A data security incident: when to report to the Authority, when to notify those affected, and what changes when the GDPR applies
A system breach, an email sent to the wrong recipient or a lost laptop are all security incidents. The Israeli Data Security Regulations set out which incidents must be recorded, which must be reported to the Privacy Protection Authority, and when the Authority may order notice to those affected. A business also subject to the GDPR works to a 72-hour timetable.
What is a security incident
Under Regulation 11 of the Privacy Protection (Data Security) Regulations, 5777-2017, a security incident is any case in which an event is discovered that raises a concern of harm to the integrity of the data, of its use without authorisation, or of use beyond authorisation. The definition is broad: it covers a concern, not only proven harm, and it applies to internal events too, such as an employee viewing data not needed for their work.
Records and procedure
The database owner is responsible for recording every security incident, as far as possible by automatic logging. The database’s security procedure must include instructions for handling incidents according to their severity and the sensitivity of the data, including revoking access rights, other immediate steps, and reporting to the database owner on the incident and the actions taken. For a database at the medium security level, the owner reviews security incidents at least once a year and considers whether to update the security procedure; at the high level the review takes place at least once a quarter. The records are kept securely for 24 months.
Severe incidents and reporting to the Authority
The duty to report to the Authority applies to a “severe security incident”, and its definition depends on the database’s security level:
| Security level | Severe security incident |
|---|---|
| High | Any incident in which data from the database was used without authorisation or beyond authorisation, or the integrity of the data was harmed |
| Medium | Such an incident involving a substantial part of the database |
| Basic, or a database managed by an individual | No definition, and no duty to report to the Authority under the regulation |
When a severe security incident occurs, the database owner must notify the Privacy Protection Authority immediately and also report on the steps taken in response. The regulation does not set a number of hours; “immediately” means not waiting for the investigation to finish. In a high-level database, unauthorised use of the data of a single person is enough for the incident to be severe. The database’s security level, set in the database definitions document, therefore directly affects the reporting duty.
Amendment 13 to the Privacy Protection Law empowered the Authority to impose administrative fines for breaches of the Data Security Regulations, according to the database’s security level, and breach of the reporting duty is one of them.
Notice to data subjects
Israeli law contains no general duty to notify those affected. Under Regulation 11, the Authority may, after consulting the head of the National Cyber Security Authority, order the database owner to notify a data subject who may be harmed by the incident. The power does not apply to the bodies listed in Section 13(e) of the Law, including the security bodies. A business may also notify those affected on its own initiative, and there is sometimes good reason to do so, for example where login credentials that can be changed were exposed.
When the incident occurs at a provider
Many incidents occur at a provider: a cloud company, a payroll system or a software vendor. The duty to report to the Authority remains with the database owner. Under Regulation 15, the agreement with a provider that has access to the database includes its duty to notify the owner of a security incident; see A provider with access to your database. An agreement that sets no timetable for notice, or lets the provider notify only after completing its own review, may prevent the owner from reporting in time.
Reporting under the GDPR
Under Article 33 of the GDPR, a controller notifies the supervisory authority of a personal data breach without undue delay and, where feasible, within 72 hours of becoming aware of it, unless the breach is unlikely to result in a risk to the rights and freedoms of individuals. A notification made after 72 hours is accompanied by reasons for the delay. A processor notifies the controller without undue delay. Every breach is documented, even if not reported.
Under Article 34, where the breach is likely to result in a high risk to individuals, the controller also communicates it to them without undue delay. This is not required where the data was protected by measures rendering it unintelligible, such as encryption; where subsequent measures ensure the high risk is no longer likely to materialise; or where individual communication would involve disproportionate effort, in which case a public communication is made instead.
An Israeli company with no establishment in the Union has no single “lead authority”. Under the guidelines of the European Data Protection Board (EDPB), it notifies every supervisory authority in a member state where affected data subjects are located. A single incident may require reporting to the Israeli Authority and to several European authorities at the same time, on different timetables.
| Topic | Israel | GDPR |
|---|---|---|
| When to report to the authority | A severe security incident, in a medium or high level database | Every breach, unless no risk is likely |
| Timing | Immediately | Without undue delay, and where feasible within 72 hours |
| Notice to those affected | On the Authority’s order | Required where there is a high risk, subject to exceptions |
| Records | Every incident; kept for 24 months | Every breach |
Fines
Breaches of the Data Security Regulations carry administrative fines under the Third Schedule to the Law, set by the database’s security level. Failing to report a severe security incident to the Authority is one of the doubled items: NIS 80,000 for a medium-level database, NIS 320,000 for a high-level database, and NIS 640,000 for a high-level database holding data on one million people or more. Failing to document an incident exposes the controller to NIS 40,000 (medium) or NIS 160,000 (high).
On 21 July 2026 the Privacy Protection Authority announced the first fine imposed under Amendment 13: NIS 256,000 on a health maintenance organisation, for failing to report immediately, as Regulation 11(d) requires, a fault that allowed unauthorised access to medical information.
