All Resources
In this article:
minus iconplus icon
Share the Article

Data Leakage Detection for AWS Bedrock

July 15, 2024
4
 Min Read
Data Security

Amazon Bedrock is a fully managed service that streamlines access to top-tier foundation models (FMs) from premier AI startups and Amazon, all through a single API. This service empowers users to leverage cutting-edge generative AI technologies by offering a diverse selection of high-performance FMs from innovators like AI21 Labs, Anthropic, Cohere, Meta, Mistral AI, Stability AI, and Amazon itself. Amazon Bedrock allows for seamless experimentation and customization of these models to fit specific needs, employing techniques such as fine-tuning and Retrieval Augmented Generation (RAG).

 

Additionally, it supports the development of agents capable of performing tasks with enterprise systems and data sources. As a serverless offering, it removes the complexities of infrastructure management, ensuring secure and easy deployment of generative AI features within applications using familiar AWS services, all while maintaining robust security, privacy, and responsible AI standards.

Why Are Enterprises Using AWS Bedrock

Enterprises are increasingly using AWS Bedrock for several key reasons:

  • Diverse Model Selection: Offers access to a curated selection of high-performing foundation models (FMs) from both leading AI startups and Amazon itself, providing a comprehensive range of options to suit various use cases and preferences. This diversity allows enterprises to select the most suitable models for their specific needs, whether they require language generation, image processing, or other AI capabilities.
  • Streamlined Integration: Simplifies the process of adopting and integrating generative AI technologies into existing systems and applications. With its unified API and serverless architecture, enterprises can seamlessly incorporate these advanced AI capabilities without the need for extensive infrastructure management or specialized expertise. This streamlines the development and deployment process, enabling faster time-to-market for AI-powered solutions.
  • Customization Capabilities: Facilitates experimentation and customization, allowing enterprises to fine-tune and adapt the selected models to better align with their unique requirements and data environments. Techniques such as fine-tuning and Retrieval Augmented Generation (RAG) enable enterprises to refine the performance and accuracy of the models, ensuring optimal results for their specific use cases.
  • Security and Compliance Focus: Prioritizes security, privacy, and responsible AI practices, providing enterprises with the confidence that their data and AI deployments are protected and compliant with regulatory standards. By leveraging AWS's robust security infrastructure and compliance measures, enterprises can deploy generative AI applications with peace of mind.

AWS Bedrock Data Privacy & Security Concerns

The rise of AI technologies, while promising transformative and major benefits, also introduces significant security risks. As enterprises increasingly integrate AI into their operations, like with AWS Bedrock, they face challenges related to data privacy, model integrity, and ethical use. AI systems, particularly those involving generative models, can be susceptible to adversarial attacks, unintended data extraction, and unintended biases, which can lead to compromised data security and regulatory violations. 

Training Data Concerns

Training data is the backbone of machine learning and artificial intelligence systems. The quality, diversity, and integrity of this data are critical for building robust models. However, there are significant risks associated with inadvertently using sensitive data in training datasets, as well as the unintended retrieval and leakage of such data. 

These risks can have severe consequences, including breaches of privacy, legal repercussions, and erosion of public trust.

Accidental Usage of Sensitive Data in Training Sets

Inadvertently including sensitive data in training datasets can occur for various reasons, such as insufficient data vetting, poor anonymization practices, or errors in data aggregation. Sensitive data may encompass personally identifiable information (PII), financial records, health information, intellectual property, and more.

 

The consequences of training models on such data are multifaceted:

  • Data Privacy Violations: When models are trained on sensitive data, they might inadvertently learn and reproduce patterns that reveal private information. This can lead to direct privacy breaches if the model outputs or intermediate states expose this data.
  • Regulatory Non-Compliance: Many jurisdictions have stringent regulations regarding the handling and processing of sensitive data, such as GDPR in the EU, HIPAA in the US, and others. Accidental inclusion of sensitive data in training sets can result in non-compliance, leading to heavy fines and legal actions.
  • Bias and Ethical Concerns: Sensitive data, if not properly anonymized or aggregated, can introduce biases into the model. For instance, using demographic data can inadvertently lead to models that discriminate against certain groups.

These risks require strong security measures and responsible AI practices to protect sensitive information and comply with industry standards. AWS Bedrock provides a ready solution to power foundation models and Sentra provides a complementary solution to ensure compliance and integrity of data these models use and output. Let’s explore how this combination and each component delivers its respective capility.

Prompt Response Monitoring With Sentra

Sentra can detect sensitive data leakage in near real-time by scanning and classifying all prompt responses generated by AWS Bedrock, by analyzing them using Sentra’s Data Detection and Response (DDR) security module.

