The premise
AI can draft a customer disclosure letter for an AI vendor incident that is honest about scope without speculating beyond facts.
What AI does well here
- Lay out what happened, what data was involved, and what you have done
- Distinguish what is known vs under investigation
- Draft customer actions and contact points
What AI cannot do
- Decide whether the incident is reportable to regulators
- Speak for the vendor
- Make the legal disclosure decision
End-of-lesson check
15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-ethics-ai-vendor-incident-disclosure-letter-creators
What should a customer-facing disclosure letter primarily focus on regarding the incident?
- Promising it will never happen again with guarantees
- Detailing all technical specifications of the security breach
- Explaining why the vendor failed and who is to blame
- Providing known facts about what happened, what data was involved, and what actions have been taken
In an AI-drafted disclosure letter, what does it mean to distinguish what is known versus what is under investigation?
- Hiding information that is under investigation to avoid confusion
- Combining facts with speculation to appear thorough
- Making assumptions about the cause based on early reports
- Clearly stating what has been confirmed versus what is still being examined
What is a key limitation of using AI to draft incident disclosure letters for AI vendors?
- AI cannot write in a professional tone
- AI cannot produce written content about security events
- AI cannot determine legal reportability requirements or speak officially for the vendor
- AI cannot access any information about the incident
Who should review an AI-drafted incident disclosure letter before it is sent to customers?
- Any available employee to save time
- No one—it can be sent immediately
- The AI system that generated the draft
- Legal counsel and the incident commander
What tone is recommended for an incident disclosure letter to customers?
- Defensive and technical to show competence
- Apologetic and overly emotional to gain sympathy
- Direct, non-defensive, and free of jargon
- Casual and reassuring to minimize concern
What does the [verify] marker indicate when used in an AI-generated disclosure draft?
- Information that is irrelevant to the incident
- Information that requires verification before inclusion in the final version
- The information is definitely true and can be published
- Information that should be deleted from the letter
What is 'third-party risk' in the context of AI vendor incidents?
- Risk that customers pose to each other
- Risk that arises from using external AI vendors and their services
- Risk from the vendor's competitors
- Risk to the vendor's employees from internal systems
What should a disclosure letter include about the data involved in an incident?
- A complete list of all data the company has ever collected
- The specific data types that were affected by the incident
- No data information, as that would alarm customers
- Only encrypted data to show security measures
Why can AI not determine whether an incident is reportable to regulators?
- AI is prohibited from mentioning government agencies
- AI systems are not sophisticated enough to write about regulations
- Legal reporting requirements depend on specific laws and contractual obligations that require human legal judgment
- Regulators do not accept AI-generated reports
What is the purpose of including contact channels in an incident disclosure letter?
- To direct customers to purchase additional products
- To meet marketing requirements
- To track which customers read the letter
- To allow affected customers to ask questions and receive support
What happens after an AI drafts an incident disclosure letter?
- The letter is deleted to prevent premature disclosure
- The letter is sent immediately to all affected customers
- The letter is posted publicly on social media
- The letter goes through legal and incident commander review before any sending decision
What does it mean for a disclosure letter to have a 'non-defensive' tone?
- Accepting responsibility and providing facts without making excuses or blaming others
- Refusing to communicate with customers about the incident
- Denying any wrongdoing or responsibility
- Making excuses for what happened and explaining why it wasn't your fault
Why must disclosure timing be reviewed by legal counsel?
- Counsel enjoys reviewing disclosure letters
- Because legal requirements and contractual obligations often dictate specific timing for disclosure
- To delay the disclosure as long as possible to avoid negative press
- Timing is irrelevant to disclosure effectiveness
What distinguishes a disclosure draft suitable for review from one ready to send?
- Nothing—AI drafts are always perfect and need no changes
- The length of the document determines readiness
- The font and formatting indicate whether it is ready
- A review draft has not yet received approval from legal counsel and incident commander
Why is it important to avoid speculation in an incident disclosure letter?
- Customers prefer reading speculative theories
- Misinformation can cause unnecessary panic, create legal liability, and damage trust
- Speculation makes the letter more interesting and engaging
- Speculation is required by disclosure regulations