AegisAI Perspectives · 130

How does a fintech on Google Workspace stop business email compromise without changing MX records?

A fintech on Google Workspace stops BEC without MX changes by reading mail through the API and checking each payment change by a call to a number on file.

TL;DR
Answer
Leave MX records alone: connect an API layer that reads Google Workspace mail after delivery, and verify every payment change via a known telephone number.
Who it's for
If you run IT or security for a fintech of 20 to 500 people on Google Workspace, with a lean team, no SOC and staff who move money.
Watch out for
A hijacked partner mailbox passes SPF, DKIM and DMARC. Judge the payment request itself, and never call the number printed in the email.
What AegisAI does
It checks every payment request against the sender's history, so a request from a hijacked vendor mailbox that passes SPF, DKIM and DMARC is still judged on what it asks for.
Book a demo

See it on your own Google Workspace or Microsoft 365 tenant. No MX change.

Or get the Q3 2026 State of the AI Attack report (free, instant download)

Not ready for a demo? Check your domain (free, about 10 seconds).

How does a fintech on Google Workspace stop business email compromise without changing MX records?

A fintech on Google Workspace stops BEC without MX changes in two layers: an API connection that reads mail after delivery, and a callback plus second approver on every payment change. IC3 logged $3,046,598,558 in BEC losses in 20255.

AegisAI connects to Google Workspace through the API, with no change to MX records21. It checks every payment request against the sender's history, so a request from a hijacked vendor mailbox that passes SPF, DKIM and DMARC is still judged on what it asks for22. If you run IT or security for a fintech of 20 to 500 people on Google Workspace, with a lean team and no SOC, this is the order of work for your payment operations and treasury staff. The product side is on the inbound email security page. The general Google Workspace steps are in how to protect Google Workspace from business email compromise. The connection itself is covered in how API-based email security works without an MX record change and Google Workspace email security without MX changes. This page is the fintech case.

What you can do by the end

  • Turn on three Google Workspace settings that cut BEC exposure without touching mail flow.
  • Write a payment callback rule that uses a number on file, never the details in the email.
  • Ask your bank to recall a fraudulent wire, and report it to law enforcement within 24 hours1.

Why does fintech infrastructure face elevated BEC risk?

Because payment operations and treasury staff can move money, and a sent wire is hard to pull back. FinCEN says such transactions are often irrevocable, which renders financial institutions and their customers unable to cancel payments or recall the funds3.

The volume keeps rising. BEC complaints to IC3 numbered 24,768 in 2025, against 21,442 in 20244. Reported losses rose from $2,770,151,146 to $3,046,598,558 over the same two years.

List every person who can release a payout or change a beneficiary, and give that list to IT today.

Short on time? Watch the explainer, or book a demo.

Which common BEC attack vectors target fintech workflows?

Three show up in payment work. A spoofed domain or employee name asks for an urgent payout, the case Gmail's spoofing protection is built for12. A real partner or vendor mailbox, taken over, asks to change bank details, and it passes sender checks. Automatic forwarding, on by default in Gmail10, lets mail from a payment account keep flowing to an outside address.

The request itself often carries no link and no file: a new account number and a deadline.

Map each vector to the payment step it targets before you choose controls.

Why does a fintech keep its MX records as they are?

An MX change sends every inbound message, from customers and bank partners, through a new system before it reaches Gmail. For a fintech, a delayed or dropped message from a bank partner is an operational incident, so mail routing changes only with a tested rollback.

An API connection avoids that change. The Gmail API is a RESTful API that lets you access Gmail mailboxes and send mail15, so protection reads mail where it already sits.

Write down every system in your mail path today before anyone proposes adding one.

Why isn't email authentication alone enough, and what can an API layer see?

Because a passing sender check is not a verdict. AegisAI research found that among AI spear phishing emails that evaded filters, 72.6 percent passed DMARC25. So the request itself has to be judged against what that sender normally asks for.

An API layer reads each message after delivery16, next to what Gmail already marks: unauthenticated emails display a question mark next to the sender's name13.

Forwarding is the setting to watch on payment accounts. Automatic forwarding is on by default, and when the setting is off the forwarding option isn't available in Gmail settings. To check messages forwarded by people in your organization, use Email Log Search11.

Call back on every payment change by email, whatever the authentication result shows.

Build the BEC prevention architecture for fintech teams in this order

  1. Publish DMARC for your own domain. DMARC tells receiving servers what action to take on mail from your domain that fails SPF or DKIM: reject, quarantine or deliver14.

  2. Turn on spoofing and authentication protection in Gmail's advanced safety settings. It covers spoofing of your domain and of employee names, plus unauthenticated email from any domain.

  3. Turn off automatic forwarding for the organizational unit that holds payment operations and treasury. It is on by default.

  4. Check Email Log Search each week for messages forwarded from those accounts.

  5. Connect an API-based layer read-only. It runs in monitoring mode first, and removal needs write access your admin grants and can revoke20.

  6. Verify all payment changes and transactions in person or via a known telephone number26. Contact vendors through telephone numbers on file rather than through a number provided in emails. For financial institutions, FinCEN says they may verify suspicious emailed payment instructions by using multiple means of communication2.

