Skip to main content
AI Policy

AI Acceptable Use & Transparency Policy

Last Updated: 7th of September 2026Effective Date: 7th of September 2026

This AI Acceptable Use & Transparency Policy ("Policy") forms part of the Outlex Terms of Service or any other written agreement governing the Customer's use of the Services ("Agreement").

This Policy explains how AI is used in Outlex, the limits of AI Outputs, prohibited and restricted uses, user responsibilities, transparency measures and human oversight expectations.

Capitalised terms not defined in this Policy have the meanings given in the Agreement.

1. About Outlex and Lexi

Outlex provides an AI legal workspace for business and professional users.

The Services may include AI-enabled features such as Lexi, document analysis, contract review, legal Q&A, drafting, redlining, summarisation, workflow automation, matter preparation, legal task support and integrations. Lawyer handoff and Human Legal Support are available only for Customers, plans, matters and jurisdictions where Outlex expressly makes those services available.

Lexi is an AI legal agent. Lexi is not a lawyer, law firm, solicitor, attorney, court, regulator, public authority or substitute for qualified professional judgment.

Where Human Legal Support is expressly available for the relevant Customer and jurisdiction, Customers may request support from appropriately qualified Professionals through the Services. Availability of Lexi or jurisdiction-specific AI functionality does not imply availability of Human Legal Support.

Unless Outlex expressly identifies content as Human-Reviewed, Customers should assume that content generated through Lexi or another AI feature has not been reviewed, approved or adopted by a Professional.

2. Purpose of this Policy

This Policy is intended to:

  • promote safe, lawful and responsible use of AI within Outlex;
  • explain the role and limits of AI Outputs;
  • define prohibited and restricted uses;
  • support transparency for users interacting with AI;
  • support human oversight and professional review;
  • support compliance with applicable artificial intelligence, data protection, equality, consumer, professional-responsibility and other laws and regulatory requirements.

This Policy does not itself determine whether a specific Customer use of Outlex is lawful. Customers remain responsible for assessing laws applicable to their own deployment and use of AI.

3. Transparency: when you are using AI

Outlex clearly identifies Lexi as an AI system so that users are informed that they are interacting with AI from the beginning of the interaction.

Outlex may use AI to generate, summarise, classify, review, compare, explain, analyse or draft content. AI-generated or AI-assisted content may appear in chat responses, document comments, summaries, redlines, task outputs, checklists, templates, suggested clauses, legal explanations, workflow recommendations and other product features.

Where relevant to the feature and context, Outlex may use labels, status indicators, descriptions, disclaimers or other user-interface cues to distinguish AI-generated content from content that has actually received human professional review.

An AI Output should not be treated as Human-Reviewed merely because a Professional may have access to the same workspace or matter.

Where applicable law requires additional transparency concerning AI-generated or AI-manipulated content, Outlex may apply or preserve appropriate technical markings, metadata, labels or other disclosure mechanisms.

4. AI-generated content marking and detectability

Where required by applicable law, Outlex will take technically feasible and proportionate measures designed to enable AI-generated or AI-manipulated content to be identified or detected as artificially generated or manipulated.

Such measures may include machine-readable markings, metadata, technical provenance mechanisms or other methods supported by the relevant AI models, infrastructure and output formats.

Customers and users must not intentionally remove, alter or circumvent AI-related technical markings or disclosures where doing so would be unlawful, deceptive or inconsistent with this Policy.

5. Human oversight

Outlex is designed to support human decision-making, not replace it.

Users must maintain a level of human oversight appropriate to the context, including by:

  • reviewing AI Outputs for accuracy, completeness and relevance;
  • checking the applicable jurisdiction, law, dates and factual assumptions;
  • considering the potential impact of acting on an AI Output;
  • determining whether professional legal or other specialist review is required;
  • escalating high-risk, unclear, unusual or sensitive matters to an appropriately qualified professional; and
  • ensuring that decisions requiring human judgment remain subject to meaningful human control.

Customers are responsible for deciding when qualified professional advice is required and whether Human Legal Support is available through Outlex for the relevant jurisdiction. Where it is unavailable, Customers should obtain independent local professional advice where appropriate.

6. Customer responsibilities

Customers and users must use the Services lawfully, responsibly and consistently with the Agreement, this Policy and applicable law.

