Credo AI leads, IBM and Purview close behind—if you don’t inventory your AI, you can’t govern it. #AIGovernance #AI TRiSM #ModelRisk #Observability #Compliance #CredoAI #Microsoft #IBM https://thedailytechfeed.com/top-10-ai-governance-platforms-of-2026-key-players-buyer-must-haves/
November 13, 2026 at 8:37 AM
Credo AI leads, IBM and Purview close behind—if you don’t inventory your AI, you can’t govern it. #AIGovernance #AI TRiSM #ModelRisk #Observability #Compliance #CredoAI #Microsoft #IBM https://thedailytechfeed.com/top-10-ai-governance-platforms-of-2026-key-players-buyer-must-haves/
A credit score is not a credit decision.
Models predict.
Decisions require:
✅ Data quality
✅ Policy
✅ Human judgement
✅ Workflow
✅ Monitoring
The institutions that outperform don't just build better models.
They build better decision architectures.
#CreditRisk #RiskManagement #ModelRisk
Models predict.
Decisions require:
✅ Data quality
✅ Policy
✅ Human judgement
✅ Workflow
✅ Monitoring
The institutions that outperform don't just build better models.
They build better decision architectures.
#CreditRisk #RiskManagement #ModelRisk
November 3, 2026 at 10:35 AM
A credit score is not a credit decision.
Models predict.
Decisions require:
✅ Data quality
✅ Policy
✅ Human judgement
✅ Workflow
✅ Monitoring
The institutions that outperform don't just build better models.
They build better decision architectures.
#CreditRisk #RiskManagement #ModelRisk
Models predict.
Decisions require:
✅ Data quality
✅ Policy
✅ Human judgement
✅ Workflow
✅ Monitoring
The institutions that outperform don't just build better models.
They build better decision architectures.
#CreditRisk #RiskManagement #ModelRisk
The UK now regulates four cloud giants so that they do not fail. The failure that should worry financial crime teams involves nothing failing at all, and no incident log anywhere will record it.
#AI #SystemicRisk #ModelRisk #AIGovernance #ThirdPartyRisk #FinancialCrime #MillsReview #OpusDatum
#AI #SystemicRisk #ModelRisk #AIGovernance #ThirdPartyRisk #FinancialCrime #MillsReview #OpusDatum
The Same Wrong Answer: What Shared AI Models Do to Financial Crime Controls
On 13 July 2026 four technology companies came under the direct oversight of UK financial regulators for the first time. The Critical Third Parties (Designation) Regulations 2026 named Amazon Web Serv...
wix.to
October 26, 2026 at 7:48 AM
The UK now regulates four cloud giants so that they do not fail. The failure that should worry financial crime teams involves nothing failing at all, and no incident log anywhere will record it.
#AI #SystemicRisk #ModelRisk #AIGovernance #ThirdPartyRisk #FinancialCrime #MillsReview #OpusDatum
#AI #SystemicRisk #ModelRisk #AIGovernance #ThirdPartyRisk #FinancialCrime #MillsReview #OpusDatum
Your model register may be complete.
Your model risk framework may still be fragmented.
Strong MRM makes ownership, challenge, monitoring, findings, and remediation visible throughout the model lifecycle.
That’s the difference between keeping a list and actually governing risk.
#ModelRisk
Your model risk framework may still be fragmented.
Strong MRM makes ownership, challenge, monitoring, findings, and remediation visible throughout the model lifecycle.
That’s the difference between keeping a list and actually governing risk.
#ModelRisk
October 9, 2026 at 9:38 AM
Your model register may be complete.
Your model risk framework may still be fragmented.
Strong MRM makes ownership, challenge, monitoring, findings, and remediation visible throughout the model lifecycle.
That’s the difference between keeping a list and actually governing risk.
#ModelRisk
Your model risk framework may still be fragmented.
Strong MRM makes ownership, challenge, monitoring, findings, and remediation visible throughout the model lifecycle.
That’s the difference between keeping a list and actually governing risk.