Data exfiltration might occur if AWS Bedrock prompt responses are used to return data outside of an organization - for example using a chatbot interface connected directly to a user facing application.

By analyzing the prompt responses, Sentra can ensure that both sensitive data acquired through fine-tuning models and data retrieved using Retrieval-Augmented Generation (RAG) methods are protected. This protection is effective within minutes of any data exfiltration attempt.

To activate the detection module, there are 3 prerequisites:

  1. The customer should enable AWS Bedrock Model Invocation Logging to an S3 destination(instructions here) in the customer environment.
  2. A new Sentra tenant for the customer should be created/set up.
  3. The customer should install the Sentra copy Lambda using Sentra’s Cloudformation template for its DDR module (documentation provided by Sentra).

Once the prerequisites are fulfilled, Sentra will automatically analyze the prompt responses and will be able to provide real-time security threat alerts based on the defined set of policies configured for the customer at Sentra.

Here is the full flow which describes how Sentra scans the prompts in near real-time:

  1. Sentra’s setup involves using AWS Lambda to handle new files uploaded to the Sentra S3 bucket configured in customer cloud, which logs all responses from AWS Bedrock prompts. When a new file arrives, our Lambda function copies it into Sentra’s prompt response buckets.
  2. Next, another S3 trigger kicks off enrichment of each response with extra details needed for detecting sensitive information.
  3. Our real-time data classification engine then gets to work, sorting the data from the responses into categories like emails, phone numbers, names, addresses, and credit card info. It also identifies the context, such as intellectual property or customer data.
  4. Finally, Sentra uses this classified information to spot any sensitive data. We then generate an alert and notify our customers, also sending the alert to any relevant downstream systems.
Data Flow Customer AWS Cloud Sentra

Sentra can push these alerts downstream into 3rd party systems, such as SIEMs, SOARs, ticketing systems, and messaging systems (Slack, Teams, etc.).

Sentra’s data classification engine provides three methods of classification:

  • Regular expressions
  • List classifiers
  • AI models

Further, Sentra allows the customer to add its own classifiers for their own business-specific needs, apart from the 150+ data classifiers which Sentra provides out of the box.

Sentra’s sensitive data detection also provides control for setting a threshold of the amount of sensitive data exfiltrated through Bedrock over time (similar to a rate limit) to reduce the rate of false positives for non-critical exfiltration events.

Example threat sensitive customer data found in Amazon Bedrock response

Conclusion

There is a pressing push for AI integration and automation to enable businesses to improve agility, meet growing cloud service and application demands, and improve user experiences  - but to do so while simultaneously minimizing risks. Early warning to potential sensitive data leakage or breach is critical to achieving this goal.

Sentra's data security platform can be used in the entire development pipeline to classify, test and verify that models do not leak sensitive information, serving the developers, but also helping them to increase confidence among their buyers. By adopting Sentra, organizations gain the ability to build out automation for business responsiveness and improved experiences, with the confidence knowing their most important asset — their data — will remain secure.

If you want to learn more, request a live demo with our data security experts.

<blogcta-big>

Discover Ron’s expertise, shaped by over 20 years of hands-on tech and leadership experience in cybersecurity, cloud, big data, and machine learning. As a serial entrepreneur and seed investor, Ron has contributed to the success of several startups, including Axonius, Firefly, Guardio, Talon Cyber Security, and Lightricks, after founding a company acquired by Oracle.

Subscribe

Latest Blog Posts

Mark Kiley
Mark Kiley
May 6, 2026
3
Min Read

Data Security for Regulated Industries in the Southeast: How NC, SC, GA, and FL Laws Impact Healthcare, Finance, and Insurance

Data Security for Regulated Industries in the Southeast: How NC, SC, GA, and FL Laws Impact Healthcare, Finance, and Insurance

I spend most of my time talking to security and compliance leaders across North Carolina, South Carolina, Georgia, and Florida. The verticals are familiar: healthcare, financial services, and insurance, exactly the industries regulators care about most, and exactly the ones sitting on some of the messiest data sprawl.

The pattern is almost always the same. Someone leans back and says:

“We’ve got hospitals in NC and FL, a shared services center in SC, a payments hub in Georgia… We’re covered by HIPAA, GLBA, PCI, maybe NYDFS…and now every state’s got its own breach law. How do we build one data security program that actually works across all of this?”

The answer isn’t another policy binder. It’s a data‑centric program that understands how state laws bite per industry and then gives you enough visibility to satisfy them all without freezing your business.

Let me walk through what that looks like for healthcare, finance, and insurance in the Southeast.

1. Healthcare: HIPAA everywhere, state law at the edges

Healthcare is where I see the most “layering” of rules, not just one‑off obligations.

