Responsible AI Policy

Last Updated: 2026-10-11

1. Purpose & Introduction

1.1 VNEXT Technologies DWC-LLC (the “ Company ”, “ we ”, “ us ”, or “ our ”) develops and provides PxM, an AI-powered strategy and performance management platform, built natively on the Microsoft Power Platform and integrated with Microsoft 365, Dynamics 365, and Microsoft Copilot (the “ Platform ”), designed to help Strategy Offices, Enterprise Project Management Offices, and Performance leaders connect strategic planning, execution, and performance measurement.

1.2 This Responsible AI Policy (“ Policy ”) sets out the principles, governance structures, and operational safeguards that govern the design, development, deployment, and ongoing operation of artificial intelligence features within the Platform (“ AI Features ”). It reflects our commitment to developing and deploying AI responsibly, ethically, and in a manner that earns and preserves the trust of our customers, their personnel, and other stakeholders, and to complying with applicable AI-specific regulation, including the EU Artificial Intelligence Act (Regulation (EU) 2024/1689, the “ EU AI Act ”), as amended by the Digital Omnibus on AI (Regulation (EU) 2026/1744) (the “ AI Omnibus ”).

1.3 This Policy is intended for a broad audience, including customers evaluating the Company during procurement, existing customers and their governance, risk, and compliance functions, regulators, and internal personnel involved in the design, development, or oversight of AI Features.

1.4 This Policy operates alongside, and should be read together with, our Data Protection Policy, our Data Processing Agreement (“ DPA ”), and the separate contractual terms entered into between the customer and the Company for the use of the Platform, each of which addresses related matters including data processing, security, and contractual commitments regarding the Platform.

2. Scope

2.1 This Policy applies to all AI Features embedded within or accessible through the Platform, including features that generate strategic recommendations, summarize performance data, draft or refine objectives and key results, surface insights or anomalies within execution or performance data, and any feature that integrates with or invokes Microsoft Copilot or other generative AI capabilities within the Microsoft ecosystem.

2.2 This Policy applies to the Company’s own development and configuration of AI Features. It does not govern the internal AI governance policies of Microsoft Corporation in respect of the underlying Microsoft Copilot, Azure AI, or other foundation model services on which certain AI Features rely; those services are governed by Microsoft’s own responsible AI commitments, as referenced in section 9 of this Policy.

2.3 This Policy does not itself create binding legal obligations between the Company and any customer. Contractual commitments regarding AI Features are set out in separate contractual terms and DPA. This Policy describes the principles and practices that inform those commitments.

3. Our AI Principles

We design and operate AI Features in accordance with the following core principles, which guide our decision-making throughout the AI lifecycle, from design through deployment and ongoing monitoring.

Principle	What It Means in Practice
Human Oversight and Agency	AI Features are designed to support and inform human decision-making, not to replace it. Material business decisions remain with the customer’s personnel, as further described in section 5.
Transparency and Explainability	We make clear when a customer’s users are interacting with an AI-generated output, mark such outputs consistently with applicable law, and provide sufficient context for users to understand, at a general level, how such outputs are produced, as further described in section 7.
Fairness and Non-Discrimination	We take reasonable steps to identify and mitigate bias in AI Features that could result in unfair or discriminatory outcomes, particularly where AI Features touch upon performance evaluation or personnel-related content.
Privacy and Data Protection	AI Features are designed and operated consistently with our Data Protection Policy and the DPA, including our commitment not to use Customer Data to train underlying foundation models without consent.
Safety, Security, and Robustness	AI Features are subject to security controls, testing, and monitoring designed to reduce the risk of misuse, manipulation, or unreliable output.
Accountability	We maintain internal governance structures and clear ownership for the responsible development and operation of AI Features, as described in section 12.
Reliability and Continuous Improvement	We monitor the performance of AI Features on an ongoing basis and take corrective action where issues are identified.

