AI, Tokenization, and Accelerated Settlement Reshape the Back Office
2025 was the year the post-trade agenda stopped being theoretical. The United States settled comfortably into T+1, the Digital Operational Resilience Act moved from countdown to compliance, tokenization graduated from proofs of concept to production issuance, and Washington passed the first federal stablecoin statute. None of it is finished. 2026 is the year firms execute on all of it at once.
That is the difficulty. The EU, UK and Swiss T+1 build, the US Treasury clearing mandate, tokenized collateral programs and AI deployment now compete for the same change budgets, the same data remediation work and the same scarce operations talent. AI in production is constrained by governance and data quality rather than model capability. European T+1 runs on remediated standing settlement instructions. Extended trading hours break the batch assumptions embedded in pricing, funding and reconciliation. Resilience supervision runs on a register of information that doubles as the dependency map every other program needs.
Which makes data modernization the most efficient dollar in the 2026 budget: the least visible item on the list, and the one that compounds across the other nine. The full report sets out all ten trends with the segment split across broker-dealers, investment managers, custodian banks and market infrastructure.
Selected Conclusions
• AI moves from pilots to production in post-trade operations. Fails prediction, reconciliation and corporate actions scrubbing shift from proof of concept to production, with narrow agentic pilots emerging under heavy guardrails; the binding constraints are governance and data quality, not model capability, and operations leaders now own AI control design.
• Tokenized funds and securities scale, forcing dual-rail operating models. Every firm touching tokenized assets runs traditional and tokenized infrastructure in parallel, and the seam between them is where operational risk concentrates; firms that map the seam before launching products will scale, and the rest will discover it in an audit.
• Post-trade data modernization underpins everything. ISO 20022 momentum, the extending identifier stack and golden-source architecture are the precondition for AI, T+1, clearing and resilience delivery alike, which makes data remediation the most efficient dollar in the 2026 budget.
Subscribers to the Journal may download the full report, including all 10 Conclusions.