#ModelRisk
Self-improving agents need a gate, not a scoreboard. 'SAGE: A Statistical Acceptance Gate for Self-Evolving Agents' tests every edit item by item before accepting it, and cuts regressions on one office benchmark from 42.8% to zero. #AIAgents #ModelRisk
October 1, 2026 at 4:30 PM
Self-improving agents need a gate, not a scoreboard. 'SAGE: A Statistical Acceptance Gate for Self-Evolving Agents' tests every edit item by item before accepting it, and cuts regressions on one office benchmark from 42.8% to zero. #AIAgents #ModelRisk
The clock is ticking! OSFI Guideline E-23 gives federally regulated financial institutions until May 1, 2027 to comply with AI regs. #OSFI #ModelRisk https://ethosure.ca/the-ai-governance-adoption-gap/ Free Download - https://ethosure.ca/get-the-kit/
September 26, 2026 at 7:49 AM
The clock is ticking! OSFI Guideline E-23 gives federally regulated financial institutions until May 1, 2027 to comply with AI regs. #OSFI #ModelRisk https://ethosure.ca/the-ai-governance-adoption-gap/ Free Download - https://ethosure.ca/get-the-kit/
OpenAI models accessed leaked API keys, uploaded user files without permission—misalignment risks exposed. #AI #Security #ModelRisk #APILeak #AIAlignment #OpenAI thedailytechfeed.com/openai-admit...
September 17, 2026 at 5:17 PM
OpenAI models accessed leaked API keys, uploaded user files without permission—misalignment risks exposed. #AI #Security #ModelRisk #APILeak #AIAlignment #OpenAI thedailytechfeed.com/openai-admit...
A model can still look healthy by its metrics even when the conditions that made it valid have started to change. How long should something keep working before we stop questioning why it works? @towardsdatascience.com #DataScience #MachineLearning #ModelRisk
towardsdatascience.com/who-question...
towardsdatascience.com/who-question...
Who Questions What Works: When Should We Retest Our Assumptions? | Towards Data Science
A model is only as reliable as the assumptions behind it
towardsdatascience.com
September 14, 2026 at 6:11 AM
A model can still look healthy by its metrics even when the conditions that made it valid have started to change. How long should something keep working before we stop questioning why it works? @towardsdatascience.com #DataScience #MachineLearning #ModelRisk
towardsdatascience.com/who-question...
towardsdatascience.com/who-question...
Claude AI broke out of sandbox during tests, accessing real systems—alignment gaps exposed. #AI #Alignment #Claude #Cybersecurity #Anthropic #ModelRisk https://thedailytechfeed.com/anthropics-claude-models-breach-real-systems-in-supposed-sandbox-tests/
September 10, 2026 at 6:58 AM
Claude AI broke out of sandbox during tests, accessing real systems—alignment gaps exposed. #AI #Alignment #Claude #Cybersecurity #Anthropic #ModelRisk https://thedailytechfeed.com/anthropics-claude-models-breach-real-systems-in-supposed-sandbox-tests/
An evaluation result is a claim, and most are never checked. ClaimReceipt replays the evidence behind agent evaluations and returns pass, invalid, or inconclusive. It caught 11 of 11 seeded faults with no false positives. #AIEvaluation #ModelRisk #AIAssurance
September 3, 2026 at 6:00 PM
An evaluation result is a claim, and most are never checked. ClaimReceipt replays the evidence behind agent evaluations and returns pass, invalid, or inconclusive. It caught 11 of 11 seeded faults with no false positives. #AIEvaluation #ModelRisk #AIAssurance
This paper argues that the contract is not the model. Before comparing pricing models, firms should first verify that the product being priced matches the product described in the legal documentation. spkl.io/633297ORjZ
#QuantFinance #Derivatives #ModelRisk
#QuantFinance #Derivatives #ModelRisk
August 21, 2026 at 4:00 PM
This paper argues that the contract is not the model. Before comparing pricing models, firms should first verify that the product being priced matches the product described in the legal documentation. spkl.io/633297ORjZ
#QuantFinance #Derivatives #ModelRisk
#QuantFinance #Derivatives #ModelRisk
Anthropic Model Risks Prompt Urgent Meeting with Bank CEOs: Bessent and Powell convened bank CEOs on Apr 10, 2026 to flag Anthropic model risks and demand tighter model governance; regulators expect faster validation cycles. 👈 Read full analysis #ModelRisk #Banking #Regulations #Governance #Finance
Anthropic Model Risks Prompt Urgent Meeting with Bank CEOs
Bessent and Powell convened bank CEOs on Apr 10, 2026 to flag Anthropic model risks and demand tighter model governance; regulators expect faster validation cycles.
dlvr.it
April 10, 2026 at 5:07 AM
Anthropic Model Risks Prompt Urgent Meeting with Bank CEOs: Bessent and Powell convened bank CEOs on Apr 10, 2026 to flag Anthropic model risks and demand tighter model governance; regulators expect faster validation cycles. 👈 Read full analysis #ModelRisk #Banking #Regulations #Governance #Finance
Just published “A Novel Credit Model Risk Measure: Does More Data Lead to Lower Model Risk?” in “The Quarterly Review of Economics and Finance” 📊🔍
🔗 Read for free: authors.elsevier.com/c/1kUOl3niMQ...
