Checks run while the customer types. Every issue fixed here is an investigation that never opens.
IBAN and BIC checks use the real ISO checksum and format rules. The name check stands in for an account pre-validation service and is simulated here.
Every held payment becomes one case with one timeline. The copilot gathers and drafts. The analyst decides. The customer's screen updates the moment they do.
The customer's view, live
Every held payment tells the customer what is happening, what happens next and by when. When we need something, we ask once.
Four clear steps tracked through the payment's UETR, from our app to the receiving bank.
An expected time, or the time of the next update. Silence is what drives support contacts.
One checklist with upload and one-tap answers. Answers are remembered for that beneficiary.
Sensitive reviews use approved neutral wording only. No free text, no reasons.
Every input is an assumption you can change. Customer numbers come from Wio's 2025 results. Volumes, rates and handling times are my estimates, to be replaced with real data in the first two weeks.
Most of these ideas exist somewhere. Leaders usually run them as separate projects. The opportunity for a cloud-native bank is to join them into one customer journey on one data spine.
| Capability | Global leaders | In the UAE | Clearway for Wio |
|---|---|---|---|
| Check the account and name before sending | UK Confirmation of Payee (since 2020, about 1.9 million checks a day). EU Verification of Payee, required for euro transfers since October 2025. J.P. Morgan's Confirm account validation. | Aani lets people pay by mobile number, so fewer IBANs are typed by hand. Name checks on IBANs are still emerging. | Pre-flight checks at submit: IBAN and BIC rules, account and name pre-validation, purpose suggestion. |
| Track every international payment | Wise shows every step. 77% of its transfers arrive in under 20 seconds and 97% within 24 hours. Swift gpi: 90% of payments reach the end bank within an hour. | Mashreq was the region's first bank on Swift gpi, with tracking in its app. Wio's clearing partner offers real-time tracking on USD, EUR and GBP. | A parcel-style tracker driven by the UETR, plus a proactive alert when something stalls. |
| AI on screening alerts | HSBC with Google: 2 to 4 times more financial crime found and 60% fewer false positives. | Emirates NBD automated its screening-alert investigations with Silent Eight (2024). | A copilot that ranks and drafts, with a human decision on every case. Auto-closure only for a validated, lowest-risk lane. |
| Fewer checks for known, low-risk patterns | Wise skips extra steps for transfers to trusted recipients and to the customer's own accounts. | Mostly handled case by case today. | Alert memory for cleared pairs, plus remembered answers per beneficiary. |
| Fast domestic money | UK Faster Payments, Brazil's Pix, India's UPI. | Aani: instant payments in under 10 seconds across 57 institutions. | The same tracker and status language for domestic and international, so the experience is consistent. |
The UK's habit of checking the name before money moves, Wise's transparency on every step, and HSBC's discipline of measuring both fewer false positives and more true hits.
Transliterated Arabic and South Asian names, many first-time beneficiaries for SMEs and freelancers, and purpose-of-payment codes. Matching and messaging have to be built for this, in English and Arabic.
Incumbents bolt these on one by one. Wio can join prevention, triage and customer messaging on one platform, and be the bank where a held payment never feels like a black box.
An event backbone joins every system on the payment's UETR. A case layer turns each held payment into one case. Rules, models and an AI copilot sit on top, and a person makes every decision.
Across every layer: tracing, access control, model evaluation, and data kept in the UAE. Illustrative. I would fit this to Wio's existing cloud platform.
| Problem | Approach | What it does | Who decides |
|---|---|---|---|
| Typos and format errors | Rules: IBAN checksum, bank code and country checks | Exact and explainable, needs no training data | Customer, with one tap |
| Purpose and documents | Classifier on payment history, plus document AI that reads invoices | Suggests the purpose and checks invoice amount and reference | Customer confirms |
| Name matching | Spelling-aware fuzzy and phonetic matching, plus secondary identifiers | Handles Arabic and South Asian name variants | Analyst |
| Which alert first | Gradient-boosted model trained on past analyst decisions, with a reason for every score | Ranks by false-positive likelihood. Never closes an alert by itself in phase 1 | Analyst |
| Which payment will stall | Time-to-credit model by corridor and partner bank | Flags delays before the customer notices, drafts the chase | Operations |
| Case file and messages | LLM copilot with read-only tools and search over case history and policies | Gathers evidence, drafts the rationale and the customer question | Analyst edits and approves |
| Reconciliation breaks | Rules plus anomaly detection on fee patterns | Clears known fee gaps, surfaces the odd ones | Operations |
A Kafka pipeline carrying 115 GB a day from 200+ offshore stations with zero data loss. Bad records were quarantined with context for replay, cutting incidents by 25%.
An AI agent limited to least-privilege tools, with a person holding final authority over every change to research data.
Answer quality up 40%, graded on held-out test sets. Response time cut from 450 ms to under 100 ms at 3,000+ requests a second.
Self-service for 500+ vendors replaced a manual queue. Support tickets fell 35% and service requests fell 45%.