By · Last updated 2026-05-29

Back to BlogGDPR & Compliance

CCPA/CPRA 2025: California AI Privacy

CPPA issued $100M+ in fines in 2024. CPRA covers 40M Californians and applies globally to most businesses. 19 sensitive data categories, automated.

May 29, 202610 minute read
CCPA CPRACalifornia privacy lawCPPA enforcementUS state privacyAI compliance

CCPA/CPRA: California Privacy Law Guide

Updated for 2026.

California's Consumer Privacy Rights Act (CPRA) took effect in 2023. It set up the California Privacy Protection Agency (CPPA). The CPPA is the first state privacy regulator in the US. In 2024, the CPPA issued over $100 million in fines. This is active enforcement — not a paper rule.

Who Must Comply

Three tests determine if the law applies. A business must meet any one of them.

  • Annual revenue of $25 million or more.
  • Personal data from 100,000 or more state consumers.
  • More than 50% of revenue from selling personal information.

The state has 40 million residents. It is the fifth largest economy in the world. Most global companies meet at least one test.

19 Categories of Sensitive Data

The law creates a special tier for sensitive personal details. Companies must give consumers extra notice and a right to limit use. The 19 types are:

  1. Social Security, driver's license, state ID, passport numbers
  2. Financial account or card numbers with access codes
  3. Precise geolocation (within 1,852 meters)
  4. Racial or ethnic origin
  5. Religious or philosophical beliefs
  6. Union membership
  7. Private email or postal mail content
  8. Genetic information
  9. Biometric records for identification
  10. Health or medical records
  11. Sexual orientation or sex life
  12. Immigration status (added in 2024)
  13. Mental health records (added in 2024)
  14. Citizenship status
  15. Financial account numbers without access codes
  16. Disability status
  17. Employment status indicators
  18. Insurance policy information
  19. Criminal or arrest records

For each category, consumers can limit how a business uses and shares it. This right is separate from the general right to opt out of sale.

What the CPPA Targeted in 2024

The CPPA's 2024 actions fell into four areas.

Data broker registration. The state requires data brokers to register with the CPPA. The agency found hundreds of unregistered brokers selling consumer profiles.

Broken opt-out tools. Many consent platforms did not give a real opt-out. The button did not work, or the opt-out only covered some uses.

AI decisions without notice. The 2025 AI rules require notice when automated tools make key decisions. Jobs, credit, and housing all count. Several 2024 cases targeted AI tools used without this notice.

Children's records. California's Age-Appropriate Design Code applies to any service minors are likely to use. Those companies must complete a Data Protection Impact Assessment. The CPPA found many companies had not done this.

State Law vs. GDPR: Key Differences

GDPR compliance does not cover you in California. The laws share goals but differ on key points.

Opt-out versus opt-in. GDPR requires opt-in consent for most sensitive-data uses. The state law uses an opt-out model. Processing is allowed until the consumer objects.

Consumer rights. Both laws give access, deletion, and correction rights. State law adds a right to opt out of automated decisions. GDPR Article 22 covers this too, but with a narrower scope.

Employee records. State law covers employee personal details fully. GDPR does too, but EU member states have their own employment rules. Staff privacy in the state often needs its own compliance track.

Sensitive types. The 19 state types partly overlap with GDPR Article 9. Immigration status and standalone account numbers are state-specific.

See the legal compliance guide for how these duties stack.

AI Vendor Requirements

The 2025 AI rules create clear duties for companies that use AI tools on consumer records.

Vendor contracts. Service providers must sign a written agreement. The agreement must cover four things. First, use records only for the stated purpose. Second, delete records when the service ends. Third, pass through consumer rights requests. Fourth, keep adequate security in place.

Automated decision notice. If an AI tool helps decide on credit, fraud, or jobs — consumers must be told. They must also get an opt-out.

AI training limits. If consumer records trained an AI model, that model has limits. It cannot be used for purposes that conflict with the original collection reason.

The direct fix for most teams: remove personal IDs before records enter any AI system. This meets the AI decision rules and cuts sensitive-type risk at the same time.

Learn how anonym.legal removes identifiers before AI processing at /security-compliance.

Key Points

State law covers most global companies with customers in the state. It adds 19 sensitive types, an active enforcement agency, and AI notice rules. GDPR compliance does not replace state compliance. The clearest step: remove personal IDs before records enter AI tools.

Read about data minimization at /docs/faq.

Sources

  • CPPA: California Privacy Protection Agency. cppa.ca.gov.
  • California Privacy Rights Act (CPRA) full text. leginfo.legislature.ca.gov.
  • CPPA: Automated Decision-Making Technology Regulations 2025. cppa.ca.gov.
  • California Age-Appropriate Design Code. leginfo.legislature.ca.gov.

Ready to protect your data?

Start anonymizing PII with 285+ entity types across 48 languages.

About this page

We update this page when our platform or the law changes.

Read our founder note for how we work.

Each change shows up in the timestamp at the top.

Related reading

We follow these rules

  • GDPR (EU 2016/679).
  • ISO/IEC 27001:2022.
  • NIS2 (EU 2022/2555).
  • HIPAA safe harbor under 45 CFR § 164.514(b)(2).

Our promise

We do not sell your data.

We do not train models on your text.

We store your files in Germany.

You can delete your account at any time.

You own your work.

Where we run

Our servers live in Falkenstein, Germany.

We use Hetzner. They hold ISO 27001 certification.

All data stays in the EU.

Backups run every day.

Need help?

Email support@anonym.legal.

We reply within one business day.

How we test

We run a full check suite on every release.

Each surface gets its own sweep script and report.

Human reviewers spot-check the output each week.

We track recall and precision on a labelled set.

Bad runs block the deploy.

What we never do

  • We never sell your information to third parties.
  • We never train models on what you upload.
  • We never keep your work after you delete it.
  • We never share keys with any outside firm.
  • We never run ads inside the product.

Plans in plain words

We sell credits, not seats.

One credit covers one short job.

Long jobs use a few credits each.

You can top up at any time.

Unused credits roll over each month.

Read the plans page for current rates.

Who built this

A small team of engineers and lawyers built this.

We ship from Europe and work in the open.

Our founder note spells out why we started.

Where to start

How the parts fit

A browser add-on cleans text inside Chrome.

A Word plug-in handles drafts in Office.

A small desktop tool works on whole folders.

An agent protocol link feeds large models safely.

All four share one core engine and one rule set.

Words from our team

We started this work after a lunch about cookies.

One friend kept getting odd ads on her phone.

We asked why a court file leaked through a draft.

We sketched the first build on a napkin that week.

By month three we had a tiny demo for a friend.

She used it on her first case the next day.

Common questions we hear

Can the tool read scanned PDFs? Yes, with OCR.

Does it work on long files? Yes, in small chunks.

Can I roll my own rule set? Yes, save it as a preset.

Does it run offline? The desktop build runs offline.

Do you keep my files? No, the cloud build wipes after each run.

Will it learn from my work? No, we never train on inputs.

A short tour of the workflow

Upload a file or paste a snippet of prose.

Pick the entities you want gone from the draft.

Choose a method: replace, mask, hash, encrypt, or redact.

Press run and watch the side panel show each hit.

Skim the result and tweak any rule that misfired.

Save the cleaned file or send it to a teammate.