What is the Right to Explanation?
The right to explanation is a legal and ethical principle that allows individuals to demand clear, understandable explanations when artificial intelligence systems make decisions that significantly affect their lives. In 2026, as AI systems increasingly determine everything from credit approvals to medical diagnoses, understanding your rights to transparency has become essential.
According to Article 22 of the GDPR, individuals have the right not to be subject to decisions based solely on automated processing that produce legal or similarly significant effects. This includes the right to obtain meaningful information about the logic involved in such decisions. While the GDPR applies primarily to EU citizens, similar principles are emerging globally, including in California's Consumer Privacy Act (CCPA) and various state-level AI regulations in the United States.
"Transparency isn't just about opening the black box—it's about ensuring people understand how AI decisions impact their fundamental rights and opportunities."
Dr. Sandra Wachter, Oxford Internet Institute
This comprehensive guide will walk you through when you can demand AI transparency, how to exercise these rights, and what to expect from the process in 2026.
When Can You Demand an AI Explanation?
Not every AI interaction grants you the right to an explanation. Understanding when this right applies is crucial for effectively exercising your rights.
Situations Where You Have Strong Rights
You can typically demand explanations in these high-impact scenarios:
- Financial decisions: Credit denials, loan rejections, insurance premium calculations, or fraud detection flags
- Employment: Automated resume screening, AI-driven hiring decisions, performance evaluations, or termination recommendations
- Healthcare: AI-assisted diagnoses, treatment recommendations, or insurance coverage determinations
- Legal and criminal justice: Risk assessment scores, bail recommendations, or parole decisions
- Education: Automated admissions decisions, scholarship allocations, or academic evaluations
- Housing: Rental application rejections or tenant screening results
- Government services: Benefit eligibility determinations or public service access
Geographic Considerations
Your rights vary significantly based on location:
- European Union: Strongest protections under GDPR and the forthcoming EU AI Act
- United States: Patchwork of state laws, with California, Colorado, and Virginia leading; sector-specific federal regulations (FCRA for credit, ECOA for lending)
- United Kingdom: UK GDPR and developing AI-specific regulations
- Canada: PIPEDA protections and proposed Artificial Intelligence and Data Act (AIDA)
Prerequisites: What You Need Before Requesting an Explanation
Before exercising your right to explanation, gather the following information:
- Documentation of the AI decision: Save emails, letters, or screenshots showing the automated decision
- Timeline records: Note dates of application, decision, and any communications
- Identification of the decision-maker: Know which company or organization made the AI-driven decision
- Relevant privacy policies: Download and save the organization's privacy policy and AI use disclosures
- Your personal data: Gather any information you submitted that the AI system processed
- Jurisdiction verification: Confirm which laws apply based on your location and the organization's location
According to EPIC's algorithmic transparency guidelines, having comprehensive documentation significantly increases the likelihood of receiving a meaningful explanation.
Step-by-Step Guide: How to Request an AI Explanation
Step 1: Verify Your Right to an Explanation
First, confirm that the decision qualifies for explanation rights:
Checklist:
☐ Was the decision made solely or primarily by AI/automated processing?
☐ Does the decision have legal or similarly significant effects?
☐ Did you receive notice that AI was used in the decision?
☐ Are you in a jurisdiction with relevant protections?
☐ Did the decision occur within the statute of limitations (typically 1-3 years)?
If you can check at least the first two boxes, you likely have grounds to request an explanation.
Step 2: Identify the Correct Contact
Locate the appropriate department or individual:
- Check the organization's privacy policy for a Data Protection Officer (DPO) or Privacy Officer contact
- Look for "AI Transparency" or "Algorithmic Decision Appeals" sections on their website
- Review the decision notice for contact information
- For EU organizations, find their GDPR representative
- For US companies, contact their legal or compliance department
Step 3: Draft Your Explanation Request
Use this template as a starting point, customizing it for your situation:
[Your Name]
[Your Address]
[Your Email]
[Date]
[Organization Name]
[Department/DPO Name if known]
[Organization Address]
Re: Request for Explanation of Automated Decision Under [GDPR/CCPA/Applicable Law]
Dear [Title/Department],
I am writing to exercise my right to receive an explanation regarding an automated
decision made by your organization on [DATE].
