takes days of frantic phone calls because no counterparty profile exists to reference. The complete absence of formal counterparty documentation means the firm cannot manage concentration risk, cannot perform credit due diligence systematically, and cannot demonstrate to regulators that it understands who it is doing business with.

None — AI cannot perform any counterparty risk analysis because no counterparty records exist in any system. Every credit decision depends entirely on individual human memory and judgment.

Counterparty profiles exist as basic entries in a shared registry — a spreadsheet or simple database — with each counterparty having a name, jurisdiction, business type classification, and internal credit rating. This provides a minimal foundation: the firm can at least list its counterparties and look up basic information about each one. However, the profiles are shallow and inconsistent. Some counterparty entries have detailed notes about the relationship history and risk characteristics, while others have only a name and a rating with no supporting context. There is no standard set of required fields — the depth of each profile depends entirely on the credit analyst who created it. Critical information is missing from most profiles: beneficial ownership structure, legal entity hierarchy (which subsidiaries belong to which parent), netting agreement status, collateral arrangements, and settlement risk characteristics. The counterparty name field contains inconsistencies — the same entity might appear as 'Goldman Sachs,' 'Goldman Sachs & Co.,' 'GS,' and 'Goldman Sachs International' with no indication that these are related entities within the same corporate group. Without legal entity hierarchy mapping, the firm cannot aggregate exposure across subsidiaries to assess group-level concentration risk. The registry proves that counterparties exist but does not provide the depth needed for meaningful credit risk management.

AI can look up basic counterparty information and generate simple lists, but cannot assess counterparty risk, map legal entity hierarchies, or identify related-entity exposure because profiles lack the depth, consistency, and relational structure needed for meaningful analysis.

Counterparty profiles follow a standardized template with comprehensive required fields: legal name, Legal Entity Identifier (LEI), jurisdiction of incorporation, legal entity hierarchy showing the relationship between the transacting entity and its parent group, internal credit rating with the methodology and date of the last review, external ratings from major agencies, netting agreement status and ISDA master agreement details, collateral arrangements including CSA thresholds and eligible collateral types, settlement risk characteristics, and regulatory classification (central counterparty, bank, insurance company, corporate, sovereign). Every active counterparty has a complete profile that meets the minimum documentation standard. The firm can now answer fundamental questions: 'What is our total exposure to the Citigroup corporate family across all subsidiaries and all products?' because legal entity hierarchies are mapped. Credit analysts have a consistent information base for every credit review. However, the counterparty profile exists as a standalone document — it is not linked to the firm's actual trading positions, exposure calculations, or collateral management records. To assess current exposure to a specific counterparty, an analyst must pull the profile for credit characteristics and then separately query the trading system for current positions and the collateral system for posted margin. The profile tells you who the counterparty is but not what your current relationship with them looks like in terms of live exposure and risk.

AI can reference standardized counterparty profiles for credit analysis, identify related entities through legal entity hierarchies, and generate counterparty risk reports. Cannot assess current exposure or collateral adequacy because the profile is disconnected from position and collateral systems.

Counterparty profiles are comprehensive, living documents that integrate static credit characteristics with dynamic exposure and relationship data. Each profile links directly to current trading positions across all products and desks, real-time credit exposure calculations including current exposure, potential future exposure, and exposure at default, collateral management records showing posted and received margin with mark-to-market values, netting set definitions with their legal enforceability assessments, and the complete history of credit events, rating changes, and limit modifications for the entity and its corporate group. A credit analyst opening a counterparty profile sees the complete picture in one place: who the counterparty is, what the firm's current exposure is across all products, what collateral is in place, what the credit assessment is, and what the historical relationship trajectory looks like. The profile supports group-level aggregation — viewing the entire corporate family's exposure, collateral, and credit status as a consolidated view. Credit committee reviews are data-driven because every relevant metric is assembled in the profile. Regulatory examinations of counterparty risk management can be conducted by walking through profiles rather than assembling information from multiple disconnected sources. The firm can now perform meaningful credit portfolio analysis, concentration risk assessment, and what-if scenario analysis on counterparty exposures.

AI can perform comprehensive counterparty risk assessment combining credit characteristics, exposure data, collateral adequacy, and relationship history. Can draft credit review memos, flag deteriorating counterparties, and model exposure scenarios. Cannot yet perform autonomous credit decisions because judgment on complex credits requires human expertise.

Counterparty profiles are formal entities in a risk knowledge graph with machine-readable semantic relationships to every relevant dimension of counterparty risk. Each counterparty entity connects to its legal entity hierarchy, beneficial ownership network, industry sector classification, geographic presence, regulatory status across jurisdictions, instrument-level exposures, netting sets, collateral agreements, credit events, rating histories, and market-implied credit metrics. The knowledge graph captures complex relationships that flat profiles cannot: 'Counterparty A shares beneficial ownership with Entity B through a holding company in Luxembourg, Entity B is a guarantor for Counterparty C, and all three entities have exposure to the same commodity sector' — enabling the firm to identify hidden concentration risks through chains of corporate relationships. AI agents can traverse the graph to answer sophisticated questions: 'If oil prices drop 30%, which counterparties in our portfolio are most vulnerable considering their sector exposure, corporate affiliations, and the firm's current hedging through CDS protection on related entities?' The ontology maps each counterparty to the relevant regulatory capital framework — standardized approach risk weights, internal ratings-based parameters, CVA capital charges — creating a direct link between the counterparty profile and its regulatory capital implications. The firm can model the impact of counterparty credit events on both economic risk and regulatory capital simultaneously.

AI can perform autonomous graph-based counterparty risk analysis, identify hidden concentration risks through corporate relationship networks, model multi-counterparty credit scenarios, and optimize the credit portfolio considering both economic risk and regulatory capital. Human credit officers review AI recommendations on complex credits.

Counterparty profiles are living intelligence entities that continuously assemble and update themselves from all available information sources in real-time. When a credit rating agency places a counterparty on negative watch, the profile updates instantly with the rating action, the agency's rationale, the market response in CDS spreads and bond prices, the impact on the firm's exposure-at-default calculations, the collateral adequacy under the new credit assessment, and AI-generated recommendations for portfolio adjustments. When news breaks about a counterparty's industry sector — a regulatory action, a competitor default, or an economic policy change — the profile automatically ingests and contextualizes the information relative to the specific counterparty's risk characteristics. The counterparty entity in the knowledge graph evolves continuously: corporate restructurings, M&A activity, new subsidiary formations, and beneficial ownership changes are detected from public filings and corporate registries and reflected in the profile's legal entity hierarchy within hours of disclosure. The profile captures not just current state but forward-looking risk assessments — AI-generated credit migration probabilities, sector stress vulnerability scores, and scenario-conditional exposure projections that update as market conditions and counterparty-specific information evolve. Credit officers work with counterparty profiles that are always current, comprehensive, and forward-looking rather than static snapshots that decay between periodic reviews.

Fully autonomous counterparty intelligence. AI continuously assembles, updates, and analyzes counterparty profiles from all available sources. Generates real-time credit assessments, exposure forecasts, and portfolio optimization recommendations. Human credit judgment is reserved for strategic relationship decisions and complex restructuring situations.

Ceiling of the CMC framework for this dimension.