Do mortgage brokers need a penetration test?
Yes, in practice. The FTC Safeguards Rule names mortgage brokers as a financial institution, and one way to meet its security-testing requirement is an annual penetration test plus vulnerability assessments every six months. If you are licensed in New York, the DFS Cybersecurity Regulation expects annual testing on top of that.
You handle the kind of data these rules were written to protect. A single loan file can carry a borrower's full income history, two years of bank statements, a Social Security number, a pulled credit report and the wire instructions for a closing. That is exactly the nonpublic personal information the Safeguards Rule covers.
Grapefruit runs the test these rules point to: a real attack on your systems, with every finding proven, at a fixed price, with the report in about 10 days.
What does the FTC Safeguards Rule require?
The Safeguards Rule, at 16 CFR 314.4(d), gives you a choice. You can continuously monitor your systems, or you can do both annual penetration testing and vulnerability assessments at least every six months, and again after any material change. Most brokerages pick the testing path because it produces a clean, dated record.
The rule defines a penetration test at 16 CFR 314.2 as a method where testers try to defeat your security from outside or inside your systems. That is what Grapefruit does: authenticated testing with your own test accounts, attacking the broker portal and its APIs the way a real intruder would.
- Annual penetration testing: one full Grapefruit test a year satisfies this half.
- Vulnerability assessments every six months: a Grapefruit maintenance plan re-tests weekly or monthly, well inside the six-month line.
- After material changes: a plan already re-tests after you ship, so a new lender integration or portal feature is covered on the next run.
- In force since June 9, 2023, so the testing expectation is current, not upcoming.
Are you a covered financial institution?
Almost certainly. The rule and FTC guidance describe financial institutions broadly and name mortgage brokers and mortgage lenders directly. If you take loan applications, pull credit, or move borrower money, you are in scope. The relevant definition sits at 16 CFR 314.2.
There is one narrow exit. Under 16 CFR 314.6, a firm that maintains customer information on fewer than 5,000 consumers is exempt from the testing requirement. Count carefully: borrowers, co-borrowers and prior applicants add up fast, and most active brokerages cross the line.
No certification is required to run the test. The rule asks for a competent tester, not a licensed one, because there is no federal penetration-testing license.
Does New York add its own rule?
If you hold a New York mortgage broker or banker license, yes. The DFS Cybersecurity Regulation at 23 NYCRR 500.1 covers entities licensed under the state's Banking Law, which includes licensed mortgage brokers and bankers. Its testing requirement is separate from the FTC's.
Section 500.5 calls for penetration testing at least annually, from both inside and outside your systems' boundaries, by a qualified party. New York does not set a six-month scan cadence the way the FTC does; it ties automated scans plus manual review to your risk assessment and to material changes.
| 23 NYCRR 500.19 exemption, meet any one | Threshold |
|---|---|
| Headcount | Fewer than 20 employees and independent contractors |
| Revenue | Under $7.5M gross annual revenue in each of the last three fiscal years from NY operations |
| Assets | Under $15M year-end total assets |
What does a mortgage broker breach look like?
It usually starts with a borrower who was never supposed to see another borrower's file. Picture a loan status page that loads a file by its ID number in the web address. Change the number and a second borrower's income documents, bank statements and SSN load, because the server never checked that the logged-in user owned that file. That is a broken access check, and it exposes everyone at once.
The costlier version touches money. A support feature or a document upload that an attacker can reach lets them alter the wire instructions attached to a closing. The borrower sends funds to the attacker's account believing it is the title company's. These are the exact failures Grapefruit hunts, and proves.
- Loan files that load across accounts through a guessable ID, a broken object level authorization flaw.
- Tampered wire or payoff instructions that redirect closing funds.
- Credit report data exposed through an unprotected API endpoint.
- Test or staging portals left reachable with real borrower data inside.
What does Grapefruit test for a brokerage?
Everything a borrower's data touches, in one test. The web portal, the REST and GraphQL APIs behind it, and any AI assistant you have added for intake or support are attacked together, because a real attack chains across them. You get findings ranked critical, high, medium or low in an online portal.
Every finding arrives with the exact request and response that proves it. Your engineer pastes the request, runs it, and watches the attack work. After the fix, they run it again and watch it fail. That is your remediation evidence, dated and ready for an examiner.
- Authenticated testing with test accounts you supply, so we see what a logged-in borrower or loan officer can reach.
- Access-control testing across loan files, documents and user roles.
- API testing mapped to the OWASP API Security Top 10 2023.
- A downloadable report with the tested hostnames, the run date and proof for every finding.
How does one test a year plus a plan cover the rule?
The Safeguards Rule's two halves map onto Grapefruit's two offers. One full penetration test a year satisfies the annual testing requirement. A Grapefruit maintenance plan re-tests your verified domains every week or month, which sits comfortably inside the six-month vulnerability-assessment line and dates each run for your file.
So one full test a year plus a maintenance plan covers both halves of the rule, and the New York annual requirement at the same time. See pricing to configure a full test, or to start a plan.
How do you start?
Add your hostname, publish one DNS TXT record and sign the testing authorization online. It takes a few minutes. Testing without authorization is a federal crime under the Computer Fraud and Abuse Act, 18 U.S.C. 1030, which is why Grapefruit takes a signed authorization and a defined scope before any test, and why setup is this quick.
Results land in your portal within hours of kickoff. Pick a full penetration test at a fixed quote, or a maintenance plan for weekly or monthly re-tests. This is a summary of public rules, not legal advice.
Questions
- Is penetration testing required for mortgage brokers?
- The FTC Safeguards Rule names mortgage brokers as financial institutions and, at 16 CFR 314.4(d), lets you meet its testing requirement with annual penetration testing plus vulnerability assessments every six months. New York licensees also face annual testing under 23 NYCRR 500.5. Grapefruit runs that test at a fixed price with the report in about 10 days.
- Are small brokerages exempt?
- Possibly. Under 16 CFR 314.6, a firm maintaining customer information on fewer than 5,000 consumers is exempt from the testing requirement. Count borrowers, co-borrowers and past applicants, because the number climbs quickly. In New York, 23 NYCRR 500.19 offers a separate exemption from the testing rule for firms under 20 employees, under $7.5M NY revenue, or under $15M in assets.
- What borrower data is at risk in a brokerage?
- A loan file can hold income history, two years of bank statements, a Social Security number, a pulled credit report and closing wire instructions. That is the nonpublic personal information the Safeguards Rule protects. Grapefruit tests whether any of it can be reached across accounts, through your APIs, or through an AI intake assistant.
- Do I need a certified tester for the FTC rule?
- No. The Safeguards Rule asks for a competent tester, not a certified or licensed one, and there is no federal penetration-testing license. 23 NYCRR 500.5 similarly requires a qualified party without naming a certification. What matters to an examiner is a real test with evidence, which is what a Grapefruit report delivers.
- How often do I have to test?
- Under the FTC path, annually for the penetration test and at least every six months for vulnerability assessments, plus after material changes. New York expects annual testing with scans at a cadence your risk assessment sets. A Grapefruit maintenance plan re-tests weekly or monthly, which covers every one of those intervals with a dated run.
- Can Grapefruit test the AI assistant on our site?
- Yes. If you have added a chat assistant for intake or borrower support, Grapefruit attacks it in the same test as your portal and APIs. A crafted message that talks an assistant into fetching another borrower's file is a real path, and it is reported alongside your web and API findings in one place.