At a federal level, you’ve got HIPAA and HITECH governing PHI. But in our region:

  • North Carolina adds the Identity Theft Protection Act and breach provisions that apply to any “personal information” of NC residents—patient or employee—stored in electronic or non‑electronic form.
  • South Carolina adds § 39‑1‑90, the general breach statute, plus industry‑specific rules for HMOs and health plans in some cases.
  • Georgia uses O.C.G.A. § 10‑1‑912 to cover personal information held by information brokers and others—think combined identity + financial data, credentials, and so on.
  • Florida goes further with FIPA (§ 501.171), which explicitly treats medical information, health insurance IDs, and account credentials as personal information, and forces you onto a 30‑day notification clock for Floridians.

In other words: if you run a health system or health plan across the Southeast, data about one patient can be subject simultaneously to:

  • HIPAA (federal)
  • NC or SC or GA or FL breach laws, depending on residency
  • Sometimes GLBA or state insurance rules if you’re handling plan or financial data as well

The “trick” is not a clever legal memo; it’s knowing, in detail:

  • What data you actually have (PHI, FIPA‑personal information, credentials, financial details, etc.)
  • Where it lives across EHR, billing, analytics, cloud storage, and SaaS
  • Whose data it is—NC vs SC vs GA vs FL residents
  • How it’s protected (encryption, masking, access controls)

That’s the only way to decide, under HIPAA and each state law, whether an incident is a “breach,” which residents are impacted, and which regulators you owe notices to.

2. Financial services: GLBA + PCI + state breach rules

Financial services in the Southeast feel the regulatory squeeze from a different angle.

Most banks, credit unions, and fintechs I work with are already used to GLBA, PCI DSS, and sometimes NYDFS 23 NYCRR 500. They’ve had to build an information security program, monitor vendors, and protect customer information for years.

Then state breach laws layer on top:

  • In North Carolina, if you hold residents’ personal information (name + SSN, account numbers, or other identity data), you’re subject to its Identity Theft Protection Act and must notify affected residents and the AG without unreasonable delay after a qualifying breach.
  • In South Carolina, § 39‑1‑90 also keys off financial account data and government‑issued identifiers, requiring notice to residents, the Department of Consumer Affairs, and credit bureaus in certain volumes.
  • In Georgia, O.C.G.A. § 10‑1‑912 focuses specifically on the kinds of identifiers that enable identity theft and account takeover—perfectly aligned with banking/fintech risk.
  • In Florida, FIPA wraps in financial account data and login credentials and gives you that hard 30‑day deadline plus penalties up to $500,000 for failure to notify.

For a regional bank or fast‑growing fintech headquartered in Atlanta or Charlotte with customers in all four states, a single misconfigured bucket or data lake can light up:

  • PCI (card data)
  • GLBA/FTC (customer information)
  • O.C.G.A. § 10‑1‑912, NC and SC breach laws, and FIPA depending on residency

It’s no accident that Sentra treats financial services and insurance as core regulated ICPs: they have high data sprawl, heavy compliance, and a real need for continuous, provable visibility into PCI and PII across multi‑cloud environments.

3. Insurance: state‑based by design, data‑centric by necessity

Insurance is almost a case study in “fifty states, fifty flavors,” but in the Southeast there’s an especially clear example in South Carolina.

If you’re an insurer or insurance licensee there, you’re dealing with:

  • The South Carolina Insurance Data Security Act (Title 38, Chapter 99), which forces you to implement a written, risk‑based information security program, oversee third‑party service providers, and report certain “cybersecurity events” to the Department of Insurance within ~72 hours of determination.
  • The general SC breach law, § 39‑1‑90, which still governs notice to residents and consumer agencies when “personal identifying information” of SC residents is exposed.

Add to that:

  • NC, GA, and FL breach laws when you hold policyholder data across state lines.
  • Federal overlays like GLBA if you’re handling financial account data, or HIPAA where you’re dealing with health plans.

What I see in practice is that insurance data estates are often more tangled than banking:

  • Core admin systems that have grown through acquisition
  • Claims platforms, document management, and imaging systems stuffed with IDs, medical information, and bank details
  • Data lakes for actuarial modeling and pricing, often with poorly documented ingestion

Under SC’s Insurance Data Security Act, the question is: Do you have “reasonable security” over your nonpublic information, and can you investigate/report a cybersecurity event quickly and accurately?

Under the breach laws (SC, NC, GA, FL), the question is: Can you prove what personal information was at risk, which residents it belongs to, and whether you hit the right notification thresholds and timelines?

You can’t do either if you don’t have a single, trusted view of your data.

