A Delphi study of 272 experts prioritized 24 AI risks: 18 carry a >10% chance of catastrophe by 2030 under business-as-usual. What it means for financial-institution risk governance.
Experts rated finance among the most AI-vulnerable sectors: fraud, market manipulation, and AI security. The three financial attack surfaces, and the engines that monitor each.
Those exposed to AI harm can't prevent it; those who can face competitive pressure against caution. Why voluntary self-governance fails, the hard instruments that fix incentives, and how to be ready.
Power centralization, competitive dynamics, inequality and environmental harm are AI risks a safety patch cannot reach. Why they persist, and how to measure exposure to them.
Dangerous capabilities and AI-enabled weapons/cyberattacks top the expert risk table. What they mean, why model-level fixes don't touch them, and how a continuous agent-monitoring layer surfaces them.
A practitioner's guide to portfolio optimisation constraints: turnover, cardinality, transaction costs, estimation error, and three diagnostics to run now.
The EU AI Act Digital Omnibus delays high-risk obligations to December 2027 — what stays in force in 2026 and what it means for financial firms.
Aspects Q2 2026 — four risk signatures in Tech & AI infrastructure: capex-financing migration, market concentration, the power constraint, and AI Act governance.
Risk Heartbeat June 2026 — non-bank leverage and private-credit concentration graduates to a flagged signal, the energy-led reflationary regime stays modal through the ECB's 11 June decision, and dollar-funding stress enters the watchlist.
Risk Heartbeat May 2026 — an energy supply shock pushes the reflationary-surprise regime to modal, cross-border payment-system stress graduates to a flagged signal, and non-bank leverage enters the watchlist.
Practitioner's playbook for time-series forecasting in finance — feature engineering, target leakage, drift monitoring, and three diagnostics to run now.
VaR vs. CVaR in stressed regimes — a benchmark on 2008 and 2020 with breach rates, magnitudes, and three notes for running CVaR in production.
Risk Heartbeat April 2026 — commodity-financing dispersion graduates, second-order Financials contagion partially resolves, new payment-system signal.
A free-tier sector risk lens — aggregate idiosyncratic risk, Risk Heartbeat coherence, sentiment compression, microstructure and regulatory signal by sector.
Two-tier AI agents — a cheap triage classifier in front of a deep LLM extractor. Cost, precision, and governance benefits, with the full design pattern.
Technical walkthrough of Issuer Scout, Microstructure Watcher, Regulatory Crawler — the AI agent stack architecture behind Risk Heartbeat at DF Analytics.
Risk Heartbeat March 2026 — Financials signal priced in, commodity-financing dispersion graduates, EU AI Act clarifying note on agent documentation.
An 11-section MRM documentation template for LLM-based agents — practical, continuous, runtime-generated. Aligned with SR 11-7 and the EU AI Act.
EU AI Act readiness checklist — 30 items across inventory, data, robustness, transparency, documentation, and vendor governance for asset managers in 2026.
ai governance financial services
Nine operational principles for AI governance in financial services — explainability, model registry, challenger models, drift monitoring, audit-readiness.
risk heartbeat february 2026
Welcome to issue #02 of Risk Heartbeat. February is a short month and a busy one — three of the four signals we covered in January moved meaningfully, and one materially new signal entered our watchlist. The format is the same as last month: what changed, three flagged signals, one macro
factor models institutional risk"
Where factor models still earn their place in 2026 and where they hide risk — residual concentration, stale loadings, non-stationary premia, and the fix.
idiosyncratic risk scoring
Epsilon scores idiosyncratic risk for every issuer in the investable universe, daily, using two-of-three coherence across price, narrative, and microstructure.
inancials liquidity stress signatures
Aspects Q1 2026 — four early-warning Financials liquidity stress signatures, the issuer clusters they identify, and the portfolio implications.