Skip to content

GRA · AI and Financial Model Risk

Spatial financeWhere capital meets hazard, exposure and transition

Resilience and finance readiness

Model risk readiness asks whether a model used in a financial decision, from credit scoring to catastrophe pricing, is inventoried, documented, validated, monitored and overseen by people who can challenge it. AI and Financial Model Risk, a the Global Risks Alliance (GRA) platform, prepares the questions and evidence structures. Validation and approval belong to the institution and its supervisor.

02تیاری

Readiness is not approval.

A readiness file describes a programme’s evidence, risks and governance so that an institution can carry out its own review. GRA does not approve, rate, underwrite, lend or invest, and it recommends no counterparty.

03Regimes

Model and AI rules side by side.

Regimeدائرہ کارKey expectation
SR 26-2 (US, April 2026)Models at banking organisations; generative and agentic AI stated to be outside scopeRisk-based, tailored model risk management commensurate with model use
PRA SS1/23 (UK)Banks with internal model approvalFive principles: identification, governance, development, validation, mitigants
EU AI Act with AI OmnibusHigh-risk AI including creditworthiness and life and health insurance pricingObligations apply from 2 December 2027
NIST AI RMFVoluntary, all sectorsGovern, map, measure, manage
04عمل

عمل کیسے چلتا ہے۔

Process explorer · Process

Resilience and finance readiness: Model risk lifecycle

10 steps3 stages

عمل

Resilience and finance readiness: Model risk lifecycle

امریکی بینک نگرانوں نے 17 اپریل 2026 کو ماڈل کے خطرے کی رہنمائی پر نظرثانی کی، Fed SR 26-2 نے SR 11-7 کی جگہ لی؛ یہ کوئی قابل نفاذ معیار طے نہیں کرتی۔

  1. 01 فہرست سازی اور تیاری

    1. 02Model inventory and materialityDataset · Internal auditAn effective inventory shows model risk for single models and in aggregate. Model purpose with model exposure sets materiality.
    2. 04Development and testingStep · Internal auditDevelopment starts from a clear purpose aligned with business use and policy. Testing checks whether the model performs as intended.
    3. 05DocumentationRecord · Internal auditAdequate documentation supports continuity of operations, the tracking of recommendations, responses and exceptions, and remediation.
  2. 02 توثیق اور چیلنج

    1. 03Effective challengeComponent · Internal auditObjective experts critically evaluate model risk and drive change across the lifecycle. They need expertise, independence and standing.
    2. 06Conceptual soundnessStep · Internal auditValidation of conceptual soundness assesses and documents model design, construction and developmental testing.
    3. 07Outcomes analysisStep · Internal auditOutcomes analysis compares model outputs with real-world outcomes. It can include back-testing or outlier analysis.
  3. 03 نظم و نسق، استعمال اور نگرانی

    1. 01Policies, roles and responsibilitiesComponent · Internal auditPolicies set expectations and a framework for assessing model risk. Roles are clear, with attention to conflicts of interest.
    2. 08Decision to use, with limits where neededDecision · Internal auditValidation generally comes before first use. Urgent earlier use calls for attention to limits, informed stakeholders and controls.
    3. 09Ongoing monitoringModel · Internal auditMonitoring checks whether the model performs as expected as products, exposures, clients, data or markets change.
    4. 10Adjust, recalibrate or redevelopStep · Internal auditWhen performance deviates meaningfully from expectations, the organisation considers adjustment, recalibration or redevelopment.

کسی بیرونی معیار یا طریقے کا خاکہ، اس کے عوامی ماخذات سے۔ یہ بیان کرتا ہے کہ فریم ورک کیا کہتا ہے، کسی عمل کا ریکارڈ نہیں۔ یہ نہیں دکھاتا کہ Nexus یا کوئی شراکت دار کوئی مرحلہ انجام دیتا، اس کی تصدیق کرتا یا اسے منظور کرتا ہے۔ دروازہ طریقے کا ایک مرحلہ ہے، کبھی Nexus کی منظوری نہیں۔ توقع ہے کہ یہ رہنمائی $30 billion سے زیادہ کے کل اثاثوں والی بینکاری تنظیموں کے لیے سب سے زیادہ متعلقہ ہوگی۔

05خلا

Generative AI sits between regimes.

With SR 26-2 placing generative and agentic AI outside its scope, US banks govern those systems through broader risk management and third-party expectations while supervisors develop specific guidance. GRA reports this as a gap in the record, not as a rule. Topic depth at generative AI in finance; oversight at the Global Risks Forum (GRF) AI oversight and accountability.

06Release

Who sees a readiness file.

A file is released only at the programme owner’s written direction, to a party the owner chooses, with no fee tied to any transaction.

07ماخذات

Where this page’s references lead.

وہ ادارے اور دستاویزات جن کے تناظر میں یہ صفحہ پڑھا جاتا ہے۔ حیثیت اور ایڈیشن بدلتے رہتے ہیں؛ ہر لنک جاری کرنے والے ادارے تک جاتا ہے۔

08صفحات

Pages in Resilience and finance readiness.

09سوالات

سوالات و جوابات۔

Does a readiness file mean finance is available?
No. It supports an independent review by the competent institution, which decides under its own rules.
AI اور مالیاتی ماڈل کا خطرہ

Bring an exposure or capital question to the Alliance.

معلومات اور استفسارات آج کھلے ہیں۔ استفسار کو ایک حوالہ نمبر ملتا ہے اور اگلے مرحلے کی نشاندہی ہوتی ہے؛ درخواستیں، ادائیگیاں، کمپیوٹنگ کی تخصیص اور رجسٹریشن صرف اس وقت کھلتی ہیں جب ان کی ٹیم، شرائط اور جائزے کی استعداد موجود ہو۔