The through‑line: regulated data, everywhere

Across all three verticals—healthcare, finance, insurance—the story in the Southeast is the same:

  • Regulators and state AGs are mostly focused on the same core assets: PII, PHI, PCI, credentials, and other data that enable identity theft, fraud, or serious privacy harm.
  • Each state adds its own timing and thresholds, but none of them give you months to figure things out once an incident happens—especially Florida with FIPA’s 30‑day rule.
  • Sector‑specific rules (HIPAA, GLBA, PCI, Insurance Data Security Acts) don’t replace state breach laws; they stack on top of them.

The only way to keep your sanity across all of that is to stop guessing and start operating from real, continuous data intelligence.

That’s exactly where Data Security Posture Management (DSPM) and Sentra come into the picture.

How DSPM helps regulated industries in the Southeast line everything up

Sentra’s DSPM platform is built around the problems that matter most to heavily regulated orgs:

  • Discover & classify regulated data everywhere.
    Sentra continuously discovers and accurately classifies PII, PHI, PCI, credentials, and other regulated data across cloud, SaaS, and on‑prem—building a single inventory your compliance team can trust.

  • Map access and exposure.
    It shows which identities (users, groups, service accounts, AI agents) can reach which sensitive datasets, and whether encryption, masking, and other controls are in place—critical for “reasonable security” and state harm assessments.

  • Align with regulations.
    For regulated industries, Sentra maps regulated data to frameworks like HIPAA, PCI DSS, GLBA, and state privacy/breach laws, with audit‑ready reporting and exportable evidence.

  • Accelerate incident response.
    When an incident hits, Sentra helps you quickly answer:
    • Which data stores were affected?
    • What kinds of sensitive data (PHI, PCI, PII, credentials) were inside?
    • How many NC/SC/GA/FL residents are likely impacted?
    • Was the data truly secured (encryption, keys) or exposed?

That’s what lets you satisfy:

  • HIPAA and FIPA timelines for a Florida hospital
  • GLBA, PCI, and O.C.G.A. § 10‑1‑912 for an Atlanta fintech
  • SC Insurance Data Security Act and § 39‑1‑90 for a Columbia‑based insurer—using one data‑centric system of record instead of a new spreadsheet for every jurisdiction.

If you want a feel for how this looks in a real, high‑stakes environment, the SoFi stories are a good reference point: they’ve talked publicly about using Sentra to build a centralized catalog of sensitive data, improve access governance, and turn cloud‑risk findings into data‑aware decisions.

Different industry, same problem: too much regulated data, not enough visibility, and too many overlapping rules to manage it manually.

Call to action

If you’re running security or compliance for healthcare, financial services, or insurance in the Southeast, you’re already living under NC, SC, GA, and FL laws—whether your playbooks fully reflect that or not.

Let’s take a concrete look at where your regulated data actually lives today, how it lines up with state and sector‑specific rules, and how Sentra’s DSPM can give you a single, trusted view across your Southeast footprint.

Request a Sentra demo

Read More
Mark Kiley
Mark Kiley
May 6, 2026
3
Min Read

Southeast Data Breach Laws Compared: NC, SC, GA, and FL Requirements on One Page

Southeast Data Breach Laws Compared: NC, SC, GA, and FL Requirements on One Page

When I talk to security and privacy leaders who cover the Southeast, the conversation almost always turns into a map.

They’ll say something like: “We’ve got data centers and staff in North Carolina and Georgia, a big insurance book in South Carolina, a hospital or call center in Florida, and our customers don’t see borders. What exactly changes when a breach touches all four states?”

They’re not asking for a law school seminar, they’re asking a simpler question:

What actually matters for my incident response plan when NC, SC, GA, and FL are all in the mix?

This is how I usually walk through it.

Why these four states matter together

A lot of organizations I work with don’t fit neatly into a single state:

  • A health system that owns hospitals in NC and FL, plus clinics just over the border in SC.
  • A fintech headquartered in Atlanta but serving customers across the Carolinas.
  • An insurer with South Carolina licenses and policyholders spread across the region.

They’re all dealing with the same cloud realities—multi‑cloud, SaaS, data lakes, AI tools—but they answer to different Attorneys General, different departments, and slightly different definitions of “personal information” and “breach.”

The patchwork looks messy on paper. The good news is there are more similarities than differences; the challenge is getting enough data visibility to make those similarities work for you.

Let’s go state by state, then pull it together.

North Carolina in practice

North Carolina’s breach framework sits in its Identity Theft Protection Act, particularly N.C. Gen. Stat. § 75‑65 and related provisions. The NC Department of Justice has a very straightforward page for businesses on “Security Breach Information,” and I share that link a lot.