Customers are responsible for:

  • ensuring that they have the right to upload, process or share Customer Data through the Services;
  • ensuring that prompts, documents and instructions are lawful and appropriate;
  • identifying the jurisdiction relevant to each matter;
  • ensuring their intended use of AI is permitted under applicable law;
  • determining whether they act as a provider, deployer, developer, operator or other regulated actor under applicable AI laws;
  • conducting impact, risk, bias or other assessments where required for their own use of AI;
  • maintaining appropriate human oversight over material decisions;
  • reviewing AI Outputs before use;
  • avoiding overreliance on AI;
  • implementing internal policies and training appropriate to their own AI deployment;
  • complying with sector-specific or professional restrictions applicable to their organisation;
  • ensuring that the Services are not used for prohibited or restricted purposes;
  • ensuring that any use of AI in their organisation complies with applicable data protection, employment, professional, regulatory and AI laws; and
  • not using Outlex to circumvent legal requirements that would apply if the activity were conducted without AI.

7. Prohibited uses

Customers and users must not use the Services to:

  • violate applicable law, regulation, court order, professional duty or third-party rights;
  • generate, facilitate or conceal fraud, deception, unlawful conduct or regulatory evasion;
  • produce illegal, discriminatory, harassing, defamatory, hateful, exploitative or harmful content;
  • infringe intellectual property, confidentiality, trade secret, privacy or data protection rights;
  • generate malware, phishing, credential theft, unauthorised access, harmful code or cybersecurity abuse;
  • impersonate another person or misrepresent AI Outputs as human-generated where that would be misleading or unlawful;
  • create or distribute deceptive deepfakes, synthetic media or manipulated content in a way that misleads others;
  • process personal data without a lawful basis or required notices;
  • submit data that the Customer is not authorised to process through the Services;
  • use the Services to make decisions in a way that unlawfully discriminates against individuals or groups;
  • bypass usage limits, security controls, access controls, monitoring, filters or technical safeguards;
  • reverse engineer, extract, scrape or misuse the Services, models, prompts, outputs or system behaviour;
  • use the Services in a way that could compromise legal privilege, confidentiality or professional secrecy without appropriate safeguards;
  • deploy intentionally manipulative, deceptive or subliminal techniques designed or reasonably likely to materially distort a person's behaviour in a way that causes or is reasonably likely to cause significant harm, where the EU AI Act applies;
  • exploit vulnerabilities associated with age, disability or specific social or economic circumstances in a manner that materially distorts behaviour and causes or is reasonably likely to cause significant harm, where the EU AI Act applies;
  • carry out unlawful social scoring of individuals or groups;
  • evaluate or predict an individual's risk of committing a criminal offence solely on the basis of profiling or assessment of personality traits or characteristics, except where expressly permitted by applicable law in conjunction with objective and verifiable facts;
  • create or expand facial-recognition databases through untargeted scraping of facial images from the internet or CCTV footage;
  • infer emotions of individuals in workplaces or educational institutions where prohibited by law;
  • perform biometric categorisation designed to infer legally protected or sensitive characteristics where prohibited;
  • carry out prohibited real-time remote biometric identification for law-enforcement purposes; or
  • engage in any other AI practice prohibited under applicable law.

8. Restricted high-risk uses

Outlex is not designed or generally offered as a high-risk or consequential automated decision-making system.

Unless Outlex has expressly assessed and enabled a specific use case and all legally required controls are in place, Customers must not use the Services to make, materially influence, recommend or substantially assist decisions concerning individuals in high-risk, consequential or heavily regulated contexts.

This includes:

  • recruitment, screening, hiring or selection of individuals;
  • decisions affecting employment terms, promotion, termination, disciplinary action, task allocation based on personal characteristics or worker monitoring or evaluation;
  • admission, assessment or access to education or vocational training;
  • determining access to credit, lending, insurance, housing or essential private or public services;
  • healthcare eligibility or material medical decisions;
  • law enforcement, criminal justice or predictive policing;
  • migration, asylum, visa or border-control decisions;
  • public-benefit or public-authority eligibility decisions;
  • biometric identification, biometric categorisation or emotion recognition;
  • judicial or administrative decision-making on behalf of a court, tribunal, regulator or public authority;
  • autonomous decisions materially affecting legal rights or obligations; or
  • another use classified as high-risk, consequential or subject to mandatory AI governance controls under applicable law.

A written commercial agreement with Outlex does not waive or replace any statutory requirement applicable to a high-risk use.

Outlex may refuse, suspend, disable or restrict a use case, require additional information or safeguards, or require a separate legal or technical assessment where Outlex reasonably believes that a proposed use may be prohibited, high-risk, consequential, unsafe or outside the intended scope of the Services.

9. Legal workflows and professional review

Outlex may support legal workflows such as contract review, drafting, negotiation support, employment documentation, corporate documentation, legal research support, compliance checklists and internal legal operations where Outlex makes those workflows available.

