Skip to content

Content Library

Back to Articles

Article April 29, 2026

Stress testing illusions exposed as AI reshapes model risk

Stress testing illusions exposed as AI reshapes model risk

Tanveer Bhatti, Model risk, Revolut

Traditional stress testing models are failing under extreme conditions, as AI accelerates complexity and uncertainty. Tanveer Bhatti argues that banks must rethink model risk management, focusing on resilience, human judgment, and system-level oversight t
  • Traditional stress testing models fail under extreme and systemic shocks
  • Models struggle to capture non repeatable events and hidden vulnerabilities
  • Cascading risks require network level analysis not isolated model testing
  • Scenario design must be coherent and grounded in causal mechanisms
  • AI introduces new risks and cannot detect its own errors
  • Future model risk management must focus on system resilience and failure controls 

The credibility of stress testing frameworks in modern financial institutions is under growing scrutiny, as extreme events expose fundamental weaknesses in how models are designed, governed, and interpreted.

According to Tanveer Bhatti, Founder and Senior Adviser on AI Governance, Model Risk and Regulatory Readiness at Fintech Investor, the challenge is not simply technical, but philosophical.

“I saw this managing model risk in a G SIB during 2008,” Bhatti said. “The models kept outputting numbers right through the crisis. Validity collapses at the moment you need it most.”

Bhatti, who was formerly Group Head of Model Risk at Revolut, said the experience of both the global financial crisis and the COVID shock revealed a deeper flaw in how institutions rely on quantitative frameworks.

Models built on repeatable patterns are being used to assess events that are inherently unpredictable.

“You're asking models built on repeatable mass phenomena to price non repeatable events,” he said. “That's a category error disguised as a quantitative one.”

Instead of relying on point forecasts, Bhatti argues that institutions must design frameworks that are capable of failing safely.

This includes setting explicit uncertainty bounds, identifying key sensitivities, and implementing hard circuit breakers when outputs fall outside credible ranges.

The challenge becomes even more acute when considering interconnected risks. During periods of stress, financial systems behave less like isolated components and more like tightly coupled networks, where disruptions cascade across functions.

“If the stress framework doesn't model that chain, you're testing nodes and calling it a network test,” Bhatti said, pointing to the way margin calls, funding pressures, and client behavior interact dynamically.

Capturing these cascading effects requires a more pragmatic approach. Bhatti emphasized the importance of triangulation, using multiple modeling techniques to approximate complex system behavior rather than relying on a single methodology.

“Analytical rigor is useless if it can't be run and explained and governed,” he said.

Scenario design, long considered the backbone of stress testing, is also evolving. Bhatti warned against simply stacking adverse outcomes without a coherent narrative, arguing that severity must emerge from the mechanics of the scenario itself.

“Plausibility doesn't come from historical frequency,” he said. “It comes from a coherent causal story with mechanics that are defensible.”

He added that scenarios must also withstand scrutiny from senior management and regulators, reflecting the institutional realities of decision making.

The rise of artificial intelligence is adding another layer of complexity. While AI can generate scenarios and identify weak signals, Bhatti cautioned that it cannot determine when it is wrong.

Events you may be interested in

View all of our upcoming events

Sign Up to Our Newsletter

Sign up

Connect Plus

Unlock more with Connect Plus membership

This content is available to Connect Plus members.