Data Leak
A data leak is when sensitive information is accidentally exposed to people who shouldn't see it, for example through a misconfigured cloud folder, a public link or a misdirected email, often without any attacker involved at first.
Also called: data exposure, information leak, accidental data exposure, leaked data
What is a data leak?
A data leak is information escaping by accident. No one had to break in: a door was left open. Common examples include a cloud storage bucket set to public, a file-sharing link that anyone can open, a spreadsheet of customer details attached to the wrong email, or code with passwords pushed to a public repository.
Data leak vs. data breach
The terms overlap. A leak describes how data was exposed, usually through a mistake. It becomes a data breach when unauthorized people access the data, which can trigger notification duties. MITRE ATT&CK documents how attackers actively search for improperly secured cloud storage, so leaked data is often found and taken quickly.
Common causes
- Public or "anyone with the link" sharing in OneDrive, Google Drive, Dropbox or S3.
- Databases and backups exposed to the internet without passwords.
- Misdirected emails and auto-complete mistakes.
- Secrets and data in public code repositories.
- Overly detailed error messages and debug pages on websites.
How to prevent data leaks
- Restrict default sharing settings in your file-sharing and cloud platforms, and review external shares regularly.
- Use data loss prevention rules and sensitivity labels to warn or block when sensitive data is shared.
- Scan for exposed cloud storage and databases, and fix misconfigurations.
- Turn on email recipient warnings for external addresses.
- If a leak is found, remove access immediately, check logs for who accessed the data, and follow the FTC's breach response steps if needed.
Related terms
Sources
- Data Breach Response: A Guide for Business (Federal Trade Commission)
- Data from Cloud Storage (T1530) (MITRE ATT&CK)
Drafted with AI assistance, then edited and fact-checked by a person. How we use AI.