4. AI Features Within the Platform

4.1 The Platform incorporates AI Features designed to support strategic planning, execution tracking, and performance measurement workflows, which may include: natural-language drafting assistance for strategic objectives and key results; automated summarization of performance data, reports, and dashboards; identification of trends, anomalies, or risks within execution and performance data; conversational or copilot-style interfaces enabling users to query Platform data using natural language; recommendations intended to support prioritization or resource allocation decisions; and robotic process automation (“ RPA ”) and workflow automation features that execute defined actions within the Platform, such as routing approvals, updating record status, or flagging initiatives for review, subject to the human confirmation requirements set out in section 5.

4.2 AI Features may be powered by a combination of the Company’s own models and logic and underlying Microsoft AI services, including Microsoft Copilot and Azure AI services, invoked within the security and compliance boundary of the customer's own Microsoft tenant, consistent with the architecture described in our Data Protection Policy.

4.3 The specific AI Features available to a customer may vary depending on the subscription plan purchased and the customer’s own Microsoft licensing and tenant configuration.

5. Human Oversight and Human-in-the-Loop Design

5.1 AI Features are designed as decision-support tools, not as automated decision-making systems that produce legal or similarly significant effects on individuals without human involvement. AI Features that produce content or output for human review, such as drafted text, summaries, or recommendations, are designed to require an affirmative action by a human user (such as review, editing, or acceptance) before such content is finalized or acted upon.

5.2 Separately, and with a higher standard of oversight, no AI Feature or RPA/workflow automation feature within the Platform may autonomously execute a consequential portfolio action (including the exclusion, reallocation, deprioritization, or cancellation of an initiative, project, or resource assignment) without an explicit, affirmative confirmation from an authorized human user prior to that action taking effect. Where an AI Feature identifies an initiative as a candidate for exclusion or a resource as a candidate for reallocation, the Platform is designed to present that determination to the user as a recommendation requiring approval, not as a completed action, and the Platform maintains a record of the human user who approved each such action.

This distinction between AI-generated content awaiting review and AI-recommended actions awaiting execution approval, applies uniformly across all AI Features and automation capabilities within the Platform.

5.3 We have not designed any AI Features to autonomously make personnel decisions, such as decisions relating to hiring, promotion, compensation, or termination, and we advise customers against configuring or using the Platform for such purposes without independent human review and, where appropriate, their own legal and human resources guidance.

5.4 Customers remain responsible for establishing internal governance and review processes appropriate to their own use of AI Features, including determining the extent to which AI-generated outputs are relied upon in business decision-making, as further described in section 11.

6. Data Governance for AI Features

6.1 The Company does not use Customer Data to train, fine-tune, or otherwise improve any underlying foundation model, including Microsoft Copilot or any other third-party AI model, except with the prior written consent of the relevant customer. This commitment is reflected contractually in the DPA.

6.2 AI Features are designed to operate on the minimum Customer Data reasonably necessary to generate the relevant output, consistent with the data minimization principles described in our Data Protection Policy.

6.3 Where a user submits a prompt or query to an AI Feature, the content of that prompt and any associated output may be logged for service delivery, troubleshooting, abuse prevention, and service improvement purposes, subject to the retention limitations and security measures described in the Data Protection Policy and DPA. Such logs are treated as Customer Data and are not used to train foundation models absent consent, consistent with section 6.1.

6.4 AI Features process Customer Data consistently with the data residency commitments described in our Data Protection Policy, operating within the region associated with the customer’s Microsoft tenant to the extent supported by the underlying Microsoft infrastructure.

6.5 Customers should not submit Special Category Data (as defined in our Data Protection Policy) to AI Features other than where strictly necessary and lawful, and should notify the Company in advance where such data may be submitted.

7. Transparency, Explainability, and AI-Content Marking

7.1 The Platform indicates to users, within the product interface, when content, recommendations, or insights have been generated or materially assisted by an AI Feature, to support informed reliance on such content.

7.2 Where an AI Feature generates synthetic text, image, audio, or video content, the Company marks such output in a machine-readable format designed to enable it to be detected as artificially generated, consistent with Article 50(2) of the EU AI Act. This obligation applies independently of, and in addition to, the user-facing labeling described in section 7.1, and is treated by the Company as a present, applicable engineering requirement.

7.3 For AI Features that generate recommendations or flag anomalies, we aim to provide users with contextual information regarding the general basis for an output, such as the underlying data points or metrics considered, to the extent consistent with the underlying model architecture and without compromising the performance or security of the Platform.

