LifeCloneAI Responsible AI and Acceptable Use Policy

Effective date: 1 September 2026

1. Purpose and scope

This policy sets minimum rules for the use of LifeCloneAI services, deliverables, automations, agents, integrations, and AI-enabled workflows. It applies to Customers, authorised users, and anyone acting through their accounts or credentials.

The policy supplements our Service Terms and each order. It does not replace a project-specific risk assessment or data processing agreement.

2. Core responsibilities

You must use the services lawfully and within your authority, provide accurate context, apply human oversight proportionate to the risk, verify material outputs before operational use or publication, respect privacy and intellectual property, protect credentials, maintain appropriate backups and approval gates, and report suspected misuse or incidents promptly.

3. Prohibited uses

You must not use the services or deliverables to:

• Break the law, evade legal obligations, or facilitate another person's unlawful conduct.
• Access, test, scan, disrupt, damage, or control a system, account, network, or dataset without appropriate authorisation.
• Create or distribute malware, credential theft, destructive code, covert persistence, or instructions intended to cause unauthorised harm.
• Deceive people through impersonation, fabricated identity, materially misleading synthetic media, fake evidence, fraudulent reviews, or deceptive commercial practices.
• Conduct harassment, stalking, intimidation, blackmail, exploitation, or targeted abuse.
• Generate or distribute child sexual abuse material, sexual content involving minors, non-consensual intimate imagery, or content that facilitates sexual exploitation.
• Unlawfully discriminate or make decisions based on protected or highly sensitive traits.
• Infer or exploit sensitive personal data in an unlawful, unexpected, or materially harmful way.
• Conduct biometric identification, emotion recognition, surveillance, or location tracking without a documented lawful basis, necessity assessment, proportionality assessment, and explicit project approval.
• Send spam, scrape or enrich personal data unlawfully, evade opt-outs, bypass platform limits, or conduct outreach without required consent, authority, and suppression controls.
• Manipulate civic participation, voting, public institutions, or political opinion through deception or unlawful targeting.
• Present AI output as professional legal, medical, financial, tax, safety, or regulatory advice when it has not been reviewed and delivered by an appropriately qualified person.
• Defeat provider safeguards, rate limits, access controls, licence restrictions, monitoring, or usage policies.
• Reverse engineer or extract protected models, data, prompts, or systems except where expressly authorised by law and contract.
• Create a material risk of death, physical injury, critical-infrastructure disruption, or another severe and foreseeable harm without a separately agreed safety case and controls.

4. High-impact decisions

AI output must not be the sole basis for a decision that determines or materially affects a person's access to employment, housing, education, credit, insurance, healthcare, legal services, immigration, public benefits, essential services, or another similarly significant opportunity.

An approved workflow touching a high-impact domain must have a documented purpose and accountable owner, meaningful human review, input and output quality checks, bias and accessibility assessment, an explanation and challenge route where required, logging and incident handling, periodic review, and a safe manual fallback.

LifeCloneAI may decline a high-impact use even if it is technically possible.

5. Accuracy, disclosure, and human review

AI systems can produce plausible but false, incomplete, biased, or outdated content. Material factual claims, calculations, citations, legal requirements, eligibility decisions, and external actions must be independently checked.

Where a reasonable person would need to know that content or an interaction is AI-generated, materially altered, or automated, you must provide an appropriate disclosure. Do not use a synthetic voice, likeness, signature, or identity without appropriate rights and consent.

6. Data, privacy, and confidentiality

Use the minimum data necessary. Do not upload personal, confidential, regulated, or proprietary data to an AI or automation provider unless the data flow is authorised, the provider is approved, the settings are appropriate, and the processing is covered by relevant privacy and data processing terms.

Do not place passwords, secret keys, payment-card data, authentication tokens, or unredacted production secrets in prompts or general chat fields. Customer materials must not be used to train a general-purpose model unless the relevant controller and rights holder expressly authorise it in writing and all legal and contractual conditions are satisfied.

7. Intellectual property and third-party content

You are responsible for checking that you have the rights needed for prompts, source materials, data, branding, likenesses, and intended output use. AI output may resemble third-party material and may not be exclusive or protectable.

Comply with open-source licences, data licences, website terms, access restrictions, and provider usage policies. Do not remove attribution, provenance, watermarking, or rights-management information where doing so is unlawful or misleading.

8. External actions and approvals

An AI recommendation, draft, score, or tool result is not approval to take an external action. Workflows that can publish content, spend money, modify production, contact people, change permissions, delete data, sign agreements, or submit regulated information must use explicit approval and verification controls proportionate to the risk.

Approval should identify the target, payload or change, account or tenant, expected effect, and rollback or recovery route. Material external changes should be independently read back.

9. Security testing and research

Security research, penetration testing, vulnerability validation, or adversarial simulation is allowed only with documented authority, scope, time window, data-handling rules, and escalation contacts. Testing must minimise disruption and follow coordinated-disclosure requirements where applicable.

10. Monitoring and records

LifeCloneAI may use proportionate logging and review to secure the services, investigate incidents, enforce this policy, and meet legal obligations. Monitoring involving personal data is governed by our Privacy Policy.

Customers should retain records appropriate to the risk, including material instructions, model or workflow versions, approvals, external-action receipts, corrections, and incidents.

11. Reporting concerns

Report suspected abuse, unsafe behaviour, rights infringement, privacy issues, or security incidents to sales@lifecloneai.com. Include enough information to identify the affected service and risk, but do not send secrets or unnecessary personal data through ordinary email.

We will review good-faith reports promptly, take proportionate steps to contain material risk, and keep the reporting party reasonably informed where lawful and appropriate.

12. Enforcement

Where reasonably necessary and proportionate, LifeCloneAI may require corrective action, limit a workflow, suspend affected access, preserve relevant evidence, or terminate an order for material or repeated breach. We may act immediately where there is a credible risk of serious harm, unlawful conduct, security compromise, or rights violation.

Where lawful and practicable, we will explain the reason, limit the action to what is necessary, and provide a contact route to challenge a mistake.

13. Changes to this policy

We may update this policy to address legal, technical, security, or service changes. Material changes will be communicated as required and will not retrospectively remove accrued contractual rights.

14. Contact

LIFECLONEAI LTD, company number 16337860, Flat 6 4 Waterhouse Lane, Kingswood, Tadworth, England, KT20 6EB.

Email: sales@lifecloneai.com