Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

What Is End-to-End Encryption?

End-to-end encryption keeps the decryption keys on the participants’ devices, so the service carrying the messages can route them but cannot read them. It is a trust model, not just an algorithm.

Illustration for the cybersecurity term End-to-End Encryption
Home / Blog / Cybersecurity Terms / What Is End-to-End Encryption?
17 November 2026

What Is End-to-End Encryption?

What Is End-to-End Encryption?

Most encrypted services decrypt your data when it arrives. A message sent over TLS is protected on the wire, decrypted at the provider, stored, re-encrypted, and delivered — which means the provider can read it, and so can anyone who compromises the provider or serves it with a legal demand. End-to-end encryption removes that middle step. Keys exist only on the participants’ devices, the server handles ciphertext it cannot open, and the operator’s honesty stops being part of your threat model.

The distinction is about who holds keys, not about which cipher is used. A service can describe itself as encrypted in every marketing sentence and still hold the keys to everything.

How the Signal Protocol Works

Nearly every modern deployment follows the design Open Whisper Systems published and WhatsApp adopted across its user base in April 2016. Three parts matter:

The ratchet delivers two properties worth naming. Forward secrecy means a key compromised today does not decrypt yesterday’s messages. Post-compromise security, sometimes called future secrecy, means that if an attacker steals the current state but then loses access, the conversation heals itself after the next exchange. Few other protocols offer the second.

What It Does Not Protect

End-to-end encryption protects content in transit between endpoints, and that is all.

  1. Metadata. Who contacted whom, when, how often, from which IP address, and message sizes are typically visible to the operator. For many intelligence purposes metadata is more useful than content.
  2. Endpoints. Malware on a phone reads messages after decryption. Commercial spyware such as Pegasus targets the device for exactly this reason, and no protocol can defend against a compromised endpoint.
  3. Backups. A chat history uploaded to a cloud backup in plaintext undoes the whole design. WhatsApp and iCloud both offer encrypted backups, and both are optional.
  4. Other participants. Anyone in the conversation can screenshot, forward, or report the content.
  5. Key distribution. If the server can quietly add a device or swap a public key, it can insert itself. This is why key verification through safety numbers or QR codes exists, and why change notifications should be enabled.

Where You Will Meet It

Signal, WhatsApp, and Matrix use it for messaging. Apple applies it to iMessage, FaceTime, and, with Advanced Data Protection enabled, to most iCloud categories. Zoom added it for meetings in 2020 after criticism of earlier claims. Password managers, Proton Mail, and Tresorit apply the model to stored data, usually by deriving keys from a passphrase the provider never sees. In enterprise settings it appears in file sharing, secure email gateways, and some payment flows.

Deploying it in a business is a governance decision as much as a technical one. Content that nobody can decrypt cannot be e-disclosed in litigation, archived for FCA record-keeping obligations, or scanned for data loss prevention. Those are solvable problems, usually by keeping regulated communication on channels designed for it, but they must be solved deliberately rather than discovered during an investigation.

The UK Regulatory Position

The Online Safety Act 2023 gives Ofcom power under section 121 to require a service to use accredited technology to identify child sexual abuse or terrorism content, including on private messaging. Providers including Signal and WhatsApp stated they would withdraw from the UK rather than weaken end-to-end encryption; the government said during the Lords stages in September 2023 that the power would not be used until scanning was technically feasible, which leaves the clause on the statute book and unexercised.

Separately, in February 2025 Apple withdrew Advanced Data Protection for new UK users after reports of a Technical Capability Notice served under the Investigatory Powers Act 2016, meaning UK iCloud accounts lost the option of end-to-end encrypted cloud backups. Whatever your view of the policy, the practical lesson for an organisation is that the availability of these features can change by jurisdiction, with little notice, and your data protection assessments should not assume otherwise.

Where messaging, collaboration, or file transfer carries sensitive data, we assess the real key custody arrangements rather than the marketing claims as part of our infrastructure and architecture review services.

FAQ – Frequently Asked Questions

Is HTTPS End-to-End Encryption?

No. HTTPS protects data between your browser and the server, and the server decrypts it on arrival. The operator, and anyone with access to their systems, can read the content. End-to-end encryption means only the communicating users hold keys, and the service in the middle sees ciphertext it cannot open.

Can My Employer Read End-to-End Encrypted Messages?

Not from the server, but often from the device. Managed mobile devices, endpoint agents, and screen capture tools can see messages after decryption, and company-owned devices may be configured to do so lawfully. If a conversation must stay private from an employer, the device matters more than the protocol.

Does End-to-End Encryption Satisfy UK GDPR?

It is strong evidence of an appropriate technical measure under Article 32, and it can reduce the risk rating of a breach if ciphertext is exposed without keys. It does not remove other obligations: metadata is still personal data, you remain responsible for lawful basis, retention, and subject access, and you must still be able to meet those duties.

Why Do Apps Show a Safety Number or QR Code?

Because end-to-end encryption only works if you have the right public key for the other person. The safety number is a fingerprint of both identity keys; comparing it in person or over a separate channel confirms nobody has been inserted in the middle. Enable change alerts, and treat an unexpected change as worth checking.

Related Terms

You May Be Interested In These

What Is Sliver?

Cybersecurity Terms

17 November 2026

What Is Sliver?

Read More
What Is the OSCP Certification?

Cybersecurity Terms

17 November 2026

What Is the OSCP Certification?

Read More
What Is a Public S3 Bucket?

Cybersecurity Terms

17 November 2026

What Is a Public S3 Bucket?

Read More