In plain terms:

  • Who’s covered? Any business or public entity that owns, licenses, or maintains “personal information” of North Carolina residents.
  • Personal information? Name + one of: SSN, driver’s license/ID, financial account or card numbers with required codes, or other identifiers that uniquely identify an individual. Encryption and redaction matter — encrypted data is generally out of scope.
  • Breach? Unauthorized access and acquisition of unencrypted/unredacted personal information, when illegal use has occurred, is likely, or creates a material risk of harm.
  • Timing? Notify affected residents “in the most expedient time possible and without unreasonable delay” consistent with law enforcement needs and scoping the breach.
  • Regulator notice? If you notify residents, you also notify the NC Attorney General’s Consumer Protection Division when the breach affects NC residents, plus credit bureaus if you notify more than 1,000 people.

NC also offers a private right of action: residents can sue if they’re injured by a violation.

From a CISO’s perspective, North Carolina is “harm‑aware” and expects you to move quickly once you know what happened and who’s at risk.

South Carolina in practice

South Carolina’s general breach statute is S.C. Code § 39‑1‑90, sitting inside Title 39 (Trade and Commerce). It reads a lot like NC’s but with its own twists.

In plain English:

  • Who’s covered? Any person or entity conducting business in SC that owns or licenses computerized or other data with personal identifying information of SC residents. It also covers entities that only maintain that data for someone else.
  • Personal identifying information? Name + SSN, driver’s license/state ID, financial account or card numbers with required codes/passwords, or other numbers used to access accounts or unique government‑issued identifiers. Publicly available data is excluded.
  • Breach? Unauthorized access to and acquisition of data (not rendered unusable by encryption/redaction) that compromises security, confidentiality, or integrity of PI, when illegal use has occurred, is likely, or creates a material risk of harm.
  • Timing? Same phrase as NC: “most expedient time possible and without unreasonable delay,” consistent with law enforcement and scoping.
  • Regulator notice? If more than 1,000 SC residents are notified, you must also notify the Consumer Protection Division of the Department of Consumer Affairs, and notify nationwide credit bureaus.

Legal summaries from Davis Wright Tremaine, Constangy, and Mintz all flag that South Carolina has both regulatory penalties ($1,000 per affected resident, by DCA) and a private right of action for injured residents.

If you’re in insurance, you also have the South Carolina Insurance Data Security Act on top of this, which I covered in a separate post,  but § 39‑1‑90 is the base layer.

Georgia in practice

Georgia’s rules are built into the Georgia Personal Identity Protection Act, specifically O.C.G.A. § 10‑1‑912. The law is older but still very much alive, and if you work in “Transaction Alley” you’ve almost certainly brushed up against it.

In plain terms:

  • Who’s covered? “Information brokers” and other entities that own or license personal information of Georgia residents, plus some public entities.
  • Personal information? Name + one or more of: SSN, driver’s license/state ID, account/credit/debit card numbers that can be used without extra info, or account passwords/PINs/access codes. Even without the name, those elements can be treated as PI if they’re enough to commit identity theft.
  • Breach? Unauthorized acquisition of an individual’s electronic data that compromises security, confidentiality, or integrity of PI, excluding good‑faith employee access.
  • Timing? Again, “most expedient time possible and without unreasonable delay” after discovery, consistent with scoping and restoring system integrity.
  • Regulator notice? Georgia doesn’t require Attorney General notice in the statute. But if you notify more than 10,000 residents, you must notify all nationwide consumer reporting agencies.

Violations are treated as unlawful practices under Georgia’s Fair Business Practices Act (FBPA), with civil penalties and AG enforcement on the table.

Insureon’s and law review summaries emphasize that Georgia has effectively woven breach duties into its broader consumer protection landscape.

Florida in practice

Florida is the outlier on one very important axis: time.

The Florida Information Protection Act of 2014 (FIPA), living in Fla. Stat. § 501.171, is one of the more aggressive breach notification laws in the U.S.

Here’s how I describe it to Florida teams:

  • Who’s covered? “Covered entities” — any commercial or government entity that acquires, maintains, stores, or uses personal information of Floridians in electronic form.
  • Personal information? Name + any of: SSN; government ID/passport/military ID; financial account/card numbers with required codes; medical history, condition, treatment, or diagnosis; health insurance policy or subscriber number; and username/email plus password or security Q&A for online accounts.
  • Breach? Unauthorized access of data in electronic form containing personal information. Good‑faith access by employees/agents is excluded; encrypted data is excluded if the keys/process weren’t compromised.
  • Timing? Notify affected individuals no later than 30 days after determining a breach occurred, with a possible 15‑day extension if you show good cause to the Attorney General.
  • Regulator and CRA notice? If 500+ residents are affected, notify the Florida Attorney General within 30 days. If 1,000+ are notified, also notify nationwide credit bureaus.