AI Outputs are AI-generated legal information, drafting, issue spotting or workflow assistance and are not formal legal advice merely because they concern legal matters.

Unless Outlex expressly identifies an AI Output as Human-Reviewed, it has not been reviewed or approved by a qualified Professional.

Customers should obtain appropriately qualified professional review before materially relying on AI Outputs for court or regulator filings, formal legal opinions, litigation strategy, complex or high-value transactions, termination of employment, insolvency, criminal matters, sanctions, immigration, tax, financial regulation or other matters where professional review is legally required or reasonably appropriate.

A jurisdiction-specific template, checklist or AI Output is not automatically filing-ready, execution-ready, complete or appropriate for every factual scenario merely because Outlex makes it available for that jurisdiction.

Human Legal Support is available only where expressly confirmed for the relevant jurisdiction. Availability of a jurisdiction in Lexi does not mean a lawyer or other Professional is available through Outlex in that jurisdiction.

10. Data use and AI model providers

Outlex uses multiple third-party AI model providers and infrastructure providers to deliver AI-enabled functionality.

Depending on the feature, task and technical configuration, Outlex may select and route a Customer request to different model providers. Customers do not necessarily select the underlying provider.

Current AI model providers may include OpenAI, Anthropic and Google Gemini, as further described in the Privacy Policy, DPA and Subprocessor List.

Outlex does not use Customer Data to train third-party foundation models by default. Outlex configures model providers to disable provider-side training or retention where available. Any optional use of Customer Data for training, fine-tuning or product improvement will require appropriate contractual permission, product settings, consent or a separate written agreement where required.

Outlex may use aggregated, anonymised or de-identified information and non-content service telemetry for reliability, security and product improvement where this does not identify Customers or individuals.

AI model providers may process prompts, relevant document content, workspace context or other Customer Data only as reasonably necessary to provide the relevant feature, subject to applicable contractual, privacy and data-processing safeguards.

11. Customer Data and sensitive information

Customers should submit only the data necessary for the relevant task.

Customers should avoid submitting special-category, sensitive, highly confidential, privileged or otherwise regulated information unless it is reasonably necessary for the relevant matter and the Customer has an appropriate legal basis and safeguards for doing so.

This includes special-category data, criminal-offence data, privileged or confidential data, disciplinary information, financial information and trade secrets.

Where Customers submit sensitive information, they are responsible for ensuring that the submission is lawful, necessary and subject to appropriate safeguards.

Customers must not use the Services to infer sensitive characteristics about individuals where doing so is prohibited by applicable law.

12. AI-generated content and external disclosure

Outlex provides the AI-interaction and AI-output transparency required from Outlex under applicable law and as described in this Policy.

Customers may separately have transparency or disclosure obligations when they deploy, publish, distribute or otherwise use AI-generated content in their own activities.

Customers are responsible for determining whether their external use of AI-generated or AI-assisted content requires disclosure to counterparties, employees, consumers, regulators, courts, the public or other persons.

Customers must comply with any disclosure, labelling, professional-sign-off or editorial-review requirement applicable to their use case.

Customers must not intentionally remove, conceal or alter AI labels, provenance information, machine-readable markings or other transparency mechanisms where doing so would be misleading, unlawful or designed to circumvent applicable transparency requirements.

Where a Customer publishes AI-generated text concerning matters of public interest, creates or distributes deepfakes, or uses AI-generated or manipulated audio, image or video content, additional transparency obligations may apply under applicable law.

13. AI literacy and training

Outlex takes proportionate measures to support an appropriate level of AI literacy among personnel and other persons who operate or use AI systems on Outlex's behalf, taking account of their role, technical knowledge, experience and context of use.

Outlex may provide internal training, guidance, documentation, policies, testing procedures or other measures appropriate to relevant roles.

Customers are separately responsible for ensuring that their authorised users have an appropriate understanding of the use and limitations of AI within their own organisation and for meeting any AI-literacy obligations applicable to them as deployers or users of AI systems.

This may include training or guidance on:

  • how outputs are generated;
  • human review;
  • hallucination, error or missing context;
  • confidentiality and data protection;
  • appropriate prompting and document upload practices;
  • escalation to Human Legal Support or external counsel; and
  • prohibited and restricted uses.

