Responsible AI Policy

Purpose

This Responsible AI Policy describes the principles and expectations Corenth Technologies LLC applies when developing, configuring, providing, and supporting artificial intelligence features.

The purpose of this policy is to encourage AI use that is lawful, transparent, privacy-aware, secure, appropriately supervised, and designed to reduce foreseeable harm.

This policy supplements the Corenth Technologies Terms of Service, Privacy Policy, Security page, and Acceptable Use Policy.

Scope

This policy applies to AI-powered features provided through the Corenth Technologies website, platform, applications, automation tools, chatbots, assistants, workflows, integrations, APIs, and related services, collectively referred to as the “Service.”

It applies to Corenth Technologies and to customers, administrators, authorized users, contractors, integrations, and automated systems that configure, access, or use those features.

Customers may need additional policies, disclosures, assessments, professional oversight, or contractual controls based on their industry, location, data, users, and intended use.

Human-Centered AI

Corenth Technologies designs and provides AI features to assist human users rather than eliminate responsibility for important judgments, communications, decisions, or outcomes.

Customers and users should maintain meaningful human oversight where AI output could affect another person, organization, legal right, financial interest, safety, employment, health, reputation, or access to essential services.

Human oversight may include:

• Reviewing AI-generated content before publication.
• Verifying important facts, calculations, and claims.
• Escalating uncertain or sensitive matters to a qualified person.
• Allowing affected individuals to ask questions or request review where appropriate.
• Preventing automated systems from making unauthorized final decisions.

Accuracy and Reliability

AI-generated output can be inaccurate, incomplete, outdated, misleading, inconsistent, or unsuitable for a particular purpose.

Corenth Technologies does not guarantee that every AI output will be correct, complete, original, current, legally compliant, or appropriate for a customer’s intended use.

Users are responsible for:

• Reviewing output before relying on it.
• Checking important facts against reliable sources.
• Testing workflows before deploying them to customers or the public.
• Evaluating whether output is suitable for the relevant audience and purpose.
• Using qualified professional review where the subject requires specialized expertise.

Fairness and Nondiscrimination

AI systems may reflect limitations, gaps, or biases contained in training data, prompts, customer data, configurations, or third-party models.

Customers must not use the Service to unlawfully discriminate against individuals or groups based on protected characteristics or other legally protected status.

Appropriate fairness review may include:

• Testing results across relevant groups and use scenarios.
• Reviewing prompts, datasets, rules, and workflow logic.
• Investigating unequal error rates or inappropriate outcomes.
• Providing human review and correction mechanisms.
• Suspending an AI-supported workflow when serious harm or discrimination is identified.

Transparency and Disclosure

Users should not be materially misled about whether they are interacting with an automated system or receiving AI-generated content.

Customers should provide appropriate notice when AI is used in customer-facing chatbots, communications, recommendations, content-generation systems, or automated workflows.

Appropriate disclosure may include:

• Identifying a chatbot or assistant as AI-powered.
• Explaining when content has been generated or materially modified by AI.
• Informing users when AI contributes to an important recommendation or decision.
• Providing a method to reach a human where appropriate.
• Avoiding false claims that AI output was personally authored or verified by a professional.

Privacy and Data Governance

AI features must be used in a manner consistent with applicable privacy laws, contractual requirements, customer notices, permissions, consent obligations, and the Corenth Technologies Privacy Policy.

Customers and users may not:

• Submit information they do not have the legal right or authority to process.
• Include unnecessary sensitive or regulated information in prompts or uploaded materials.
• Use AI features for unauthorized surveillance, profiling, identification, or monitoring.
• Use personal information for purposes that are incompatible with required notices, permissions, or legal bases.
• Attempt to obtain another customer’s data, prompts, documents, outputs, or confidential information.

Safety and Security

AI features should be designed and configured to reduce foreseeable risks involving unauthorized access, prompt injection, malicious content, data exposure, credential theft, insecure integrations, abusive automation, and harmful output.

Relevant safeguards may include:

• Access controls and role-based permissions.
• Secure handling of API keys and provider credentials.
• Input and output validation.
• Usage limits and rate controls.
• Logging of important administrative actions.
• Restrictions on tools, integrations, and sensitive workflows.
• Human escalation for serious safety or security concerns.

