Photo by Phong Thanh from Pexels

Nokia is stepping up its game in the AI world, joining the AI Pact—a new initiative designed to help companies prepare for the European Union’s AI Act. The Act, which came into effect in August 2024, sets out rules for how AI systems can be used based on their level of risk. With major compliance requirements kicking in over the next three years, Nokia’s early involvement shows it’s ready to tackle the challenge head-on.

What is the AI Pact?

The AI Pact is a voluntary initiative launched by the European Commission to encourage organizations to anticipate and implement the AI Act’s requirements ahead of the mandated deadlines.

Officially effective as of August 1, 2024, the AI Act introduces comprehensive regulations on the use of AI technologies, particularly those categorized as high-risk. While some of the Act’s provisions are already enforceable, the complete set of requirements will only come into force after a transitional period, expected to end within the next few years.

To support companies in navigating these upcoming changes, the AI Pact seeks voluntary commitments from businesses, urging them to align their operations with the new standards before they become legally binding. The Pact was first introduced through a call for interest in November 2023, drawing responses from over 550 organizations across various sectors, including tech giants, SMEs, and startups.

Its structure revolves around two main pillars designed to guide participants through the complexities of AI governance: Pillar I, which fosters a community for sharing best practices, and Pillar II, which encourages proactive compliance.

A Deeper Look into the Two Pillars

Pillar I serves as the gateway for companies to engage in the AI Pact network, offering a collaborative environment where participants can exchange insights, experiences, and strategies on implementing the AI Act. Through workshops and online platforms, the AI Office provides organizations with guidance on their responsibilities under the AI Act, enabling them to understand and prepare for the upcoming compliance requirements.

Participants are encouraged to share internal policies, successful compliance strategies, and even challenges faced along the way, helping to build a robust knowledge base that benefits all members of the network.

Pillar II, on the other hand, is focused on facilitating early compliance by encouraging companies to publicly commit to specific actions that align with the AI Act. These commitments, known as ‘declarations of engagement,’ outline the concrete steps companies are taking—or plan to take—to meet the Act’s requirements.

These pledges include timelines for implementation and may involve incremental objectives, allowing companies to gradually integrate AI governance practices.

Setting the Standard in AI Governance

Nokia’s research powerhouse, Nokia Bell Labs, has been at the forefront of these AI innovations, focusing on Responsible AI long before the regulations took shape. With six guiding principles—fairness, reliability, privacy, transparency, sustainability, and accountability—Nokia Bell Labs continues to influence the company’s AI strategies.

As Nokia gears up to comply with the new rules, it’s not just playing catch-up; it’s setting the pace for AI governance in the tech industry. This move cements Nokia’s role as a leader in building AI solutions that are not just cutting-edge but also ethical and sustainable.

Pledges and the Path Forward

On September 25, 2024, the European Commission celebrated the first wave of signatories to the AI Pact, including over 100 companies such as IBM, Microsoft, and SAP—together with Nokia. These companies pledged to take early action on three core commitments: adopting an AI governance strategy, identifying high-risk AI systems, and promoting AI awareness among their workforce.

Additionally, companies are encouraged to make further tailored commitments, such as ensuring human oversight of AI systems, mitigating risks, and transparently labeling AI-generated content like deepfakes.

The AI Pact’s pledges are designed to be flexible, allowing companies to shape their commitments based on their specific needs and operations. These pledges are not legally binding but serve as a framework for companies to demonstrate their commitment to ethical AI practices. Organizations are invited to report on their progress within 12 months of their pledge, offering transparency and accountability in their journey toward full compliance with the AI Act.