FIPA also:

  • Requires “reasonable measures” to protect and secure personal information in electronic form.
  • Imposes disposal requirements for customer records.
  • Allows civil penalties up to $500,000 per breach for failure to notify in time.

The Florida AG’s guidance and University of Florida’s privacy resources both underline just how broad FIPA is compared to many state laws.

If you operate across all four states, it’s usually FIPA’s 30‑day clock and wider definition of personal information that ends up setting your effective minimum.

The big picture: how the four states line up

When you zoom out, a few patterns emerge that matter more than any single section number.

1. All four states care about largely the same kinds of data.
They all center on data that can be used for identity theft and financial fraud: SSNs, government IDs, account numbers, and access credentials — with Florida adding explicit coverage for health and insurance data and online account logins.

2. All four have encryption/redaction safe harbors.
If data is rendered unusable (typically via strong encryption and sound key management), you’re often outside the breach definition, though you still need to be able to prove that to regulators.

3. NC, SC, and GA use similar “as soon as practicable” timing; FL sets a hard 30‑day line.
North Carolina, South Carolina, and Georgia all talk about notifying “in the most expedient time possible and without unreasonable delay,” giving you a bit more flexibility as long as your scoping work is defensible. Florida is explicit: 30 days, with a very short extension available in special cases.

4. Regulator notification thresholds vary.

  • NC: AG notice when residents are notified; plus CRAs if >1,000 notified.
  • SC: Department of Consumer Affairs and CRAs if >1,000 notified.
  • GA: CRAs if >10,000 residents notified; no AG trigger in the statute.
  • FL: AG if ≥500 residents; CRAs if ≥1,000.

5. NC and SC explicitly include some form of private right of action.
Georgia and Florida handle enforcement more through AG and regulator mechanisms, but Georgia’s FBPA overlay can still expose you to significant civil risk.

For multi‑state CISOs, that usually leads to two practical decisions:

  • Use the strictest timing and definition as your internal baseline — often FIPA plus any sector‑specific rules like HIPAA or GLBA.
  • Invest in data‑centric visibility so you’re not stuck reinventing your data map in every incident.

What this means for multi‑state security teams

Almost every organization I see trying to juggle these four states runs into the same wall: they don’t have a live map of where their sensitive data actually lives and who it belongs to.

So when something does go wrong, they spend critical days or weeks trying to answer:

  • Which databases, buckets, and SaaS tenants were in the blast radius?
  • What types of data were in each — SSNs, medical info, login credentials, insurance IDs, bank details?
  • How many NC/SC/GA/FL residents show up across those stores?
  • Was the data encrypted, masked, tokenized — or just sitting there?

That’s why I keep coming back to Data Security Posture Management (DSPM) in these conversations.

A platform like Sentra continuously:

  • Scans cloud, SaaS, and on‑prem data stores to discover and classify sensitive data — PII, PHI, PCI, credentials, and more.
  • Builds a living inventory of what you have, where it lives, how it’s protected, and who or what can access it.
  • Provides regulation‑aware context, so you can quickly say, “this dataset is in scope for NC/SC/GA/FL breach laws, HIPAA, GLBA, etc.”

When an incident hits, instead of starting with a blank whiteboard, you start with:

  • A list of affected data stores and their contents
  • A breakdown of sensitive data types, including the ones each state’s law focuses on
  • A much faster, more defensible way to estimate how many residents in each state are impacted

The SoFi story is a good parallel even though it’s not Southeast‑specific. In their webinar and blog with Sentra, SoFi’s team explains how they used DSPM to build a centralized, accurate catalog of sensitive data across a sprawling cloud estate, map it to compliance requirements, and improve data access governance — all without slowing engineering down.

That same pattern is exactly what Southeast organizations need to live with NC, SC, GA, and FL laws at once.

If you’re responsible for data security across North Carolina, South Carolina, Georgia, and Florida, and you’re not sure how your current visibility would hold up under a multi‑state breach, now is the time to find out, not when four clocks are already running.

See how Sentra can give you a single, continuously updated view of sensitive data across your Southeast footprint, so you can meet each state’s breach requirements with facts instead of guesswork.

Request a Sentra demo

Read More
Mark Kiley
Mark Kiley
May 6, 2026
3
Min Read

FIPA vs HIPAA: Florida Healthcare Data Breach Obligations Compared (with Real‑World Patterns)

FIPA vs HIPAA: Florida Healthcare Data Breach Obligations Compared (with Real‑World Patterns)

When I sit down with CISOs and privacy officers in Florida hospitals and health systems, the same question comes up again and again, usually right after we finish walking through an incident tabletop:

“Okay, but after a breach, who do we really answer to first? HIPAA or FIPA?”