Decision Details:
- Type of decision: [e.g., credit denial, job application rejection]
- Reference number (if applicable): [NUMBER]
- Date of decision: [DATE]
Under [GDPR Article 22/CCPA Section 1798.185/applicable regulation], I request:
1. Confirmation that automated decision-making or AI was used
2. Meaningful information about the logic involved in the decision
3. The significance and envisaged consequences of such processing
4. The specific factors that influenced the decision in my case
5. Information about the data sources used
6. Whether the decision can be appealed or reviewed by a human
Please provide this explanation in clear, accessible language within the legally
required timeframe [30 days for GDPR, 45 days for CCPA].
I have attached [relevant documentation] to verify my identity and the decision in question.
Thank you for your prompt attention to this matter.
Sincerely,
[Your Signature]
[Your Name]
[Screenshot suggestion: Example of a completed request letter with sensitive information redacted]
Step 4: Submit Your Request Through Multiple Channels
Maximize your chances of a response by submitting through multiple channels:
- Email: Send to the DPO or privacy contact (keep a copy with timestamp)
- Certified mail: For important decisions, send a physical letter with tracking
- Online portal: Use the organization's privacy request portal if available
- Phone follow-up: Call 3-5 business days after submission to confirm receipt
According to IAPP best practices, multi-channel submission increases response rates by approximately 40%.
Step 5: Document Everything
Create a comprehensive record:
Documentation Folder Structure:
📁 AI_Explanation_Request_[Organization]_[Date]
📄 Original_Decision_Notice.pdf
📄 Request_Letter_Sent.pdf
📄 Email_Confirmation.pdf
📄 Tracking_Numbers.txt
📄 Timeline_Log.txt
📄 Response_Received.pdf (when available)
📁 Supporting_Documents/
📄 Privacy_Policy_Snapshot.pdf
📄 Terms_of_Service.pdf
📄 Application_Materials.pdf
Step 6: Wait for Response (and Know Your Timelines)
Response timelines vary by jurisdiction:
- GDPR (EU/UK): 30 days (extendable to 60 days for complex requests)
- CCPA (California): 45 days (extendable to 90 days)
- FCRA (US credit decisions): 30 days for adverse action explanations
- State-specific laws: Typically 30-60 days
Set calendar reminders for follow-up dates.
Understanding the Explanation You Receive
What to Expect in a Proper Explanation
A meaningful AI explanation should include:
- Confirmation of automation: Clear statement that AI/automated processing was used
- Type of AI system: General description (e.g., "machine learning model," "rules-based system")
- Key decision factors: The main variables that influenced your specific outcome
- Data sources: What information was used (credit reports, application data, etc.)
- Decision logic: How factors were weighted or combined (in accessible language)
- Human oversight: Whether humans reviewed or can review the decision
- Appeal process: How to challenge or request reconsideration
"A good explanation doesn't require understanding the mathematics—it requires understanding what mattered in your case and why."
Andrew Burt, Chief Legal Officer, Immuta
Real-World Example: Credit Decision Explanation
Here's what a compliant explanation might look like:
Sample Explanation from Financial Institution:
"Your credit application was processed using our automated underwriting system,
which uses machine learning to assess credit risk.
In your specific case, the following factors most significantly influenced
the decision:
1. Credit utilization ratio (38% weight): Your current utilization of 78%
exceeded our threshold of 50%
2. Length of credit history (25% weight): 2.5 years falls below our
preferred minimum of 5 years
3. Recent credit inquiries (20% weight): 4 inquiries in the past 6 months
4. Payment history (17% weight): One late payment in the past 12 months
Data sources: Experian credit report dated [DATE], application information
you provided on [DATE]
You have the right to request human review of this decision. Please contact
our customer service team at [NUMBER] within 60 days to initiate a review."
