Do insurance agencies need penetration testing in New York?
Yes, unless you qualify for a limited exemption. The NY DFS Cybersecurity Regulation, 23 NYCRR 500.5, requires covered entities to have penetration testing at least annually, from both inside and outside their systems' boundaries, by a qualified internal or external party. The Second Amendment that set this took effect November 1, 2023.
Licensed insurance agencies and producers are covered. Under 23 NYCRR 500.1, a covered entity is anyone operating under a license, registration or charter under New York Banking Law, Insurance Law or Financial Services Law. An agency or producer holding a New York insurance license sits squarely inside that definition.
The FTC Safeguards Rule can reach insurance businesses too, where they are significantly engaged in financial activities. For a New York agency, 500.5 is the sharper line, since it names annual penetration testing in plain terms and ties the ongoing scanning to your own risk assessment.
What makes an agency a covered entity?
Your New York insurance license is the trigger. 23 NYCRR 500.1 defines a covered entity as any person operating under a license, registration or charter under the Banking Law, Insurance Law or Financial Services Law. Agencies, brokers and individual producers licensed to place coverage in New York are covered entities and owe the Part 500 obligations.
- Insurance agencies and brokerages licensed under New York Insurance Law.
- Individual producers and agents holding a New York license.
- Managing general agents and wholesalers operating under a New York license.
- Entities that also fall under the FTC Safeguards Rule where they handle consumer financial data.
What applicant data does an agency hold?
An agency collects some of the most sensitive data a consumer ever hands over, because underwriting needs it. A single applicant file can carry health, financial and identity details together, which is why a breach of an agency hits harder than its size suggests.
- Health questionnaires, conditions and medical history for life and health lines.
- Social Security numbers, dates of birth and driver's license details.
- Financial data: income, assets, bank and payment information for premium finance.
- Property, vehicle and business details tied to the risk being covered.
- Beneficiary and household information gathered during the application.
A breach scenario: one applicant's health file opens another's
Agencies increasingly take applications through a client portal or an agency management system with web access. The weak point is the document or application view. If an applicant can open their health questionnaire at an address that carries their record number, and the server does not confirm who is asking, changing that number loads another applicant's file: their conditions, their Social Security number, their finances. This is broken object level authorization, API1 in the OWASP API Security Top 10 2023.
Health data makes the fallout worse than a typical lead leak. One ownership-check gap on the API behind the portal lets a script walk every record number and collect medical histories across your whole book, the kind of exposure a DFS covered entity is expected to find before an attacker does.
Grapefruit proves it by exploiting it with a test account you supply, then captures the request and response behind the finding. Your developer or your software vendor pastes the request, watches it return a file it should not, adds the ownership check, and runs the request again to confirm it now fails.
What does 500.5 require, and what does it not?
23 NYCRR 500.5 asks for two things that are easy to confuse. First, penetration testing at least annually, from inside and outside your systems' boundaries, by a qualified party. Second, automated scans plus manual review, run at a frequency your risk assessment sets, and after any material change. The annual test and the ongoing scans are separate obligations.
New York does not set a six-month scanning cadence. That six-month figure belongs to the FTC Safeguards Rule. Under 500.5, you decide how often to scan based on your own risk assessment, and you re-examine after material changes. A qualified party is required, and no certification is named.
- Penetration testing at least annually, from inside and outside the boundary, by a qualified party.
- Automated vulnerability scans plus manual review at a risk-based frequency you set.
- A fresh assessment after any material change to your systems.
- No fixed six-month scan rule under NY DFS; that cadence is FTC only.
Are you under the 500.19 limited-exemption thresholds?
Some small agencies qualify for a limited exemption. Under 23 NYCRR 500.19(a), a covered entity is exempt from 500.5 if it meets any one of three thresholds. Meeting a single threshold is enough. Even when you qualify, the applicant health and financial data you hold is exactly what an attacker wants, so testing still earns its cost.
| Meet any one and you are exempt from 500.5 | Threshold |
|---|---|
| Headcount | Fewer than 20 employees and independent contractors |
| New York revenue | Under 7.5 million dollars gross annual revenue from NY operations in each of the last three fiscal years |
| Total assets | Under 15 million dollars in year-end total assets |
How does Grapefruit meet 500.5 for your agency?
Grapefruit tests your agency from both inside and outside the boundary, as a qualified party, which is the heart of the annual 500.5 obligation. One full test a year plus a Grapefruit maintenance plan covers the regulation cleanly: the full test is your annual penetration test, and the plan handles the ongoing scanning at whatever cadence your risk assessment sets.
Because the plan re-tests your verified domains every week or month, you clear any risk-based frequency you choose and pick up material changes on the next run. Each run is dated in your portal with findings ranked critical, high, medium or low and the request-and-response proof behind them, ready evidence when DFS or a carrier asks.
- Inside-and-outside annual penetration test by a qualified party, satisfying 500.5.
- Ongoing re-tests weekly or monthly, matching the risk-based scan frequency you set.
- A dated record of every run kept in your portal.
- Findings with captured request and response, replayable before and after a fix.
How do you start?
Add your hostname, publish one DNS TXT record and sign the testing authorization online. It takes a few minutes. Choose a full penetration test at a fixed quote for your annual 500.5 test, with the report in about 10 days, or a maintenance plan for weekly or monthly re-tests that cover your risk-based scanning.
Plan prices are on the page with card checkout, so there is no demo call to schedule. See pricing to choose a plan, or configure a full test for your fixed quote. Testing a system without authorization is a crime under 18 U.S.C. 1030, so Grapefruit works from a signed scope. This is a summary of public rules, not legal advice.
Questions
- Does NY DFS require insurance agencies to run penetration tests?
- Yes. 23 NYCRR 500.5 requires covered entities to have penetration testing at least annually, from inside and outside their systems' boundaries, by a qualified party. Licensed New York insurance agencies and producers are covered entities under 23 NYCRR 500.1, so the obligation applies unless you meet a 500.19 limited exemption.
- Does New York require six-month vulnerability scans like the FTC rule?
- No. The six-month scan cadence belongs to the FTC Safeguards Rule, not NY DFS. Under 23 NYCRR 500.5, you run automated scans plus manual review at a frequency your own risk assessment sets, and again after any material change. The annual penetration test is separate and fixed; the scan frequency is yours to justify.
- Which small agencies are exempt from 500.5?
- Under 23 NYCRR 500.19(a), an agency is exempt from the penetration testing requirement if it meets any one of three thresholds: fewer than 20 employees and contractors, under 7.5 million dollars in New York revenue in each of the last three fiscal years, or under 15 million dollars in year-end total assets. One threshold is enough.
- What counts as a qualified party for the test?
- 23 NYCRR 500.5 requires a qualified internal or external party but names no certification, and there is no US penetration-testing license. What matters is a tester who can actually defeat your security from inside and outside and show the path. Grapefruit performs authorized testing under a signed scope, since testing without permission is a crime under 18 U.S.C. 1030.
- Does the FTC Safeguards Rule also apply to our agency?
- It can. Insurance businesses significantly engaged in financial activities can fall under the FTC Safeguards Rule as well. For a New York agency, 500.5 is usually the clearer obligation because it names annual penetration testing directly. One Grapefruit annual test plus a plan covers the DFS requirement and supports Safeguards coverage where it applies.
- How much does insurance agency penetration testing cost?
- Grapefruit quotes a full penetration test at a fixed price with the report in about 10 days, and lists maintenance plan prices on the pricing page with card checkout. Traditional web application pentests commonly run 5,000 to 30,000 dollars, according to Astra's cost guide, and more for complex multi-role apps. See pricing for current figures.