You can feel the tension under that question. On one side, the HIPAA Breach Notification Rule with its 60‑day outside limit. On the other, Florida’s Information Protection Act (FIPA) with a 30‑day requirement that feels like a sprint from day one.

The short version, something I repeat a lot, is:

In Florida healthcare, you don’t get to choose. You have to satisfy both HIPAA and FIPA. The only way that feels sane is if you truly understand where your data lives, what kind of data it is, and who it belongs to before anything goes wrong.

Let me unpack that.

Two overlapping worlds: HIPAA and FIPA

First, a quick refresher on what each law is trying to do.

HIPAA’s Breach Notification Rule

HIPAA is a federal law. For healthcare entities, the Breach Notification Rule says that when you have a breach of unsecured PHI (protected health information), you must notify:

  • Affected individuals
  • The U.S. Department of Health and Human Services (HHS), and
  • Sometimes the media (if >500 individuals in a state or jurisdiction are affected)

without unreasonable delay and no later than 60 days after discovering the breach, unless an exception applies.

The rule expects you to perform a risk assessment: look at what PHI was involved, who accessed it, whether it was actually viewed or acquired, and how much risk there is that the information has been compromised. If the probability of compromise is low, it might not be a reportable HIPAA breach; if it’s not low, it is.

The University of Florida’s privacy office has a nice summary of how HIPAA’s Privacy Rule interacts with state law—they point out that where state law is more protective, it can effectively sit “on top of” HIPAA. That’s exactly what FIPA does in Florida.

FIPA: Florida’s Information Protection Act

FIPA, codified at Fla. Stat. § 501.171, is a state law that doesn’t just apply to healthcare—it applies broadly to businesses and government entities handling Floridians’ personal information.

A few key points that matter for hospitals and plans:

  • It defines “personal information” more broadly than just PHI: medical data, health insurance identifiers, financial data, and even login credentials (username + password or security Q&A) for online accounts are all in scope.
  • It requires notice to affected Florida residents within 30 days of determining a breach occurred, with a narrow 15‑day extension if the Attorney General agrees you have good cause.
  • If 500 or more Florida residents are affected, you also have to notify the Florida Attorney General’s Office within that same 30‑day window.
  • If 1,000+ are affected, you must notify credit reporting agencies as well.

Florida’s own Attorney General and university guidance spell out just how wide this net is: FIPA is about data security and rapid transparency when Floridians’ personal information—not just PHI—has been exposed.

Where HIPAA and FIPA overlap—and where they don’t

In most of the scenarios I see in Florida healthcare, HIPAA and FIPA are not competing—they’re stacked.

Here’s how that usually looks in practice.

Same incident, two definitions

Say you have an intrusion into a cloud backup that holds:

  • Clinical notes and lab results (PHI)
  • Insurance subscriber IDs and plan information
  • Patient portal usernames and hashed passwords
  • Billing data with partial account numbers

From HIPAA’s point of view, you’re asking:

  • Was unsecured PHI involved?
  • Did unauthorized individuals access, use, or acquire it?
  • Does the risk assessment show a low probability of compromise or not?

From FIPA’s point of view, you’re asking:

  • Did unauthorized access of data in electronic form containing “personal information” occur?
  • Does that personal information match FIPA’s definitions—medical history, health condition, diagnosis, health insurance IDs, financial data, credentials?
  • Was it unsecured (unencrypted or otherwise usable), and is there a realistic risk of harm?

Most of the time, the answer is “yes” on both sides. You’ve got PHI, and you’ve got FIPA‑personal information sitting right next to it.

Two clocks, one reality

If you accept that both laws apply, you’re now staring at:

  • HIPAA’s 60‑day maximum, and
  • FIPA’s 30‑day maximum for Florida residents and potentially the Attorney General.

In conversations, I try to be blunt about this: you don’t get to “pick” the friendlier timeline. The conservative, and frankly safest, approach is to treat the stricter FIPA 30‑day clock as your governing SLA for Florida residents, and then layer HIPAA and HHS reporting on top.

The University of Florida’s guidance on HIPAA vs state law makes the same point in more formal language: where state law is more protective, that’s the bar you have to hit.

Real‑world patterns I see in Florida healthcare

I won’t name organizations, but I can share the kinds of incidents and questions I see over and over.