[Screenshot suggestion: Annotated example explanation highlighting key required elements]
Advanced Strategies: When Initial Requests Fail
Escalation Path
If you don't receive a satisfactory response, follow this escalation sequence:
- Send a follow-up request (Week 4-5): Reference your original request and applicable deadlines
- Contact executive customer service (Week 6-7): Escalate to senior management or executive support
- File a regulatory complaint (Week 8-10): Submit complaints to relevant authorities
- Seek legal assistance (Week 10+): Consult with privacy attorneys or consumer protection organizations
Regulatory Complaint Resources
File complaints with these authorities based on your location:
- EU/EEA: Your national Data Protection Authority (DPA) - find yours at EDPB member list
- UK: Information Commissioner's Office (ICO)
- California: California Attorney General's Office
- Credit decisions (US): Consumer Financial Protection Bureau (CFPB)
- Employment (US): Equal Employment Opportunity Commission (EEOC)
Using Subject Access Requests (SARs) for Deeper Insight
If explanations are insufficient, file a broader data access request:
Enhanced SAR Template Addition:
"In addition to the explanation requested above, I am also exercising my right
to access under [GDPR Article 15/CCPA Section 1798.110] to receive:
1. All personal data you hold about me that was used in the automated decision
2. The categories of personal data processed
3. The recipients or categories of recipients of my data
4. The retention period for this data
5. Technical documentation about the AI system (to the extent it relates to
my personal data processing)
6. Any available information about the training data used for the AI model"
Tips & Best Practices for 2026
Maximizing Your Chances of Success
- Act quickly: Submit requests within 30 days of the decision while it's still fresh and documented
- Be specific: Reference exact dates, decision types, and applicable laws
- Stay professional: Maintain courteous, business-like tone even if frustrated
- Know the law: Cite specific legal provisions relevant to your jurisdiction
- Request human review: Always ask for the option of human reconsideration
- Join collective actions: If many people face similar issues, group complaints carry more weight
Common Mistakes to Avoid
- Being too vague: "I want to know about AI" is less effective than specific requests
- Accepting boilerplate responses: Generic explanations that don't address your specific case are inadequate
- Missing deadlines: Don't let response windows expire without follow-up
- Requesting impossible information: Trade secrets and proprietary algorithms may be legitimately withheld, but decision factors in your case cannot be
- Going to court immediately: Exhaust administrative remedies first; regulatory complaints are faster and free
Leveraging New 2026 Tools and Resources
Take advantage of emerging support services:
- AI explanation templates: Organizations like the Electronic Frontier Foundation (EFF) offer jurisdiction-specific templates
- Automated request platforms: Services like DataRequests.org help generate and track requests
- Legal aid organizations: Many privacy-focused nonprofits offer free consultations
- AI literacy programs: Understanding AI basics helps you ask better questions and evaluate responses
"The right to explanation is only as powerful as people's willingness to exercise it. In 2026, we're seeing a significant increase in successful requests as both individuals and organizations become more sophisticated."
Cynthia Rudin, Professor of Computer Science, Duke University
Common Issues & Troubleshooting
Issue 1: "We Don't Use AI" (But You Suspect They Do)
Solution: Request clarification on what automated processing or decision-support tools are used. Many organizations use AI but don't label it as such. Ask specifically about:
- Automated scoring systems
- Risk assessment tools
- Algorithmic ranking or filtering
- Machine learning models
- Decision support software
Issue 2: Receiving Only Generic Information
Solution: Respond with a clarification request:
"Thank you for your response dated [DATE]. However, the information provided
appears to be general information about your AI system rather than specific
factors that influenced my individual case.
I am specifically requesting:
- The factors that were most influential in MY decision
- The values or scores for these factors in MY case
- How MY data compared to approval thresholds
Please provide case-specific information as required under [applicable law]."
Issue 3: "This Information is Proprietary"
Solution: Clarify that you're not requesting trade secrets, just how the system was applied to your data. According to GDPR Recital 63, trade secret protection cannot completely deny the right to information. Request:
- General categories of factors (not exact algorithms)
- Relative importance of factors in your case
- Whether you met or missed specific thresholds
- What data was used about you specifically
Issue 4: No Response After Legal Deadline
Solution: Immediately file a regulatory complaint. Include:
- Copy of your original request with proof of delivery
- Documentation showing the deadline has passed
- Any partial or inadequate responses received
- Impact of the decision on you (financial, professional, personal)
Issue 5: Explanation Reveals Discriminatory Factors
Solution: This is serious and may indicate illegal discrimination. Take these steps:
- Document everything immediately
- File complaints with civil rights enforcement agencies (EEOC, HUD, etc.)
