LTRT Responsible AI Use Policy

LTRT recognizes the transformative power of artificial intelligence (AI). To realize its potential without compromising quality, data sovereignty, and our reputation, AI must be paired with our expertise and commitment to ethical and responsible use.

AI is not a neutral technology. Its impact depends on how it is designed, deployed, and applied, and misuse can have lasting consequences. LTRT’s approach is therefore cautious, ensuring that AI strengthens rather than undermines our expertise, values, and reputation.

This policy establishes Left Turn Right Turn’s AI principles, presents a framework for the use of AI, and discusses data sovereignty and security considerations. It gives employees the tools and know-how needed to apply AI tools in alignment with company values, legal standards, and the highest levels of data security to protect our clients.

2. AI Overview

Artificial intelligence (AI) is “technology that enables computers and machines to simulate human intelligence and problem-solving capabilities.” In this policy, the term ‘AI’ will be used to describe any artificial intelligence software with broad capabilities that surpass narrow-AI like predictive text and spellcheck. Since 2022, AI tools, applications, and integrations have proliferated, becoming a widely used and disruptive technology in professional and academic contexts. Further permeation of AI tools and workflows is expected to continue over the coming years, as companies increasingly integrate AI into their products. These changes promise efficiency and productivity gains, but they also introduce threats. 

Use of AI is not neutral. Internal processes of how AI technologies operate are opaque, and outputs may contain hidden biases, inaccuracies, unsubstantiated/unsupported claims, and/or ethical risks. Concerns range from data privacy and the erosion of human expertise to broader questions about fairness, originality, and accountability. AI also introduces environmental concerns due to the high energy-use and water-consumption needs of the software. Furthermore, datasets used to develop and train most large language models included, without proper consent, copyrighted text, art, data, and other media. Datasets also include the biases of human-generated material, absorbing assumptions and prejudices . Taken together, these threats warrant a measured approach when LTRT staff use of AI tools. 

This policy aims to provide LTRT Staff with the support they need to responsibly use AI in their work. 

3. AI Principles

LTRT has identified the following five principles that define how we will use AI responsibly in our work.

• AI is a tool: The use of AI is meant to enhance, not replace, the stellar quality of work our people produce. 

• People are our strength: The quality of our work depends on our people. AI use should never dilute our expertise. 

• Personal responsibility: We are always responsible for the quality of our work, whether it is created by us independently , or supported by AI. When using AI tools, we must validate accuracy, quality, and flow. 

• Secure use: AI use must always align with LTRT's IT Security and Privacy policies, including data ownership and data sovereignty. These policies can be found here: 

LTRT Garage - Documents\98Internal\11Policies

• Alignment with values: The use of AI must reflect LTRT's values, commitments and ethics, including professional integrity, accessibility and environmental sustainability commitments.

• With purpose and intention: AI should be used for a purpose – either to improve the quality of our research/work, automate repetitive tasks, or improve personal efficiency. The latter should include the effort to validate the accuracy of the outputs. We don’t want our skills and smarts to deteriorate, so we should use AI with specific intentions in mind.

4. AI-Use Requirements

This section details the requirements for use of AI in our work at LTRT. 

What AI tools can be used?

For any generative purposes, only Microsoft 365 Copilot may be used. “Generative” purposes means using AI to support in development of content, copy editing, translation, querying, or analyses. Microsoft 365 Copilot, within our IT system, is the only permitted tool for these purposes. 

What can we do with Microsoft 365 Copilot?

When using Microsoft 365 Copilot, the user must be transparent about its use with the person that assigned them the work. This includes a) what content it helped to develop, and b) how it was used (e.g., “I used it to develop preliminary content that I then modified”, or “I used it to proof read and simplify the language”, etc.). This responsibility is the same as if someone had delegated a task to a colleague or affiliate. 

A separate document called Permitted AI Use Cases is a compendium provided to staff, and serves as a living document that will be updated periodically. The aim of this document is to provide clear, consistent, and comprehensive guidance on how Microsoft 365 Copilot can be used in our work. 

For further information about how Microsoft 365 Copilot can be used to support LTRT work, please see the Responsible AI Use Policy – Sample Use Case Compendium. 

What AI tools are not permitted?

At this point, other generative AI tools are not permitted. AI tools embedded in applications like Quickbooks and Expensify (e.g., receipt scanning), or defaults in search engines (e.g., Google's AI generated summaries) may be used, but results must always be validated in accordance with the core principles.

What about client-facing work?

If, at any point, a document that includes content created with Microsoft 365 Copilot is going to a client, the writer must ensure that it is sufficiently modified to use the cohesive voice of the project team rather than the voice of AI. In short, any AI-generated content used in a document should be indistinguishable from our own content. While AI Detection systems are not perfect, if a client or a member of the public were to use one and it comes back with high probability of AI Use, it could result in significant reputational impacts, whether we actually used it significantly or not. 

The only exception is if the client provided explicit permission to use Microsoft 365 Copilot in the development of content or deliverables. However, the writer must clearly and plainly state that AI was used to support the development of that content. 

Regardless, a disclaimer should be used to indicate that AI was used in the report or slide deck. (see following section)

5. AI Use Acknowledgement

The follow sample acknowledgement/disclaimer should be used at the end of a document using AI.

This work was completed by Left Turn Right Turn with the assistance of Microsoft 365 Copilot AI. We are committed to the responsible and ethical use of AI in our work, and enterprise data protection is in effect. You can learn more about our Responsible AI Use Policy by visiting our website or by contacting us at info@ltrt.ca. 

6. Expectations for Affiliates and Vendors

Any affiliates, vendors, or contractors working with LTRT are expected to follow the same standards outlined in this framework. Our partners must not use AI in ways that undermine our ethical commitments, data privacy standards, or the quality of services we deliver to clients. Affiliates and Vendors who do not have a license for Microsoft 365 Copilot that includes the same level of security and privacy protections will not be permitted to use the tool. 

When entering into a new project with Affiliates and Vendors the LTRT Project Manager should inquire with the Affiliate or Vendor Project Manager about their approach and policies concerning AI. 

7. Continuous Review

AI is evolving rapidly, and LTRT’s policies must evolve alongside it. The Responsible AI-Use Policy and the Permitted AI Use Cases document are therefore considered living documents. They will undergo review regularly to ensure they remain aligned with best practices, emerging regulations, and the needs of our clients and staff. Interim updates may also be issued in response to significant changes in technology, risks, or opportunities; if the either document is updated Staff will be notified through an announcement on Microsoft Teams. 

Staff are encouraged to provide feedback and raise concerns about unintended consequences, ethical dilemmas, or new ideas for responsible AI use. By treating this policy as an evolving framework, LTRT ensures that our approach to AI remains both cautious and adaptive, protecting our values while allowing us to explore innovation in ways that add real value to our work.