The latest from product teams at Bretton AI.
Payment screening now recognizes a broader range of phonetically equivalent name variations across romanized spellings.
Phonetic spelling differences can create avoidable mismatches and missed matches in global screening programs.
More accurate person-name matching with less manual review, enabled automatically for payment-screening customers.
Agents can now return consistent, structured results alongside the narrative, configured to fit the systems and workflows that use them next.
Consuming agent results in another system meant parsing free text or prompt-engineering the format.
Integrate agent results into your own systems in whatever format you need, which speeds onboarding.
Citations can now be shown or hidden on the narrative page and in exported PDFs and workbooks, in bulk from the investigations and agents page or one case at a time.
Analysts move finished narratives into case-management systems and sometimes need the text without inline citations.
Choose whether citations appear in what you export or copy, without reformatting by hand.
Workbench is now available as an assistant on completed agent cases, with access to the case data, the organization's skills, artifacts, and direct narrative updates.
Analysts had to leave the case or move findings by hand to dig deeper.
Investigate, reason, and update a case narrative without leaving the case.
You can rerun a case from the point that needs attention, update relevant inputs, and preserve the original case history.
Changing an input or configuration meant creating a whole new case and losing history.
Rerun only the stages you need, and keep the audit history intact.
An agent can now choose a narrative template based on the case's final disposition, from a list you configure, rather than using one narrative for every outcome.
Different outcomes need different write-ups.
Narratives match the disposition automatically. Off by default, configurable in Builder.
Transaction monitoring now includes a native workspace for reviewing activity, prioritizing alerted transactions, and exploring patterns without leaving Bretton.
Teams were exporting transactions to spreadsheets to analyze them, which was slow and hard to audit.
Review and analyze monitored transactions without leaving Bretton or building a spreadsheet.
Investigation narratives are now fully editable as formatted text. Citations render as clickable footnotes, tables are editable, and citations renumber automatically as you add or remove them.
Editing used to be piecemeal and easy to get out of sync.
Analysts edit and finalize a narrative in one place, with citations and an audit log that stay correct.
Customer metadata now appears directly in the case attachment view, with a download option.
Analysts had to look this up separately.
Less context switching during a review.
Bulk alert ingestion now supports more screening workflows and secure enterprise delivery methods for large alert volumes.
Enterprise programs move alerts in large, sensitive batches and need secure, reliable ingestion.
Move very large alert volumes into Bretton securely, with errors returned for correction.
A new operating model that combines AI-native execution with experienced subject-matter experts, human oversight, quality control, and program development.
Financial institutions need to adopt AI without taking on the capital investment and operational burden of building a new program alone.
Bretton's forward deployed engineers and industry experts build, train, tune, and operate AI-native programs around customer outcomes.
Build, run, and govern AI agents across the financial back office, with managed deployment and audit-ready output from day one.
The banking back office no longer has to grow every time the bank does.
Financial institutions can transform the back office with agent-first operations built around their data and policies.
Builder now lets teams tailor agent behavior to the role each entity plays in an investigation.
One generic skill prompt for every entity type produced noisy findings.
More precise multi-entity investigations.
Agent runs now include a persistent, chronological record of the actions and evidence behind each result.
Teams need a clear record of what an agent did and how it reached its output.
Agent work is easier to review, trust, and explain.
Builder now makes it easier to configure transaction monitoring agents around an institution's existing transaction data.
Standing up a transaction monitoring agent for a new data format took custom engineering.
Stand up transaction monitoring agents faster, with the agent working from the full transaction data.
Screening-alert exports now align with the current table view and support more flexible filtering.
Exports did not line up with the on-screen view.
Export exactly what you are looking at, filtered how you need it.
Builder is live. Teams can create, configure, test, and publish agents for workflows like enhanced due diligence and transaction monitoring, without engineering. Launched as Agent Builder.
Creating and updating agents used to require engineers.
Business and operations teams stand up and maintain their own agents. Use cases expanded beyond financial crime into compliance, fraud, risk, and other banking operations.
New match tags give programs finer control over the entity attributes used during screening.
Programs need granular control over what counts as a match.
Tune matching and dispositions to your risk rules, from the tag settings.
Investigation artifacts can be exported to a shareable, high-quality PDF, with sources included.
Moving investigation output into compliance workflows meant manual copy and paste, with formatting drift.
Download a clean, sourced report in one step.
Organizations can upload their own documents, policies, procedures, and guidance into a per-organization knowledge base that Workbench searches and reads during an investigation.
Anything customer-specific, like an internal PEP-escalation policy, used to live outside the tool in the analyst's head.
Drop your compliance documents in once, and investigations use them automatically.
Prompt templates now support variables, such as a business name or website, that you fill in when you use the template.
The same investigation setup applies to many different entities.
Reuse a setup across cases without rewriting it each time.
Bretton now evaluates research and enrichment quality against verified reference data and uses reviewed outcomes to improve performance over time.
Research quality has to be measured, not assumed, especially where the work is open-ended.
Customers receive more consistent, well-grounded research and richer entity profiles over time.
Workbench now provides a detailed, reviewable record of investigation activity and clearer sourcing behind enrichment results.
Users could see loading states but not an in-depth record of the investigation.
