Maximum fine reference · Updated April 2026

AI Compliance Penalty Calculator

What is the maximum fine your business faces for non-compliance? Find the penalty structure for every major AI regulation — EU AI Act, GDPR, Colorado AI Act, NYC Local Law 144, and more.

Get my personalized fine exposure estimate — free

Penalty Structure by Regulation

Penalties are always the higher of the fixed amount or the revenue percentage. Multiple violations can stack. Revenue = global annual turnover of the entire group.

🇪🇺

EU AI Act

August 2026
Prohibited AI (Article 5)Critical

Using AI for real-time biometric surveillance in public spaces, social scoring, or other prohibited practices.

€30,000,000
or 6% of global annual turnover
High-Risk AI violations (Article 9–17)High

Deploying high-risk AI (hiring, credit, healthcare, education) without required conformity assessments.

€20,000,000
or 4% of global annual turnover
Incorrect information to authoritiesSignificant

Providing false, incomplete, or misleading information to market surveillance authorities.

€10,000,000
or 2% of global annual turnover

Key notes:

  • Fines apply to the higher of the fixed amount OR the revenue percentage
  • SMEs may receive reduced fines at authority discretion
  • Applies to any company with EU customers or employees
  • Prohibited AI rules: February 2025. General rules: August 2026.
🇪🇺

GDPR (Article 22 / AI Decision-Making)

Active now
Most serious violations (Art. 83(5))Critical

Unlawful automated decision-making with legal effects, no human review, or missing transparency disclosures.

€20,000,000
or 4% of global annual turnover
Less serious violations (Art. 83(4))High

Technical violations — missing DPA notifications, inadequate documentation, processor agreement gaps.

€10,000,000
or 2% of global annual turnover

Key notes:

  • GDPR is actively enforced — €4.5B in fines issued since 2018
  • Art. 22: individuals have right not to be subject to solely automated decisions
  • Every EU customer interaction with AI may trigger Art. 22 obligations
  • Ireland, France, and Spain are most active enforcers
🇬🇧

UK AI / UK GDPR

Active now
Most serious violationsCritical

Same scope as EU GDPR Article 83(5) — unlawful automated decisions, no human review.

£17,500,000
or 4% of global annual turnover
Standard violationsHigh

Documentation failures, processor agreements, notification gaps.

£8,700,000
or 2% of global annual turnover

Key notes:

  • ICO (Information Commissioner's Office) is the UK regulator
  • UK AI regulation framework expected 2025–2026 via sector regulators
  • UK GDPR mirrors EU GDPR post-Brexit
🇺🇸

Colorado AI Act (SB 24-205)

June 30, 2026
Violations of consumer rightsSignificant

Failing to provide required disclosures, impact assessments, or appeal rights for high-risk AI decisions.

$2,000 per violation
or Civil penalty per consumer affected
Willful violationsHigh

Knowing violation of consumer AI rights after receiving notice.

$20,000 per violation
or Enhanced civil penalty for intentional violations

Key notes:

  • Enforced by Colorado Attorney General
  • Applies to developers and deployers of "high-risk AI" affecting CO residents
  • "High-risk" = AI making consequential decisions in employment, housing, credit, insurance, education, healthcare
  • 2-year cure period for good-faith violations
🇺🇸

NYC Local Law 144 (Bias Audit)

Active since Jan 2023
Civil penaltiesSignificant

Not conducting annual bias audit of AI hiring tools or failing to post results publicly before using the tool.

$500/day
or Per day of violation
Continued violationsHigh

Continued use of automated employment decision tool without bias audit after receiving notice.

$1,500/day
or Repeat violations after initial citation

Key notes:

  • Applies to employment decisions about NYC candidates or employees
  • Annual bias audit must be completed by an independent auditor
  • Results must be posted on company website at least 10 days before tool use
  • Covers resume screening, interview scheduling, performance evaluation AI
🇺🇸

Illinois AI Video Interview Act (AAIA)

Active now
ViolationsSignificant

Using AI video interview analysis without prior written notice to candidates or sharing candidate data.

Actual damages
or Plus attorney fees per plaintiff

Key notes:

  • Illinois employees can bring private lawsuits
  • Must provide advance notice and consent before AI video interview analysis
  • Cannot share applicant videos with third parties without consent
  • Penalties include legal fees — class action risk
🇪🇺

EU Digital Services Act (DSA)

Active now (VLOPs Feb 2024)
Non-complianceHigh

Platforms with 45M+ EU users — recommender system transparency failures, failure to audit algorithms.

€—
or 6% of global annual turnover
Repeated/serious violationsCritical

Systemic violations by Very Large Online Platforms (VLOPs) or Search Engines (VLOSEs).

Temporary access ban
or European Commission can suspend access to EU market

Key notes:

  • Applies to platforms (marketplaces, social networks, app stores) with EU users
  • Recommender systems must offer non-profiling alternative
  • Annual independent audits required for VLOPs
  • Smaller platforms: lighter obligations under tiered framework

Example Calculations

Real-world scenarios showing how penalties are calculated. These are illustrative examples — actual enforcement varies by regulator.

SaaS startup, $2M revenue, EU customers, uses ChatGPT for automated customer triage

Max: €80,000
EU AI Act (High-Risk) + GDPR Art. 22

EU AI Act: 4% × €2M = €80,000. GDPR adds up to additional 4%.

US e-commerce, $10M revenue, uses AI for hiring decisions in NYC

Max: $182,500/yr
NYC Local Law 144

$500/day × 365 days = $182,500 per year of non-compliance

Enterprise software, $500M revenue, global customers, high-risk AI in healthcare

Max: €30,000,000+
EU AI Act + GDPR

EU AI Act: max €30M for prohibited AI. GDPR: up to €20M or 4% = €20M. Can stack.

HR tech startup, $5M revenue, CO + IL customers, AI resume screening

Max: $2,000 per applicant
Colorado AI Act + Illinois AAIA

CO: $2K per affected consumer. With 1,000 applicants = $2M maximum exposure.

Important: About These Figures

These are maximum statutory penalties. Regulators rarely impose the maximum fine. Actual enforcement considers: severity, intent, cooperation with regulators, self-reporting, and remediation steps.

However: GDPR regulators have issued maximum fines (Meta: €1.2B, WhatsApp: €225M). The risk is real, especially for larger companies or egregious violations.

Penalties can stack across regulations if multiple laws apply. A US company with EU customers making automated hiring decisions could face EU AI Act + GDPR + NYC LL144 all at once.

This tool is for informational purposes only and does not constitute legal advice. Consult qualified legal counsel for specific compliance guidance.

Know your exact exposure

The penalty calculator shows the maximum. ComplianceIQ calculates your actual exposure based on your specific AI systems, jurisdictions, and risk factors — then generates the documents to eliminate it.

Calculate my real exposure — free

Related Tools