- Contact civil rights legal organizations
- Consider joining class action efforts if others are affected
- Report to relevant regulatory authorities
Organizations like the ACLU and Algorithm Justice League can provide guidance.
FAQ: Right to Explanation in 2026
Do I have to pay for an AI explanation?
No. Under GDPR, CCPA, and most privacy laws, organizations cannot charge fees for explanation requests unless they are manifestly unfounded or excessive (e.g., multiple identical requests in short succession).
Can I request an explanation for any AI decision?
Not necessarily. The right typically applies to decisions that have legal or similarly significant effects. Receiving AI-generated content recommendations on social media generally wouldn't qualify, but an AI-driven credit denial would.
What if the AI decision was partially made by humans?
If AI played a significant role in the decision, you can still request information about the automated components. The organization should clarify what parts were automated and what parts involved human judgment.
How technical will the explanation be?
Explanations must be in "clear and plain language" accessible to average individuals. You shouldn't need a computer science degree to understand the basic factors that affected your case.
Can my employer use AI to make decisions about me?
Yes, but with limitations. In 2026, many jurisdictions require employers to notify employees about AI use in hiring, performance evaluation, and termination decisions. You generally have the right to know when AI is used and to request human review.
What if I'm not satisfied with the explanation?
You can request clarification, file regulatory complaints, request human review of the decision, or pursue legal remedies. Many privacy laws include provisions for compensation if your rights are violated.
Conclusion: Empowering Yourself in the AI Age
The right to explanation is a powerful tool for accountability in our increasingly AI-driven world. In 2026, as artificial intelligence systems make more consequential decisions about employment, credit, healthcare, and fundamental opportunities, understanding and exercising these rights has never been more important.
Key takeaways for exercising your right to explanation:
- Know when you have the right: Focus on decisions with significant legal or practical effects
- Document everything: Maintain comprehensive records from the initial decision through resolution
- Be specific and persistent: Generic requests get generic responses; specific, well-documented requests get results
- Use available resources: Privacy organizations, templates, and regulatory authorities are there to help
- Don't accept inadequate responses: You're entitled to meaningful, case-specific information
- Escalate when necessary: Regulatory complaints and legal action are legitimate tools when organizations don't comply
Next Steps
- Review recent AI-driven decisions: Identify any decisions in the past year that may warrant explanation requests
- Bookmark key resources: Save templates, regulatory contact information, and relevant laws
- Stay informed: Privacy and AI regulations continue evolving in 2026; subscribe to updates from organizations like IAPP or EFF
- Educate others: Share this knowledge with friends, family, and colleagues who may face AI-driven decisions
- Advocate for stronger protections: Support legislative efforts to expand transparency rights in your jurisdiction
Remember: transparency is not just a legal right—it's essential for ensuring AI systems serve humanity fairly and equitably. By exercising your right to explanation, you're not only protecting yourself but contributing to a more accountable AI ecosystem for everyone.
References
- GDPR Article 22 - Automated individual decision-making
- General Data Protection Regulation (GDPR) - Complete Text
- California Consumer Privacy Act (CCPA) - Official Resource
- EU Artificial Intelligence Act - Official Information
- Canada's Artificial Intelligence and Data Act (AIDA)
- EPIC - Algorithmic Transparency Guidelines
- IAPP - How to Make a GDPR Request
- European Data Protection Board - Member List
- UK Information Commissioner's Office - Make a Complaint
- Consumer Financial Protection Bureau - Submit a Complaint
- EEOC - How to File a Charge of Employment Discrimination
- Electronic Frontier Foundation - Digital Privacy Resources
- DataRequests.org - Automated Privacy Request Platform
- GDPR Recital 63 - Right to Information
- American Civil Liberties Union - Civil Rights Resources
- International Association of Privacy Professionals
Disclaimer: This article provides general information about the right to explanation and is current as of February 01, 2026. It does not constitute legal advice. For specific legal situations, consult with a qualified attorney in your jurisdiction. Privacy laws and AI regulations continue to evolve, so always verify current requirements in your location.
Cover image: AI generated image by Google Imagen