AI Disclosure Notice

1. Where We May Use AI

AI tools may be used to assist with activities such as:

• Workflow analysis
• Information summarization
• Ticket summarization
• Classification
• Technical-triage assistance
• Retrieval of approved knowledge
• Identification of missing information
• Drafting
• Documentation
• Workflow recommendations
• Jira or issue-ticket preparation
• Customer-update preparation
• Knowledge-content preparation
• Automation development
• Internal productivity

The exact use of AI differs by project.

2. AI Is Not Infallible

AI-generated outputs may be:

• Incorrect
• Incomplete
• Misleading
• Outdated
• Based on insufficient context
• Inconsistent

AI output should not automatically be treated as verified fact.

In particular, a model-generated technical hypothesis does not become a confirmed root cause simply because an AI system produced it.

3. Human-Governed Automation

Our general approach is to give AI authority that is proportionate to the risk of the action.

Lower-risk activities may be more highly automated.

Higher-risk activities should generally use defined approval or escalation controls.

Depending on the workflow, human review may be required for:

• Customer-facing commitments
• Unverified technical conclusions
• Changes affecting important accounts
• Billing or refund decisions
• Contractual matters
• Security-sensitive actions
• High-severity issue closure
• Priority changes
• Sensitive customer communication

4. Disclosure of AI Interaction

Where you directly interact with an automated AI system operated by Jerry J. Hudson, and disclosure is appropriate or legally required, we intend to make the automated nature of that interaction reasonably clear.

This is particularly important when an interaction might otherwise reasonably be mistaken for direct human communication.

AI-transparency requirements are evolving in the United States. For example, Colorado legislation includes disclosure requirements for specified consumer interactions with AI systems, and relevant AI requirements are now in force following the state's 2025 amendments.

5. Website Scorecard

The B2B SaaS Escalation Health Scorecard is intended as an operational self-assessment.

Its result is not:

• A legal opinion
• A security certification
• A compliance certification
• A guarantee of business performance
• A complete AI-readiness assessment
• A substitute for a paid workflow audit

Where automated scoring or AI-assisted personalization is used, recommendations are based on the information supplied and the logic configured for the assessment.

6. AI in Client Engagements

Client AI workflows will be designed in accordance with the agreed-upon scope.

Before production deployment, relevant considerations may include:

• Permitted data sources
• System permissions
• Actions the AI may perform
• Actions the AI may only recommend
• Human-approval points
• Logging
• Monitoring
• Exceptions
• Error handling
• Escalation paths
• Customer-communication controls
• Knowledge-source restrictions

7. Client Responsibility

Clients remain responsible for determining whether a particular use of AI is appropriate for their organization, customers, contracts, industry, and legal obligations.

Jerry J. Hudson does not provide legal or regulatory certification merely by designing an AI workflow.

8. Personal and Confidential Information

Do not enter sensitive customer information, confidential credentials, or restricted business information into public Website AI features or forms unless specifically instructed to do so through an approved process.

When an AI provider processes client information as part of an engagement, data handling should be addressed in the agreed-upon project architecture.

9. AI Providers

Jerry J. Hudson may use third-party AI models, automation systems or related services.

Those providers may change model behavior, availability, features, prices, policies and technical interfaces.

10. Intellectual Property

AI-assisted content can involve complex intellectual-property considerations.

The applicable Terms of Service, SOW, and third-party provider terms govern ownership and permitted use of client deliverables.

Clients should not assume that every AI-generated output is unique or eligible for exclusive intellectual property protection.

11. No Fully Autonomous High-Risk Decision Making by Default

Jerry J. Hudson's standard consulting approach does not begin with unrestricted autonomous control over sensitive business decisions.

The preferred approach is to establish defined authority, appropriate human review, and clear escalation paths based on the risk of each action.

12. Continuous Monitoring

AI systems and underlying models change over time.

Production workflows may therefore require ongoing:

• Testing
• Monitoring
• Prompt or rule updates
• Knowledge updates
• Performance review
• Permission review
• Workflow adjustment

An automation that performs appropriately when launched should not automatically be assumed to remain appropriate indefinitely.