Healthcare Cybersecurity: Regulators Want Proof, Not Promises

Every hospital board today wants to talk about AI. Regulators want to talk about something far more basic: are the systems that hold patient data securely configured, and can you show the evidenc
A recent article from the Center for Internet Security (CIS) looks at how US healthcare regulators are enforcing this. The lessons travel well. Thai hospitals, clinics, labs and health-tech firms face the same two questions under our own laws.
Enforcement is about what was missing before the attack
In the United States, the HHS Office for Civil Rights (OCR) enforces HIPAA. According to CIS, OCR settled four ransomware cases in April 2026 for a combined total above US$1.1 million. Every one of them traced back to the same root cause: no proper risk analysis.
A few months later, OCR fined the employee health plan of a national retailer US$450,000 for the same failure. It was not a hospital, and it was not a giant breach. The message is clear. Regulators judge what you had in place before the incident, not only how you handled it.
The US is also waiting on an updated HIPAA Security Rule, now expected in July 2027. CIS points out that the current rule remains fully enforceable in the meantime. "Waiting for the new rules" is not a defence.Slow user experience. Inspecting encrypted traffic on underpowered devices can slow everyone down.
Why this matters in Thailand
Thai healthcare organisations sit under two laws that ask for the same kind of proof:
Personal Data Protection Act B.E. 2562 (PDPA). Health data is sensitive personal data. Controllers must have appropriate security measures and report qualifying breaches to the PDPC within 72 hours. Administrative fines for sensitive-data violations can reach THB 5 million, on top of civil and criminal exposure.
Cybersecurity Act B.E. 2562. Public health is a designated critical information infrastructure (CII) sector. CII organisations must follow codes of practice, run risk assessments and audits, and report incidents to the NCSA.
In both cases, a policy document alone will not satisfy an investigator. They will ask what you actually configured, when, and how you know it is still that way.
Hardening is continuous work, and AI adds to it
OCR's January 2026 cybersecurity newsletter, cited by CIS, encouraged healthcare organisations to adopt standard security baselines. It described hardening as ongoing work, not a one-off project.
This matters more as AI arrives. Every new AI tool brings new servers, cloud workloads and connections to patient data. Each one needs the same secure baseline as the systems you already run.
Breaches still hit healthcare hardest
CIS cites IBM's 2026 Cost of a Data Breach Report. Healthcare breaches averaged US$6.64 million, making it the most expensive industry for the 13th year running. The global average reached a record US$4.99 million.
Time matters too. Breaches that ran longer than 200 days averaged US$5.65 million, against US$4.32 million for those contained sooner. A hardened, well-documented baseline reduces the openings an attacker can use and makes anomalies easier to spot.
Four gaps we see again and again
CIS identifies four recurring gaps. They match what Greenwill sees in Thai organisations, in healthcare and beyond:
Findings without follow-through. The risk assessment is done, the report is filed, and nothing changes. Regulators increasingly ask what you did about the findings.
No common baseline. Each server, cloud workload or medical-device PC is configured by whoever built it, so standards differ across the estate.
Configuration drift. A system that was hardened on day one slowly drifts through patches, vendor changes and quick fixes.
ConfiguraScattered evidence. Scan results sit with security, policies with compliance, change records with IT. When an auditor asks, someone rebuilds the story by hand.tion drift. A system that was hardened on day one slowly drifts through patches, vendor changes and quick fixes.
The rot cause is the same: no repeatable way to define "secure", apply it consistently, measure it over time, and show the results
A practical way forward with free CIS resources
You do not need a large budget to start. CIS provides:
CIS Benchmarks – free, consensus-based secure configuration guides for more than 100 technologies, including Windows, Linux, databases and cloud platforms. Each recommendation explains why it matters and how to apply it.
CIS Critical Security Controls – a prioritised set of safeguards. The Benchmarks map to the Controls, so you can turn technical findings into a management-level remediation plan.
Automated assessment tools – such as CIS-CAT, which compare a system's settings with its Benchmark and report conformance. Results become dated evidence you can hand to an auditor.
Start this week: the one-system test
Pick one system that holds patient data, such as the HIS database server or a PACS workstation. Ask two questions:
How is it configured, compared with a recognised baseline?
Where is the evidence, and how old is it?
If the answer takes more than a day to find, you have found your starting point.
Start this week: the one-system test
Greenwill Solution helps Thai organisations turn security policies into evidence. We assess systems against CIS Benchmarks, fix the gaps, monitor for drift, and keep audit-ready records that support PDPA and Cybersecurity Act obligations. Contact our team for a no-obligation conversation about your most critical system.
Inspired by Healthcare Cybersecurity: What Regulators Are Looking For Right Now, Center for Internet Security. This article is for awareness only and is not legal advice.



Comments