Interactive analysis of where agentic AI can drive operating-efficiency and expense-productivity gains across Citi's wholesale businesses, with financial estimates grounded in Citi's disclosed FY2025 figures and external benchmarks.

Wholesale Banking · Operations Technology

Turning Citi's proven AI momentum into measurable operating-efficiency gains.

An opportunity map for agentic AI across Services, Markets and Banking — grounded in Citi's disclosed FY2025 financials and real industry benchmarks.

$1.5–3.0B
est. AI-attributable run-rate at maturity
~3–5%
of the $55.1B expense base
8
prioritized opportunities · 2026–2028
▼ Scroll to explore
The mandate

Drive AI strategy, operating efficiency, and expense productivity.

Citi delivered record FY2025 revenue and positive operating leverage across all five businesses — while committing to keep bending the cost curve. The wholesale franchise, with its scale and process intensity, is where AI compounds fastest.

$0B
FY2025 total operating expense
0%
FY2025 efficiency ratio
target ≈ 60% in 2026
$0B
run-rate savings still to capture (≈$2–2.5B)
0%
2026 RoTCE target (10–11%)
The wholesale franchise

Three businesses, one operations spine.

Services — the engine

Treasury & Trade Solutions (cross-border / international payments, USD clearing) + Securities Services. FY2025 revenue $21.3B · opex $10.8B · cross-border value $416B.

Markets

Fixed Income + Equities (incl. Prime). High-volume, low-latency. FY2025 revenue $22.0B · opex $14.1B.

Banking

Investment Banking (M&A, ECM, DCM) + Corporate Banking. FY2025 revenue $8.2B · opex $4.5B.

Shared spine: onboarding/KYC, reference & client data, trade & payment processing, reconciliations, settlement, controls, and regulatory reporting — where most addressable expense (and AI opportunity) concentrates.

The opportunity map

Eight opportunities, ranked by impact & feasibility.

Each estimate is built bottom-up from a disclosed segment cost pool × an addressable share × a conservative AI savings rate — and anchored to a documented "quantum-leap" benchmark. Click any card to expand.

Live ROI model

Pressure-test every assumption.

Pick an opportunity and adjust the addressable cost pool and AI savings rate. The financial impact and the portfolio total recompute live. Defaults reflect the conservative ranges in the analysis.

$0.2B$6B
0%40%
This opportunity → / yr run-rate
Portfolio run-rate potential (all 8, at maturity)
—
Not strictly additive (KYC & controls overlap segment rows). AI is one lever within Citi's ~$2–2.5B savings target. Benchmarks suggest these ranges are conservative.
See it in action

Agentic AI on a live wholesale case.

From copilot to autonomous multi-step agent — grounded in Citi data and bank-grade governance. Pick a scenario and run the agent.

Before · manual
After · agentic
Basis

Ready
Prioritization

Impact vs. feasibility.

Bubble size = estimated annual run-rate impact. Hover any bubble for detail. Sequencing weighs ROI certainty, cost-pool size, and data/regulatory dependencies.

Sequencing

A 2026–2028 delivery roadmap.

A discussion starter

One way to frame the first 90 days.

The inside view — which you and the team have — decides the real sequence. This is an outside-in frame to react against: rough order of magnitude throughout, and much of it likely already in flight.

Weeks 1–4
Listen & baseline
Baseline the spineSTP by flow · exception volumes & aging · cost per payment / trade / KYC file
Capacity auditWhat deployed AI (code review, Stylus) has freed — and where it went
Governance runwayWhat Model Risk & Compliance need to see before agentic automation ships
Weeks 5–8
One or two lighthouse wins
Scale the SDLC leverHighest certainty — already proven inside Citi
Payment-exception & screening triageCross-border flow · visible in-year
Ship inside a quarterSmall enough to deliver, visible enough to anchor the mandate
Weeks 9–12
Lock the scaling frame
One-page scorecardRun-rate vs. $2–2.5B target · STP · cost per exception · hours redeployed · adoption · MRM cycle time
Self-funding logicEarly savings finance the next wave
Into the 2027 planGovernance pattern pre-agreed with Model Risk
Week-one questions — likely already in flight
1. Where does STP break most, by flow and business — and what does each break cost?
2. What is the exception aging profile in payments and post-trade, and the cost per case?
3. How much capacity has deployed AI actually created — and where was it redeployed?
4. Which client- and reference-data gaps block the next wave of use cases?
5. What evidence will Model Risk & Compliance expect before autonomous steps run in regulated flows?

Illustrative by design — the real sequence, numbers, and scorecard belong to the team. The intent: show where hands-on outside help could contribute from week one.

The basis

Every estimate has a documented quantum leap behind it.