  7. Require a second approver for beneficiary changes. FinCEN also says financial institutions may verify by contacting others authorized to conduct the transactions.

  8. Write the recall plan. If a fraudulent transfer goes out, contact your financial institution immediately and request a recall of the funds6. FinCEN has had greater success recovering funds reported to law enforcement within 24 hours.

Run steps 1 to 4 this week: none of them changes mail routing.

Compare audit-ready controls for fintech BEC protection

ControlWhat it catchesWhat it missesInbound email security
DMARC on your domainMail from your domain that fails SPF or DKIMA real partner mailbox that was taken overJudges the request even when sender checks pass
Gmail spoofing protectionSpoofed domain and employee namesRequests from real, authenticated accountsCatches fake executives, vendors and payment requests with no link or file18
Forwarding off, Email Log SearchMail leaving by automatic forwardA message read inside the mailboxPulls one bad email from every inbox it reached, forwards included19
Callback on a number on fileA payment change, verified via a known telephone numberNothing, if staff skip it under a deadlineNo edge. This control sits outside email, with your treasury team
Blocking before deliveryOnly possible for a system in mail flowNeeds an MX changeNo edge. It reads mail through the API after delivery

The last two rows are no edge for the inbound layer: the callback is a treasury control, and the API reads mail after delivery. Fill one copy of this table with your own settings before you decide what to add.

How AegisAI handles this

AegisAI connects to Google Workspace through the API, with no change to MX records. It checks every payment request against the sender's history, so a request from a hijacked vendor mailbox that passes SPF, DKIM and DMARC is still judged on what it asks for. One bad email is pulled from every inbox it reached, forwards included. Runs in monitoring mode first.

Book a demoSee inbound email security

See your own records: Check your domain.

How does a fintech stop one BEC attack in a worked example?

Illustrative example, figures are fictional: Larkspur Pay (fictional) is a 140 person fintech on Google Workspace with two people in IT. At 14:05 a message from its real card program partner asks treasury to send the month's settlement to a new account by 16:00. Every sender check passes, because the partner's mailbox was taken over. The API layer flags it: the partner has sent 31 settlement notices in two years and never changed bank details by email. The treasury analyst calls the partner on the number in the contract file, not the one in the signature. The partner confirms no change. With write access granted, the message is pulled from the three inboxes it reached, including one forward to the finance lead.

Walk one real payment change request from last quarter through the same steps.

How to measure it in 30 days

Track three numbers, each from a record you already keep.

  • Payment change requests received by email, and how many were verified by a callback to a number on file, from the treasury approval log.
  • Payment and treasury accounts with automatic forwarding on, and their forwarded messages, from the Admin console and Email Log Search.
  • Messages the API layer flagged that reached payment staff, from its verdict log in monitoring mode.

To see what an API layer finds on your own tenant, book a monitoring-mode demo. Put the three figures side by side on day 30 and decide on them.

Questions people ask

How does a fintech on Google Workspace stop business email compromise without changing MX records?

With two layers. AegisAI connects to Google Workspace through the API, with no change to MX records, and checks every payment request against the sender's history. Beside it, treasury verifies payment changes in person or via a known telephone number, as the FBI advises.

Does DMARC stop BEC at a fintech?

Not on its own. DMARC tells receiving servers what to do with mail from your domain that fails SPF or DKIM. Of AI spear phishing emails that evaded filters, 72.6 percent passed DMARC.

What should treasury do first after a fraudulent wire?

Contact your financial institution immediately and request a recall of the funds. FinCEN has had greater success recovering funds when BEC wires are reported to law enforcement within 24 hours.

Should a fintech turn off automatic forwarding in Gmail?

For payment and treasury accounts, yes, unless a written business need exists. It is on by default, and Email Log Search shows messages forwarded by people in your organization.

Do regulators require a specific email control for fintechs?

No neutral source cited here names one. FinCEN says financial institutions may verify suspicious emailed payment instructions by using multiple means of communication. Ask your bank partner which controls your agreement names.

Sources

  1. (FBI IC3, Internet Crime Report 2025)
  2. (FinCEN, Updated Advisory on Email Compromise Fraud Schemes, FIN-2019-A005, 16 July 2019)
  3. (FBI Private Industry Notification 20210317-001, Business Email Compromise Actors Targeting State, Local, Tribal, and Territorial Governments, 17 March 2021)
  4. (Google for Developers, Gmail API overview, 18 September 2026)
  5. (Google Workspace Admin Help, Let users automatically forward their Gmail, 26 October 2026)
  6. (Google Workspace Admin Help, Advanced phishing and malware protection, 26 October 2026)
  7. (Google Workspace Admin Help, Set up DMARC, 26 October 2026)
  8. (AegisAI, Inbound email security)
  9. (AegisAI, How to protect Google Workspace from business email compromise, 21 October 2026)
  10. (AegisAI, How API-based email security works without an MX record change, 26 October 2026)
  11. (AegisAI Blog, Google Workspace email security without MX changes)
  12. (AegisAI Blog, Half of AI Phishing Reaches the Inbox, updated 30 October 2026)

Recent posts

Every email, investigated.