We released advanced slider mode this week — a long-requested capability that lets enterprise compliance teams set per-category risk scores manually, rather than relying on automatic preset generation. Available today to enterprise customers through the API and partner console integrations.
The feature is straightforward: ten sliders, one per category, 0–100 each. Scores roll up to the total via our weighted methodology, with the level (green / orange / red) derived from the total score using the standard thresholds. Changes save with the usual audit log entry, and the updated scores flow immediately into all downstream reports.
Why did we build it? Customers with mature compliance programmes told us that preset mode — pick a primary category, pick a level, system generates plausible scores — was not precise enough for their highest-stakes wallets. A wallet might be orange because of a single high-severity signal; or it might be orange because of moderate exposure across three categories. Those two cases produce the same aggregate level but warrant different analyst treatment, and that nuance was getting lost.
Advanced mode gives the granularity back. You can, for instance, score a suspected Lazarus intermediate as 95% stolen funds, 60% darknet, 45% mixer, and 12% sanctions, and that scoring will survive into every PDF, every API response, and every dashboard view. Your investigators' judgement is encoded faithfully into the data layer.
For customers already using preset mode, nothing changes. The two modes coexist — we expect preset to remain the fastest path for the majority of records, with advanced used selectively. A switcher at the top of the wallet editor lets you toggle as you work.
Behind the scenes, we also invested in the UX details: live preview of the total score as you move sliders, a derived level badge that updates in real time, and a "regenerate from preset" option if you want to start from an auto-fill and tune from there. We took inspiration from audio-engineering mixers — the idea that each input should be independently tuneable while the master fader shows the aggregate.
What's next? We're working on three things. First, risk profiles — saved slider configurations that you can apply with one click across a batch of addresses. Second, bulk edit — apply a delta to dozens of records at once. Third, an ML-assisted scoring suggestion that reviews the address's on-chain activity and proposes a starting score for analyst adjustment. All three are in internal dogfood and should ship in Q3.
As always, feedback welcome at product@amlregister.com. The advanced mode ship was directly shaped by customer feedback; we take input seriously.