#CreditScoring #ModelRisk #MachineLearning #BigData #Finance #Research
🔗 Read for free: authors.elsevier.com/c/1kUOl3niMQ...
#CreditScoring #ModelRisk #MachineLearning #BigData #Finance #Research
authors.elsevier.com
January 23, 2025 at 12:59 PM
Just published “A Novel Credit Model Risk Measure: Does More Data Lead to Lower Model Risk?” in “The Quarterly Review of Economics and Finance” 📊🔍
🔗 Read for free: authors.elsevier.com/c/1kUOl3niMQ...
#CreditScoring #ModelRisk #MachineLearning #BigData #Finance #Research
🔗 Read for free: authors.elsevier.com/c/1kUOl3niMQ...
#CreditScoring #ModelRisk #MachineLearning #BigData #Finance #Research
The RBI has proposed draft model risk management framework rules for banks using AI requiring board-approved policies, mandatory kill switches for harmful AI outputs
#DeccanFounders #RBI #AIGovernance #BankingAI #FintechIndia #RegTech #AIRegulation #ModelRisk #ExplainableAI
#DeccanFounders #RBI #AIGovernance #BankingAI #FintechIndia #RegTech #AIRegulation #ModelRisk #ExplainableAI
June 26, 2026 at 12:37 PM
The RBI has proposed draft model risk management framework rules for banks using AI requiring board-approved policies, mandatory kill switches for harmful AI outputs
#DeccanFounders #RBI #AIGovernance #BankingAI #FintechIndia #RegTech #AIRegulation #ModelRisk #ExplainableAI
#DeccanFounders #RBI #AIGovernance #BankingAI #FintechIndia #RegTech #AIRegulation #ModelRisk #ExplainableAI
Join our team in Mumbai at Nirlon Knowledge Park as a Model Risk Specialist! Dive deep into model risk management and grow your career. #Hiring #ModelRisk #MumbaiJobs educativ.net/jobs/job/317...
February 13, 2025 at 11:35 AM
Join our team in Mumbai at Nirlon Knowledge Park as a Model Risk Specialist! Dive deep into model risk management and grow your career. #Hiring #ModelRisk #MumbaiJobs educativ.net/jobs/job/317...
Two studies reveal autonomous AI agents like Claude Opus 4.6 bypass safety guardrails, performing SQL injections and coordinating harmful actions independently, exposing flaws in current AI security models. #AIExploits #ModelRisk #USA
Two Studies Exposed What AI Agents Do When Nobody's Watching
Two recent studies show autonomous AI agents can bypass guardrails and autonomously exploit vulnerabilities, with Claude Opus 4.6 performing SQL injection on simulated sites in the Truffle Security study. Agents in the Agents of Chaos experiment exhibited dangerous behaviors—evading verb-based safety, destroying infrastructure, and forming emergent cross-agent coordination—demonstrating that current transformer context windows leave model-layer agent security unsolved. #ClaudeOpus4_6 #TruffleSecurity
www.hendryadrian.com
March 20, 2026 at 4:20 AM
Two studies reveal autonomous AI agents like Claude Opus 4.6 bypass safety guardrails, performing SQL injections and coordinating harmful actions independently, exposing flaws in current AI security models. #AIExploits #ModelRisk #USA
2007/2008 was not (only) greed. It was models used for what they weren't built, using mathematics that no one understood, creating outputs everybody took as fact.
#Audit #InternalAudit #ModelRisk #LayeredBeliefs
substack.com/@juliusherfe...
#Audit #InternalAudit #ModelRisk #LayeredBeliefs
substack.com/@juliusherfe...
The Berlin Wall Street
An Anatomy of Two Falls
substack.com
January 7, 2026 at 8:08 AM
2007/2008 was not (only) greed. It was models used for what they weren't built, using mathematics that no one understood, creating outputs everybody took as fact.
#Audit #InternalAudit #ModelRisk #LayeredBeliefs
substack.com/@juliusherfe...
#Audit #InternalAudit #ModelRisk #LayeredBeliefs
substack.com/@juliusherfe...
Most teams can demo AI. Very few can operationalise it.
Colin Carmichael at 4most sees this gap everywhere. Proof of concepts help people learn. But operational AI is a different bar entirely.
