Public Key Infrastructure (PKI) is the set of hardware, software, policies, and Certificate Authorities that issue and manage digital certificates, using public and private key pairs to authenticate identities and encrypt communication instead of relying on passwords.
PKI works by issuing X.509 digital certificates that bind a public key to a verified identity, whether a website, device, or user. A Certificate Authority signs each certificate, and browsers or applications validate the certificate’s signature chain back to a trusted root before establishing an encrypted connection using the public key. This replaces passwords with cryptographic proof of identity.
Key Takeaways
- PKI combines asymmetric and symmetric encryption. Asymmetric encryption authenticates the connection and exchanges a session key; symmetric encryption then handles the actual data transfer efficiently.
- X.509 is the certificate format PKI relies on. It standardizes how identity, public key, validity period, and issuer information are encoded across TLS, digital signatures, and CRLs.
- Two-tier CA architecture is the recommended default. An offline root CA plus one or more online issuing CAs balances security, scalability, and manageability for most organizations.
- CRLs, Delta CRLs, and OCSP all check revocation status. CRLs list every revoked certificate, Delta CRLs list only recent changes, and OCSP checks a single certificate’s status in real time.
- PKI extends well beyond website encryption. Code signing, email signing (S/MIME), Wi-Fi authentication, VPN access, and IoT device identity all rely on the same certificate infrastructure.
How PKI Encryption Works
PKI relies on asymmetric encryption for identity and key exchange, then hands off to faster symmetric encryption for the actual data.
Asymmetric encryption uses a mathematically linked public and private key pair: data encrypted with the public key can only be decrypted with the matching private key. This solves the key distribution problem symmetric encryption faces, since the public key can be shared openly. Algorithms like RSA and ECDSA, with key sizes from 2048 bits upward, provide this asymmetric encryption; longer keys are exponentially harder to break but slower to compute.
Because asymmetric encryption is comparatively slow, PKI uses it only to authenticate the connection and exchange a symmetric session key. That session key then handles the bulk data transfer using faster symmetric encryption, which is exactly how a TLS handshake and protocols like S/MIME combine both encryption types for security and speed.
What Digital Certificates Contain
A digital certificate is the credential PKI issues, binding a public key to a verified identity under the X.509 standard.
Certificates can be self-signed for internal use between trusted parties, or signed by a Certificate Authority for public use, such as securing a website. A CA verifies the requester’s identity, for example confirming domain ownership, before signing. The X.509 format standardizes this structure across TLS/SSL, digital signatures, and Certificate Revocation Lists, so any compliant system can parse a certificate the same way.
Two-Tier vs Three-Tier CA Architecture
Most organizations choose between two architectures for structuring their Certificate Authority hierarchy.
| Two-tier | Offline root CA + online issuing CA | Most organizations; the recommended default |
| Three-tier | Offline root + offline policy/intermediate CA + online issuing CA | Organizations needing formal administrative or policy boundaries between issuing CAs |
Three-tier architecture adds security and flexibility but also cost and management overhead; without a real administrative or policy need, the middle tier often goes unused, which is why two-tier remains the standard recommendation for most PKI deployments.
How Certificate Revocation Checking Works
Three mechanisms let a relying party confirm a certificate has not been revoked before its natural expiration.
- CRL (Certificate Revocation List). A CA-signed list of every certificate it has revoked; browsers check the site’s certificate against this list.
- Delta CRL. Lists only the certificates revoked since the last full CRL publication, making updates faster to distribute.
- ARL (Authority Revocation List). A CRL variant listing revoked Certificate Authority certificates rather than end-user certificates.
- OCSP. Defined in RFC 6960, OCSP checks a single certificate’s status in real time rather than downloading a full list, at the cost of returning less detail than a CRL.
Where PKI Is Used Beyond Websites
PKI secures far more than HTTPS connections across modern IT environments.
- Digital signatures and code signing. PKI verifies that software or a document has not been tampered with and confirms who signed it.
- Certificate-based authentication. Smart cards, Wi-Fi authentication (802.1X), and VPN access all use certificates in place of passwords.
- Secure email (S/MIME). Both sender and recipient use CA-signed certificates to encrypt and authenticate email content.
- IoT device identity. PKI assigns unique digital certificates to smart devices, letting manufacturers track, update, and secure them at scale.
How Encryption Consulting Helps
Encryption Consulting’s PKI Advisory Services cover assessment, design and implementation, Certificate Policy and Certificate Practice Statement development, and PKI-as-a-Service for organizations that want a customizable, high-assurance PKI without managing the infrastructure themselves. Backed by ISO/IEC 27001:2022 and SOC 2 certified practices.
Frequently Asked Questions
What is the difference between PKI and encryption?
PKI is the broader framework of certificates, keys, and Certificate Authorities that manages digital identity and trust. Encryption, specifically asymmetric encryption, is one of the cryptographic tools PKI relies on to authenticate identities and exchange keys securely.
Why does PKI use both symmetric and asymmetric encryption?
Asymmetric encryption is more secure for authentication and key exchange but too slow for bulk data. PKI uses asymmetric encryption to authenticate the connection and exchange a session key, then switches to faster symmetric encryption for the actual data transfer, combining the security benefits of both.
What is the difference between a Root CA and a Subordinate CA?
A Root CA sits at the top of the trust hierarchy, is typically kept offline and air-gapped, and signs certificates for subordinate CAs. A Subordinate (or Intermediate) CA operates online, issues certificates to end users and devices, and can be recreated if compromised without affecting the root; a compromised root, by contrast, requires rebuilding the entire trust chain.
Does an Extended Validation certificate look different in the browser?
Not anymore. Extended Validation certificates involve the most thorough identity verification of the three validation levels, but major browsers removed the distinct visual treatment, such as displaying the company name in the address bar, in 2019. EV’s value today is the depth of verification, not a visible browser indicator.
What is OCSP and how is it different from a CRL?
OCSP, defined in RFC 6960, checks a single certificate’s revocation status in real time by querying a responder service, which is faster than downloading a full Certificate Revocation List. The tradeoff is that OCSP returns less detail (essentially good, revoked, or unknown) and has less robust security guarantees than CRL-based checking.
Build PKI That Scales With Your Organization
Explore Encryption Consulting’s PKI Advisory Services for assessment, design, and implementation support, or see PKI-as-a-Service for a managed alternative.