14. Outlex safeguards

Outlex aims to maintain reasonable safeguards for AI use, which may include:

  • user-facing AI disclosures and disclaimers;
  • human escalation and Human Legal Support options where expressly available for the relevant jurisdiction;
  • access controls and authentication;
  • data protection and subprocessor controls;
  • contractual and technical controls designed to prevent Customer content from being used to train third-party foundation models by default, consistent with Outlex's published data-use commitments;
  • logging, monitoring and abuse prevention;
  • quality review, testing and product improvement processes;
  • security controls and incident response;
  • AI-interaction transparency controls;
  • AI-output labelling or technical marking measures where applicable;
  • feature controls by jurisdiction where Outlex makes a feature available;
  • prohibited and high-risk use restrictions;
  • testing and evaluation appropriate to material product changes;
  • documented AI-risk and compliance review processes; and
  • AI-literacy measures for relevant Outlex personnel.

These safeguards may evolve as the Services, technology and applicable legal requirements develop.

Outlex will not represent a voluntary safeguard as satisfying a mandatory legal requirement unless Outlex has assessed that requirement and implemented the necessary measures.

15. Integrations and third-party services

Customers may enable integrations with third-party services, such as Google Drive, Microsoft / Microsoft 365, Slack or other supported tools.

When Customers enable integrations, Customer Data may be exchanged between Outlex and the relevant third-party service according to the Customer's configuration, the integration permissions and the third party's terms.

Third-party integrations may themselves contain AI functionality or be subject to separate AI, privacy and acceptable-use terms. Outlex does not control third-party AI functionality that is independently provided by the relevant third party.

Customers are responsible for ensuring that they have the right to connect those services and process the relevant data through Outlex.

16. Monitoring, enforcement and suspension

Outlex may use appropriate technical and operational measures to monitor security, abuse signals, policy violations, service integrity and reliability, subject to applicable privacy and data protection law.

If Outlex reasonably believes that a Customer or user has violated this Policy, Outlex may take appropriate action, including:

  • requesting clarification or additional information;
  • requiring the Customer to stop or modify the relevant use;
  • disabling features or integrations;
  • removing or restricting access to content;
  • restricting specific jurisdictions, workflows, models, integrations or features without suspending the Customer's entire account where a narrower intervention is reasonably sufficient;
  • suspending or terminating access to the Services;
  • notifying relevant authorities where legally required.

Where reasonably practicable and lawful, Outlex will seek to apply proportionate enforcement measures considering the nature, severity, frequency and potential impact of the relevant violation.

17. Regional application

EU/EEA. Where the EU AI Act applies, Outlex applies this Policy subject to the obligations applicable to Outlex's role and the relevant AI functionality. Outlex prohibits use of the Services for AI practices prohibited by Article 5 of the EU AI Act. Outlex applies transparency measures for direct human interaction with Lexi and, where applicable, measures relating to identification or detectability of AI-generated content. Customers that use Outlex as deployers of AI systems remain responsible for legal obligations applicable to their own deployment, including human oversight, transparency, AI literacy, impact or risk assessment and high-risk-system requirements where applicable.

United Kingdom. Outlex applies the same core transparency, responsible-use and human-oversight principles to UK Customers. UK Customers must use the Services consistently with applicable UK data protection, equality, consumer-protection, professional-services, employment and sector-specific laws and regulatory requirements. Lexi is identified as an AI system and does not represent itself as a human lawyer, solicitor or other regulated professional. Human Legal Support for UK matters is available only where expressly confirmed by Outlex and provided under an appropriate arrangement with a suitably qualified Professional.

United States. AI regulation in the United States may apply at federal, state or local level and may vary depending on the relevant functionality, Customer, individual and use case. Outlex applies the global transparency, prohibited-use, human-oversight and restricted-high-risk-use principles in this Policy to US use of the Services. Customers must not use Outlex to make or materially influence consequential decisions concerning individuals where applicable law requires disclosures, assessments, anti-discrimination safeguards, human review, appeal rights or other controls unless those requirements have been independently assessed and satisfied. Lexi must not be represented as a licensed attorney, law firm or equivalent to services provided by a licensed professional. Human Legal Support for US law matters is not currently available unless expressly confirmed by Outlex and provided through an attorney or law firm appropriately licensed or authorised for the relevant jurisdiction and matter. State-specific requirements may supplement this Policy.

18. Updates to this Policy

Outlex may update this Policy from time to time to reflect changes in the Services, AI features, legal requirements, regulatory guidance, provider configuration or risk controls.

Outlex may also update this Policy to reflect changes to applicable AI laws, regulatory guidance, model providers or the regulatory classification of particular features.

Where changes are material, Outlex may provide notice through the website, in-product notice, email or another reasonable method. Where reasonably appropriate, Outlex will provide notice before a material change takes effect.

19. Contact

For questions about this Policy or Outlex's use of AI, contact:

privacy@outlex.ai