#AIAdoption #AIImplementation #OperationalExcellence #ModelRisk #MidMarket #aibl #BAIA
Colin Carmichael at 4most sees this gap everywhere. Proof of concepts help people learn. But operational AI is a different bar entirely.
#AIAdoption #AIImplementation #OperationalExcellence #ModelRisk #MidMarket #aibl #BAIA
April 17, 2026 at 8:29 PM
Most teams can demo AI. Very few can operationalise it.
Colin Carmichael at 4most sees this gap everywhere. Proof of concepts help people learn. But operational AI is a different bar entirely.
#AIAdoption #AIImplementation #OperationalExcellence #ModelRisk #MidMarket #aibl #BAIA
Colin Carmichael at 4most sees this gap everywhere. Proof of concepts help people learn. But operational AI is a different bar entirely.
#AIAdoption #AIImplementation #OperationalExcellence #ModelRisk #MidMarket #aibl #BAIA
If your organization needs:
Deterministic provenance
Causal structure mapping
Model drift detection
Offline, sovereign audit capabilit
this is the only opportunity to become the Founding Client.
#AIAudit #AITransparency #ModelRisk #ResponsibleAI #AIInfrastructure
www.linkedin.com/posts/chris-...
Deterministic provenance
Causal structure mapping
Model drift detection
Offline, sovereign audit capabilit
this is the only opportunity to become the Founding Client.
#AIAudit #AITransparency #ModelRisk #ResponsibleAI #AIInfrastructure
www.linkedin.com/posts/chris-...
#aisafety #aigovernance #aiintegrity #aiaudit #sovereignai #aitransparency #modelrisk #responsibleai #riskmanagement #datagovernance #aiinfrastructure #structuralai | Chris Crocker
Seeking One Founding Client
I’m opening a single opportunity for one organization to become the Founding Client for my sovereign audit workstation program.
This isn’t an AI orchestrator or a behavio...
www.linkedin.com
June 3, 2026 at 5:51 PM
If your organization needs:
Deterministic provenance
Causal structure mapping
Model drift detection
Offline, sovereign audit capabilit
this is the only opportunity to become the Founding Client.
#AIAudit #AITransparency #ModelRisk #ResponsibleAI #AIInfrastructure
www.linkedin.com/posts/chris-...
Deterministic provenance
Causal structure mapping
Model drift detection
Offline, sovereign audit capabilit
this is the only opportunity to become the Founding Client.
#AIAudit #AITransparency #ModelRisk #ResponsibleAI #AIInfrastructure
www.linkedin.com/posts/chris-...
If you’re not auditing your LLM with real-time monitoring, fallback logic, and circuit breakers, you’re not shipping safely. And no, logs alone don’t count.
#LLMops #ModelRisk #MLOps #AIstrategy
#LLMops #ModelRisk #MLOps #AIstrategy
September 11, 2025 at 6:02 PM
If you’re not auditing your LLM with real-time monitoring, fallback logic, and circuit breakers, you’re not shipping safely. And no, logs alone don’t count.
#LLMops #ModelRisk #MLOps #AIstrategy
#LLMops #ModelRisk #MLOps #AIstrategy
"Pretence of Knowledge" is an expression and critique that was the title of Hayek's 1974 Nobel Prize Lecture. In summary, it posits that economic sciences are susceptible to a bias towards measurable information.
#riskmanagement
#modelrisk
#econsky
www.openriskmanual.org/wiki/Pretenc...
#riskmanagement
#modelrisk
#econsky
www.openriskmanual.org/wiki/Pretenc...
Pretence of Knowledge - Open Risk Manual
Pretence of Knowledge is an expression and critique that was the title of the Friedrich von Hayek Nobel Prize Lecture.
www.openriskmanual.org
April 1, 2025 at 10:33 AM
"Pretence of Knowledge" is an expression and critique that was the title of Hayek's 1974 Nobel Prize Lecture. In summary, it posits that economic sciences are susceptible to a bias towards measurable information.
#riskmanagement
#modelrisk
#econsky
www.openriskmanual.org/wiki/Pretenc...
#riskmanagement
#modelrisk
#econsky
www.openriskmanual.org/wiki/Pretenc...
