SR 26-2 was the headline event of April. It will not be the last. Between now and early 2028, the regulatory calendar lands obligations across three streams simultaneously — regulatory capital, AI and model risk — in three jurisdictions on three different timetables. This is the planning view.
Download Full Briefing (PDF)The regulatory work between now and early 2028 sits across three streams that touch the same teams (regulatory capital, AI, model risk) and is supervised in three jurisdictions on three different timetables. This is the planning matrix.
| United Kingdom | United States | European Union | |
|---|---|---|---|
| Capital | Basel 3.1 from 1 Jan 2027; FRTB-IMA delayed to 1 Jan 2028. Strong & Simple final phase for SDDTs from Jan 2027. Pillar 2 rebasing data due 31 March 2026. | Basel III Endgame reproposed 19 March 2026; comments close 18 June 2026. Final rule expected late 2026; trading-book threshold raised from $1B to $5B. | CRR3 in force since January 2025; FRTB postponed to 1 Jan 2027. EBA ESG Risk Management Guidelines apply from 11 Jan 2026 (SNCIs from 11 Jan 2027). |
| AI | SS1/23 already covers AI/ML in scope. AI named a 2026 supervisory priority. AI Consortium report and biennial survey expected H2 2026. No UK AI-specific rules planned; PRA approach remains technology-neutral. | SR 26-2 explicitly excludes generative and agentic AI. Interagency RFI expected H2 2026 covering MRM and banks' use of generative, agentic and AI-based models. Resulting principles likely 2027. | The Digital Omnibus (Regulation (EU) 2026/1744, in force 27 July 2026) deferred Annex III high-risk obligations, which cover credit scoring, to 2 Dec 2027 and Annex I to 2 Aug 2028. Article 50 transparency obligations apply from 2 Aug 2026. |
| Model Risk | SS1/23 in force since May 2024; annual self-assessments and remediation continue. SS5/25 climate-risk gap analysis due 3 June 2026. SMF accountability remains the binding constraint on group MRM. | SR 26-2 replaced SR 11-7 on 17 April 2026. Principles-based, non-enforceable, narrower in scope, >$30B threshold. Validation rigour valued over structural independence. | ECB Guide to Internal Models updated 28 July 2025 with new ML section, refined IRB validation and CRR3 alignment. ESA Guidelines on ESG scenario analysis expected H1 2026. |
Items below are in chronological order of the next binding action point. Filter by stream and jurisdiction. Click an item for the full detail in the briefing PDF or, for later additions, the source or related LINXS briefing.
The window to 2028 is heavier than 2024 to 2025, and it does not flatten in the second half. Three workstreams matter most for a group risk function: Basel 3.1 model-permission discipline into January 2027; EU AI Act readiness for December 2027 — deferred, not cancelled; and a group AI policy that holds the SS1/23 line for the models MRM can validate, governs GenAI on its own terms, and stays ready for whatever the US RFI surfaces.
The risk is not failure on any one item. It is capacity.The Chief Risk Officers who finish 2027 cleanly will be the ones who kept MRM for the models it can validate, governed GenAI around the use case, and mapped their evidence once so that jurisdiction-specific submissions draw from it.
LINXS Advisory works with banks on regulatory roadmap design, SS1/23 self-assessments, EU AI Act conformity assessment, FRTB and Basel 3.1 model permission packs, and DORA TLPT scoping across the US, UK and EU.
Get in Touch Visit LINXS Advisory