1. The “multi‑system PHI + PII” breach

A compromised account or misconfigured service touches more than just the EHR. It hits:

  • The EHR or clinical data warehouse
  • The revenue cycle system with bank and card info
  • A file share holding scanned IDs and insurance cards
  • An S3 bucket or Azure Blob used for data science

Suddenly, the incident isn’t “just a HIPAA issue.” It’s HIPAA + FIPA + maybe PCI + maybe GLBA. Teams realize they don’t have an accurate, current inventory of what’s actually stored in each of those places, or how many Florida residents show up in each dataset.

2. Portal and credential‑driven incidents

FIPA’s inclusion of usernames and email addresses with passwords or security Q&A as personal information is a big deal for patient portals and mobile apps.

When I walk through credential stuffing or phishing scenarios with Florida teams, the question isn’t just, “Did PHI get accessed?” It’s also, “Did we expose enough to let someone log in as this person and see their PHI or transact in their name?”

From FIPA’s perspective, a stash of valid portal credentials is personal information, even before a single clinical note is viewed.

3. The “is this a breach under one but not the other?” corner case

Occasionally, we run into situations where the HIPAA risk assessment suggests a low probability of compromise (for example, strong encryption and good evidence no data left the environment), but the team is still queasy about Florida’s expectations under FIPA.

In those moments, I’ve seen the best outcomes when organizations lean on data‑driven evidence: encryption posture, key management details, access logs, and a clear map of what data was in the blast radius. That’s what convinces AGs and regulators, not vague assurances.

Why a data‑centric view matters more than ever

The common thread in all of this: you can’t make good HIPAA or FIPA decisions if you don’t really know your data.

Over and over, I see the same pain points:

  • PHI and FIPA‑personal information spread across EHR, billing, imaging, analytics platforms, M365, Google Workspace, and niche SaaS apps.
  • Multiple copies of the same sensitive datasets in test and dev, created in a hurry and then forgotten.
  • No single, up‑to‑date view of which systems contain medical info, insurance IDs, financial data, and credentials for Florida residents.

That’s why I keep steering the conversation toward data‑centric security and Data Security Posture Management (DSPM) instead of just more perimeter tools.

A DSPM platform like Sentra continuously:

  • Discovers and classifies sensitive data across cloud, SaaS, and on‑prem, including PHI, FIPA‑personal information, PCI, and other regulated data.
  • Builds a live inventory of where that data lives and how it’s protected (encryption, masking, labels, retention).
  • Shows who and what can access it—doctors, nurses, back‑office staff, vendors, AI assistants, service accounts.

So when you’re faced with a potential breach, you’re not scrambling to reconstruct all of that from scratch. You already know:

  • Which systems in the incident path actually hold PHI and FIPA‑personal information
  • How many Florida residents are likely involved
  • Whether the data was truly secured or not

Sentra customers in healthcare, like Valenz Health, have used this approach to scale PHI protection post‑merger, as highlighted in Sentra’s case studies and industry pages. The specifics of their story are different from yours, but the underlying move is the same: get out of the spreadsheet business and into continuous, factual visibility.

How I suggest Florida healthcare teams think about HIPAA + FIPA

When we build joint playbooks with Florida customers, the conversation usually ends up here:

  • Treat HIPAA and FIPA as a combined requirement, not two separate worlds.
  • Use DSPM to create a single, accurate view of PHI + FIPA‑personal information across all your environments.
  • Let that data intelligence drive both your breach risk assessments and your notification decisions.
  • Anchor your timelines to the stricter FIPA 30‑day deadline for Florida residents, and then layer HIPAA/HHS obligations on top.

Once you do that, the question, “HIPAA or FIPA first?” stops being so theoretical. You’ve got the evidence to satisfy both.

Call to action

If you’re in Florida healthcare and you’re not sure how you’d really perform under a combined HIPAA + FIPA breach scenario, now’s the time to find out—before the clock starts.

Let’s take a look at where your PHI and FIPA‑personal information really live today, and how Sentra’s DSPM can help you move from guesswork to defensible, data‑driven decisions.

Request a Sentra demo

Read More
Expert Data Security Insights Straight to Your Inbox
What Should I Do Now:
1

Get the latest GigaOm DSPM Radar report - see why Sentra was named a Leader and Fast Mover in data security. Download now and stay ahead on securing sensitive data.

2

Sign up for a demo and learn how Sentra’s data security platform can uncover hidden risks, simplify compliance, and safeguard your sensitive data.

3

Follow us on LinkedIn, X (Twitter), and YouTube for actionable expert insights on how to strengthen your data security, build a successful DSPM program, and more!

Before you go...

Get the Gartner Customers' Choice for DSPM Report

Read why 98% of users recommend Sentra.

White Gartner Peer Insights Customers' Choice 2025 badge with laurel leaves inside a speech bubble.