7.4 We maintain user-facing documentation describing the general functionality and intended use of AI Features, made available to customers through the Platform’s Documentation.

7.5 Customers acknowledge that certain underlying AI and machine learning models, including large language models such as those underlying Microsoft Copilot, are inherently complex, and that full technical explainability of every individual output may not be possible. We are nonetheless committed to providing meaningful, general-level transparency regarding how AI Features operate, and to maintaining the machine-readable marking commitments described in section 7.2

8. Fairness and Bias Mitigation

8.1 We take reasonable steps, appropriate to the nature and risk of each AI Feature, to identify and mitigate the risk that AI Features produce biased, unfair, or discriminatory outputs, including where AI Features process data that may reference or relate to individual employees, such as performance commentary or ownership of objectives.

8.2 Where AI Features are used to summarize or evaluate performance-related content, we design such features to support balanced, evidence-based summarization rather than to independently assign performance ratings or rankings, and we encourage customers to apply their own internal review processes before relying on such summaries for personnel-related purposes.

8.3 We periodically review the design and, where feasible, the outputs of AI Features for indications of systemic bias, and we take corrective action where material issues are identified, consistent with our monitoring practices described in section 10.

9. Safety, Security, and Third-Party AI Governance

9.1 AI Features are subject to the same technical and organizational security measures applicable to the Platform generally, including access controls, encryption, and logging and monitoring of access to Customer Data processed by AI Features.

9.2 Where AI Features rely on Microsoft Copilot or other Microsoft AI services, such services are subject to Microsoft’s own responsible AI principles, safety mitigations, content filtering, and data handling commitments applicable to its commercial cloud and Copilot offerings. We select and configure Microsoft AI services with regard to Microsoft’s published responsible AI and enterprise data protection commitments for Copilot and Azure AI services.

9.3 Where technically supported by the underlying AI services, we apply reasonable content filtering and safety guardrails designed to reduce the risk that AI Features generate harmful, offensive, or clearly inaccurate content, and we test AI Features for robustness against common categories of misuse, such as prompt injection, prior to release.

9.4 New or materially modified AI Features are subject to internal testing and review prior to general release, proportionate to the risk profile of the feature, including functional testing, security review, and, where applicable, review for potential bias or fairness concerns.

10. Risk Management, Monitoring, and Evaluation

10.1 We assess AI Features on a risk-based approach, applying a greater degree of scrutiny, testing, and human oversight design to AI Features that carry a higher risk of harm, such as those touching on performance evaluation or personnel-related content, relative to lower-risk features such as general content summarization. This approach is designed to align with the risk-based structure of the EU AI Act.

10.2 For higher-risk AI Features, we conduct an internal assessment prior to release considering the intended use and users of the feature, the categories of data processed, the potential for bias or discriminatory impact, the degree of human oversight built into the feature, whether the feature may fall within Annex III of the EU AI Act, and applicable regulatory considerations.

10.3 We monitor the operation of AI Features on an ongoing basis, including monitoring for performance degradation, unexpected outputs, security incidents, and customer-reported issues, and we maintain internal channels for personnel and customers to escalate concerns regarding AI Feature behavior.

10.4 Where monitoring or feedback identifies a material issue with an AI Feature, including bias, inaccuracy, or unsafe behavior, we will take corrective action proportionate to the risk identified, which may include adjusting the feature's configuration, adding safeguards, restricting availability of the feature, or, where necessary, disabling the feature pending remediation.

11. Prohibited and Restricted Uses

Customers and their Authorized Users must not use AI Features within the Platform to:

• Make or substantially automate decisions regarding hiring, promotion, compensation, discipline, or termination of any individual, without independent human review and compliance with applicable employment law.

• Generate content intended to discriminate against, harass, or unlawfully profile any individual or group on the basis of a protected characteristic.

• Attempt to identify, re-identify, or infer sensitive information about an individual that was not knowingly and lawfully provided to the Platform.

• Circumvent, probe, or attempt to manipulate the safety guardrails, content filters, or intended operation of any AI Feature (including prompt injection or attempts to extract system prompts or underlying model configuration).

