Document name | Tele2 AI principles and governance |
Policy holder | EVP Corporate Affairs (Torkel Sigurd) |
Other interest holders | Head of AI & Automation, Head of Legal |
Policy last revised (date) | April, 2024 |
Version | 1.0 |
Policy approved by (name / date) | Group Leadership Team |
Policy valid as of | April, 2024 |
Policy valid until | Next revision, latest April 2027 |
TELE2 AI PRINCIPLES &
GOVERNANCE
Ensuring responsible AI within Tele2
Tele2's purpose is to enable a society of unlimited possibilities. We play an important role in realizing the fundamental ideas of our time - as a part of the societal infrastructure, as communication provider and digital integrator. We have been breaking new grounds for 30 years and will continue to do so, since the challenger mindset is part of our heritage.
Artificial Intelligence (AI) has the potential to help us on this journey. However, with all possibilities associated with AI also comes a responsibility. The use of AI faces us with new challenges and ethical concerns.
We are committed to conducting our business in a responsible manner in all areas and we have a vision to lead in sustainability . This is why we have developed the AI principles and governance model set out in this document to ensure that we mitigate any negative impacts this technology may have on society or the environment. Tele2's customers, shareholders and other stakeholders should be confident that how we do business does not only comply with legal requirements, but also meets the highest standards of ethics and integrity.
Information Classification: Public
Version Table | |||
Revision | Date | Prepared and approved by | Information |
1.0 | April, 2024 | Helena Wanhainen (Head of Legal), Henrik | Original version |
Atteryd (Head of AI & Automation) |
Information Classification: Public
Objective and Scope
This document sets out the AI principles that Tele2 is committed to following when designing, developing or using AI. The document also sets out the governance model we apply to ensure that the AI principles are followed in practice.
Information Classification: Public
The Tele2 AI principles
Accountability
The human always remains responsible. Our solutions come with a clear definition of who is responsible for which AI system or feature. We are in charge of our products and services. And, we know who is in charge for partner or third party solutions.
Purpose
The use of AI shall always support Tele2's purpose, our business goals, and customers.
Fairness
AI systems should be designed, implemented and used in a way that avoids discrimination based on sex, transgender identity or expression, ethnicity, religion or other belief, disability, sexual orientation or age. Bias or unfair treatment based on socioeconomic status or other personal conditions should also be avoided. We will strive to identify and remedy discrimination or unintended bias in the data, design and implementation of AI systems, minimizing the likelihood that our use of AI reinforces bias and discrimination in society.
Privacy and security
AI systems should protect the individual's right to privacy and ensure security of data, while being compliant with applicable data protection laws and industry standards. Our AI systems follow the general Tele2 privacy and security policies for data collection, storage and use, as well as GDPR and relevant EU and local legislation and we shall always ensure the integrity and availability of AI systems with the use of relevant and up-to date security measures. We apply a risk-based thinking aimed at identifying, assessing, and managing risks to proactively address potential issues.
Transparency
We ensure transparency in relation to customers, employees and other stakeholders. The decision-making processes and algorithms used in AI systems should be transparent and explainable. When people directly interact with an AI system, it should be clear to the users that this is the case. When AI systems take or support decisions, an adequate level of understanding of how the conclusions are arrived at needs to be ensured.
Reliability and safety
AI technologies should be developed and deployed in a manner that ensures their reliability, consistency, quality, robustness, and safety during the entire lifecycle, under normal and unexpected conditions. Systems should resist harmful manipulation. We can stop and deactivate AI-systems at any time (kill switch).
Sustainability
We are mindful of our AI systems' environmental footprint, and we strive for optimal dataset(s) selection, frequency of updates, storage location, and algorithm choice for the systems to not only succeed but to do so in the most environmentally friendly way. Our AI solutions benefit our customers and society at large, contributing to Tele2's wider ESG initiatives and policies.
Information Classification: Public
AI Governance within Tele2
To ensure that the above fundamental principles are being considered when developing or interacting with AI, Tele2 has put a governance framework in place.
The governance framework entails, inter alia, more detailed rules and guidelines for the use of AI in different areas within Tele2 based on the guiding principles. To ensure that such rules and guidelines are effectively implemented and used by the organisation, the policy part of the framework is supported by three important "pillars" - or operating processes and implementation tools - namely: 1. Documentation & ownership, 2. Tools & processes and 3. Awareness.
Documentation & ownership
Our aim is that all AI systems used or developed within Tele2 should be documented and have a designated owner.
Tools & processes
Tele2 has put an organisational model in place that provides clear escalation channels for employees working on AI to raise concerns about ethical dilemmas. Our aim is to first and foremost provide relevant tools and processes for the people closest to an AI system to make sure that they themselves can make the ethical and compliance considerations necessary to comply with the governance framework and document their decisions. However, since ethical and compliance impacts of AI can be complex to determine and require multifunctional perspectives, Tele2 has also created an AI Ethics & Compliance Committee to deal with more complex and high-risk cases. The Committee is also responsible for updating the detailed rules and guidelines. The Committee consists of representatives from different units and functions and includes a variety of expertise representing inter alia technical, commercial, legal and security perspectives. Any high-risk or business critical decisions are escalated to executive level.
Awareness
We will engage in AI and ethics training and knowledge sharing to ensure awareness of the importance of ethics in developing and deploying AI across the organisation.
Information Classification: Public
Attachments
- Original Link
- Original Document
- Permalink
Disclaimer
TELE2 AB published this content on 22 April 2024 and is solely responsible for the information contained therein. Distributed by Public, unedited and unaltered, on 22 April 2024 13:29:02 UTC.