Safeguards may vary by feature, provider, integration, plan, configuration, and identified risk. Corenth Technologies reviews and improves relevant controls as the production Service evolves.

Prohibited AI Uses

AI features must not be used to generate, facilitate, distribute, promote, or support prohibited activity.

This list is not exhaustive. Additional restrictions appear in the Corenth Technologies Acceptable Use Policy.

High-Impact Uses

AI features must not serve as the sole or final decision-making authority for matters that may significantly affect a person’s rights, safety, livelihood, reputation, or access to essential services.

High-impact areas may include:

• Employment, hiring, promotion, or termination.
• Credit, lending, banking, or financial eligibility.
• Housing, tenancy, or property access.
• Healthcare, diagnosis, treatment, or patient safety.
• Insurance eligibility, pricing, or claims.
• Education admissions, grading, or discipline.
• Legal rights, legal status, or law enforcement.
• Government benefits, public services, or essential resources.

Customers using AI in these areas are responsible for qualified human review, appropriate testing, documentation, transparency, correction procedures, appeal processes, and compliance with applicable requirements.

Content and Intellectual Property

Customers are responsible for ensuring that prompts, documents, images, recordings, datasets, instructions, and other materials submitted to AI features may be lawfully used.

AI-generated content may resemble existing material, may not be unique, and may require additional review before commercial use, publication, registration, licensing, or distribution.

Users should:

• Review output for possible copyright, trademark, privacy, publicity, confidentiality, and contractual concerns.
• Avoid requesting unlawful reproductions of protected material.
• Obtain required permissions and licenses before using third-party content.
• Avoid representing AI-generated content as guaranteed to be original or free from third-party rights.

Third-Party AI Providers

Corenth Technologies may rely on third-party AI models, infrastructure providers, hosting providers, databases, integrations, and other technology vendors.

Third-party providers may have their own technical limitations, availability conditions, acceptable-use rules, privacy practices, data-retention settings, and safety controls.

Corenth Technologies reviews and maintains information about relevant AI providers and subprocessors used by the Service, including:

• The providers and models used by each feature.
• The information transmitted to each provider.
• Provider retention and training settings.
• Processing locations and transfer mechanisms.
• Security and contractual protections.
• Model limitations and supported use cases.
• Provider outage and replacement procedures.

Testing, Monitoring and Incident Response

Corenth Technologies uses risk-based procedures for evaluating AI features before and after production release.

Those procedures may include:

• Testing important prompts and workflow scenarios.
• Evaluating accuracy, relevance, safety, and reliability.
• Reviewing failures, inappropriate output, and customer reports.
• Tracking material changes to providers, models, prompts, or configurations.
• Restricting or disabling features that create unacceptable risk.
• Documenting serious incidents and corrective actions.

Customer Responsibilities

Customers remain responsible for the way they configure, deploy, supervise, and use AI features.

Customer responsibilities include:

• Using AI features only for lawful and authorized purposes.
• Providing required notices, disclosures, and consent.
• Reviewing output before relying on or distributing it.
• Protecting account credentials and access to the customer workspace.
• Testing workflows before production use.
• Maintaining appropriate human oversight.
• Investigating complaints, errors, and harmful outcomes.
• Complying with professional, industry, privacy, consumer, employment, and other applicable requirements.

Customers should stop or restrict an AI-supported workflow when they become aware that it is producing unsafe, discriminatory, materially inaccurate, unlawful, or unauthorized results.

Changes to This Policy

Corenth Technologies may update this policy as the Service, AI providers, models, features, risks, laws, customer needs, and business practices change.

Updated versions will be posted on this page with a revised “Last updated” date.

Material changes may also be communicated through email, an account notice, an in-product message, or another reasonable method where appropriate.

Contact Information

Questions, concerns, or reports regarding responsible AI, harmful output, AI-supported decisions, or this policy may be sent to:

cdean@corenthtech.com is the current contact for responsible AI questions, concerns, and reports. Corenth Technologies may introduce dedicated AI governance, legal, privacy, or safety contact addresses in the future.