• Circumvent, disable, or strip the machine-readable marking applied to AI-generated content described in section 7.2.

• Generate any content prohibited under the EU AI Act or other applicable law.

• Rely on AI-generated output as the sole basis for a material business, financial, legal, or personnel decision without independent human review.

• Use AI Features to generate content that infringes the intellectual property, privacy, or other rights of any third party.

These restrictions supplement, and should be read together with, the Acceptable Use clause set out in our Terms of Use and any acceptable use provisions of the separate contractual terms entered into between the customer and the Company. Violation of this section 11 may result in suspension of access to AI Features or the Platform.

12. Governance and Accountability

12.1 The Company maintains internal ownership and accountability for responsible AI practices, including a designated function or individual responsible for overseeing compliance with this Policy.

12.2 Material decisions regarding the introduction of new AI Features, or material changes to existing AI Features, are subject to review by relevant internal stakeholders, which may include product, engineering, security, legal, and data protection functions, proportionate to the risk profile of the feature.

13. Regulatory Alignment

13.1 We monitor the evolving regulatory landscape applicable to artificial intelligence in the jurisdictions in which we and our customers operate, and take reasonable steps to align our practices with such requirements as they become applicable to the Platform, including without limitation, the EU AI Act (Regulation (EU) 2024/1689), as amended by the Digital Omnibus on AI (Regulation (EU) 2026/1744.

13.2 Independently of the EU AI Act, Article 22 of the GDPR restricts decisions based solely on automated processing, including profiling, that produce legal or similarly significant effects. Our human oversight design, described in section 5 of this Policy, is intended to support customer compliance with this requirement.

13.3 We similarly monitor relevant guidance issued by data protection authorities regarding automated decision-making and profiling under the UK GDPR, and applicable UAE regulatory guidance on the use of AI, and take reasonable steps to align our practices accordingly.

13.4 Where a customer’s use of the Platform is subject to sector-specific or jurisdiction-specific AI regulation, including potential classification of a particular use case as high-risk under Annex III of the EU AI Act, the customer is responsible for assessing its own compliance obligations arising from its use of AI Features, and the Company may provide reasonable cooperation and available documentation to support such assessments.

13.5 Given the pace of change in AI-specific regulation, the Company treats this section 13 as subject to more frequent review than the remainder of this Policy, and will update it as further implementing measures, harmonized standards, or guidance are issued under the applicable laws.

14. Training and Awareness

14.1 Personnel involved in the design, development, deployment, or oversight of AI Features receive training appropriate to their role regarding the principles set out in this Policy, applicable data protection and AI-specific regulatory requirements, and the identification and escalation of AI-related risks.

14.2 We provide customers with reasonable documentation and guidance regarding the intended use, capabilities, and limitations of AI Features to support customers' own internal training and change management efforts.

15. Incident Response for AI-Related Issues

15.1 Where the Company identifies, or is notified of, a material issue involving an AI Feature, including a security incident, a material bias or fairness concern, or output that is materially inaccurate or unsafe, the Company will investigate the issue and take corrective action proportionate to the risk presented, consistent with its general incident response and Personal Data Breach notification obligations described in the Data Protection Policy and DPA.

15.2 Customers may report concerns regarding the behavior of AI Features through the Company’s designated support channels, and, where the concern relates to a Personal Data Breach, through the escalation procedures set out in the DPA.

16. Changes to This Policy

We may update this Policy from time to time to reflect changes in our AI Features, changes in applicable law or regulatory guidance (including further implementing measures under the EU AI Act and the AI Omnibus), or evolving industry best practice regarding responsible AI. Material changes affecting customers will be communicated in accordance with the notice provisions of the applicable contractual terms. The “Last Updated” date at the top of this Policy indicates when it was most recently revised.

17. Contact Information

17.1 Questions, feedback, or concerns regarding this Policy or the Company's AI Features may be directed as follows:

VNEXT Technologies DWC-LLC

Email : contact.pxm@vn3xt.com

17.2 Customers seeking further detail regarding the technical operation of specific AI Features should contact their designated account or support representative in accordance with the separate contractual terms they entered into with the Company.