A transparent, reviewable record of every investigation.
You can attach PDFs, spreadsheets, and other documents directly in an investigation by drag-and-drop or file picker, and ask the agent to analyze them.
Analysts had to copy and paste document content or switch tools.
Upload shareholder registers, incorporation documents, or statements and analyze them in place.
Transaction monitoring now completes reviews more efficiently, brings relevant counterparty context into the workflow, and supports editable narratives in the interface.
The older pipeline was slow and costly on high-volume cases and locked the narrative to a fixed template.
Much faster, lower-cost transaction monitoring, with editable output.
Workbench now offers default prompt templates and lets analysts save their own.
Analysts repeat similar investigations often.
Faster, more consistent starts to common investigations.
Country and jurisdiction names now resolve to a single canonical standard across the platform, so variants like Ivory Coast and Cote d'Ivoire, or Myanmar and Burma, are treated as the same place.
Conflicting country data will cause avoidable mismatches.
More accurate name and jurisdiction matching.
Payment screening now handles a broader range of Chinese name variations and cross-language representations.
These matches previously all fell into ‘needs investigation.’
Higher automation and less manual review for programs screening Chinese entities.
Payment Screening is now a first-class product for reviewing payment parties against sanctions and watchlists, resolving alerts efficiently, and working with existing payment-screening systems.
Payment screening is a false-positive-heavy workflow where slow or unclear review can hold up legitimate funds.
Faster, clearer payment-screening review, in-product and by API.
Screening-alert citations now render as complete, working links.
Citations were sometimes missing or malformed, which undercut auditability.
Analysts can trust the citations and click through to the source.
Authorized teams can now manage screening settings and disposition rules directly in the platform, with access controls and support for program-specific configurations.
Changing screening behavior used to require a manual request.
Compliance teams tune their own rules and tags, with clear limits on who can edit.
Organizations can tailor how politically exposed person matches are classified and routed for review based on their risk policy.
PEP risk appetite varies by institution, and analyst attention should go to the highest-risk cases.
More control over PEP decisioning from the Screening Alerts configuration experience.
Document recency and expiration checks now use an explicit reference date, so valid documents are no longer wrongly flagged as expired or future-dated.
Ambiguity about the current date was producing false flags.
Fewer false positives and more reliable KYC and KYB onboarding assessments.
Adverse media can now be searched in an entity's local language, with results translated back to English for review.
Critical adverse media often exists only in local-language sources.
Broader regional coverage and no manual translation, so important local information is not missed.
Enhanced due diligence can now assess more than one website for a single company entity.
Many businesses operate several websites, and assessing only one misses risk.
More complete due diligence for multi-site businesses, with less manual effort.
Organizations can securely connect relevant data from Databricks to Bretton workflows.
Much of the context that improves compliance work lives in the data warehouse.
Databricks users connect their data securely and give Bretton more to work with.
Bretton integrates with Unit 21 to bring transaction alerts and relevant case history into the review workflow.
Clients on Unit 21 needed their alerts and case history to flow into Bretton.
Teams review and act on Unit 21 cases as soon as they are created, with historical context included.
Transaction monitoring now provides deeper counterparty context and clearer alert narratives, bringing relevant review findings into one report.
Analysts were assembling this context by hand.
Faster, better-informed decisions, with the relevant context already in the report.
Enhanced due diligence can now run as a periodic review or a triggered review, carrying the cause and prior context of the re-review into the case and its narrative.
Reviewing an already-onboarded customer is different from onboarding one, and that context matters.
Run periodic and event-triggered re-reviews as a repeatable playbook, not a fresh case each time.
A new Aged Adverse Media tag helps teams distinguish older findings from more current risk signals.
Outdated adverse media generates review work that rarely changes the outcome.
Less time on stale alerts and more attention on higher-risk findings, with controls available in Screening Alerts configuration.
Employment matching now understands the meaning of employer names, so minor differences like ‘Apple’ and ‘Apple Inc.’ are recognized as the same employer.
Small naming differences were causing missed matches.
More accurate alerting and fewer false negatives, so teams decide faster and with more confidence.
Address verification now considers additional indicators that may signal address or entity risk.
A registered agent at an address can signal address clustering or shell-company risk and raises questions about whether a company really operates there.
More rigorous address analysis and more confidence in location-risk assessments.
Adverse media search now works more like an analyst doing research, using shorter, more targeted queries to find more relevant results.
The older approach missed findings a human researcher would have caught.
Broader, more relevant adverse media coverage.
Enhanced due diligence and transaction monitoring now provide a deeper view of account activity, important counterparties, and patterns that may require attention.
Line-by-line transaction review is slow and easy to lose the thread of.
Teams see the high-level trends and the flagged patterns without reading every transaction.
RFI Analysis reviews customer documentation, including identity documents, bank statements, and payslips, against your RFI criteria, and checks each document for identity, location, and income discrepancies.
Reviewing request-for-information responses by hand is slow and inconsistent.
Faster, more consistent RFI review, with discrepancies surfaced automatically.
A two-way integration with Sumsub brings alerts into Bretton and returns completed review outcomes to the existing workflow.
Sumsub users needed their alerts handled without leaving their existing setup.
Alerts flow both ways automatically, and false positives can close on their own.