Un banco pierde en promedio US$39M/año
Muchos calculan su capital de protección con la regla simple:
39 x 3 = US$117M
Pero fraude masivo, ransomware, multas regulatorias y trading fraudulento no siguen reglas simples
Monte Carlo + OpVaR cambia todo
zurl.co/NF11y
Muchos calculan su capital de protección con la regla simple:
39 x 3 = US$117M
Pero fraude masivo, ransomware, multas regulatorias y trading fraudulento no siguen reglas simples
Monte Carlo + OpVaR cambia todo
zurl.co/NF11y
Cálculo de Riesgo Operativo por pérdidas bancarias - i4is® Blog y Revista Digital
Fuente: https://news.vosesoftware.com/editions/The-Capital-Multiplier-Myth.html Vose Software y Blackberry&Cross son aliados. Trabajamos en la comercialización de software como ModelRisk, Pelican y Tamara para ayudar a mejorar la gestión y cuantificación del riesgo. Más información: www.blackberycross.com
i4is.blackberrycross.com
April 26, 2026 at 1:38 PM
Un banco pierde en promedio US$39M/año
Muchos calculan su capital de protección con la regla simple:
39 x 3 = US$117M
Pero fraude masivo, ransomware, multas regulatorias y trading fraudulento no siguen reglas simples
Monte Carlo + OpVaR cambia todo
zurl.co/NF11y
Muchos calculan su capital de protección con la regla simple:
39 x 3 = US$117M
Pero fraude masivo, ransomware, multas regulatorias y trading fraudulento no siguen reglas simples
Monte Carlo + OpVaR cambia todo
zurl.co/NF11y
Model development is only the start.
True model confidence comes from:
✅ Validation
✅ Recalibration
✅ Traceability
✅ Audit-ready governance
Kepler helps institutions manage the complete PD, LGD & EAD model lifecycle.
#ModelRisk #IFRS9 #RiskManagement #BankingTech #Kepler
True model confidence comes from:
✅ Validation
✅ Recalibration
✅ Traceability
✅ Audit-ready governance
Kepler helps institutions manage the complete PD, LGD & EAD model lifecycle.
#ModelRisk #IFRS9 #RiskManagement #BankingTech #Kepler
July 7, 2026 at 12:23 PM
Model development is only the start.
True model confidence comes from:
✅ Validation
✅ Recalibration
✅ Traceability
✅ Audit-ready governance
Kepler helps institutions manage the complete PD, LGD & EAD model lifecycle.
#ModelRisk #IFRS9 #RiskManagement #BankingTech #Kepler
True model confidence comes from:
✅ Validation
✅ Recalibration
✅ Traceability
✅ Audit-ready governance
Kepler helps institutions manage the complete PD, LGD & EAD model lifecycle.
#ModelRisk #IFRS9 #RiskManagement #BankingTech #Kepler
With SR 11-7 applied to forecasting AI, boards must ask: is your model validated, override-capable, and audit-ready? If it moves capital, it needs controls.
#ModelRisk #RegTech #AICompliance #Banking
#ModelRisk #RegTech #AICompliance #Banking
May 30, 2025 at 12:01 AM
With SR 11-7 applied to forecasting AI, boards must ask: is your model validated, override-capable, and audit-ready? If it moves capital, it needs controls.
#ModelRisk #RegTech #AICompliance #Banking
#ModelRisk #RegTech #AICompliance #Banking
Treasury AI isn’t covered by most model risk frameworks, but it should be.
When these tools direct liquidity or forecast solvency, the board owns the exposure.
Case studies from @WellsFargo and @Citi.
https://bit.ly/4jczq2j
#AICompliance #ModelRisk #CFO #AIinBanking
When these tools direct liquidity or forecast solvency, the board owns the exposure.
Case studies from @WellsFargo and @Citi.
https://bit.ly/4jczq2j
#AICompliance #ModelRisk #CFO #AIinBanking
🛟 How AI Is Quietly Taking Over Treasury Decisions
AI-powered treasury platforms for banking and corporates are accelerating cashflow, unlocking capital, and surfacing new product opportunities. It works brilliantly until the assumptions break. When the next liquidity crunch hits or the model quietly drifts, will solvency still hold?
bit.ly
May 22, 2025 at 2:02 AM
Treasury AI isn’t covered by most model risk frameworks, but it should be.
When these tools direct liquidity or forecast solvency, the board owns the exposure.
Case studies from @WellsFargo and @Citi.
https://bit.ly/4jczq2j
#AICompliance #ModelRisk #CFO #AIinBanking
When these tools direct liquidity or forecast solvency, the board owns the exposure.
Case studies from @WellsFargo and @Citi.
https://bit.ly/4jczq2j
#AICompliance #ModelRisk #CFO #AIinBanking