IRIA News

6 minute read

The European Commission announced the first 100 signatories of its EU Artificial Intelligence (AI) Pact, marking a key step toward ensuring compliance with the forthcoming AI Act. The pact is a voluntary initiative aimed at encouraging companies to proactively apply the principles of the AI Act before its full enforcement.

Several notable AI technologies-related firms, including Google, OpenAI, and Microsoft, have come on board with the EU’s AI Pact. However, some of the leading tech giants such as Meta, Apple, and Mistral, abstained from signing the pact raising concerns about its effectiveness and its role in shaping the future of AI regulation in Europe.

The AI Pact was introduced as a way for companies to voluntarily pledge adherence to certain principles that align with the EU’s AI Act, which entered into force on August 1, 2024. While full compliance with the Act is not mandatory until 2026, the Commission is urging companies to begin preparing now to ease the transition.

The signatories of the AI Pact commit to three core actions:

• Developing an AI governance strategy to promote responsible AI usage within their organizations.
• Mapping AI systems that are likely to be classified as “high-risk” under the AI Act.
• Promoting AI literacy among staff to ensure ethical AI development.

In addition to these core commitments, more than half of the signatories have voluntarily pledged to take further actions, including ensuring human oversight of AI systems, mitigating potential risks, and clearly labeling AI-generated content, such as deepfakes. These measures are designed to foster transparency, trust, and ethical practices in AI development as Europe moves toward full enforcement of the AI Act.

The European Commission views the AI Pact as a preparatory step for industries, creating a foundation of collaboration and compliance before the Act’s full implementation. As Thomas Regnier, a spokesperson for the Commission, emphasized, “Artificial intelligence offers many benefits to all of us, but it also carries risks. We encourage all companies to join this initiative.”

The AI Pact also complements broader EU efforts to position itself as a leader in AI innovation. The Commission has launched various initiatives, including the AI Factories, which are designed to provide start-ups and industry players with access to vital resources such as data, computing power, and talent. The goal is to accelerate AI development in key sectors like healthcare, automotive, defense, and agriculture.

Despite the EU’s efforts, the absence of major AI companies from the pact has sparked debate about the overall efficacy of the initiative. While Google, Microsoft, and OpenAI (creator of ChatGPT) are notable signatories, the exclusion of other key players raises questions about whether the pact can truly drive change across the entire AI ecosystem.

Critics argue that the voluntary nature of the pact may limit its impact. Without the involvement of industry giants like Meta and Apple, whether the pact can effectively set the standard for AI governance and regulation in Europe remains uncertain.

The European Commission, however, remains optimistic. As Thomas Regnier pointed out, participation in the pact is voluntary but essential for planning the future of AI in Europe. “Signatories number more than a hundred, with Google and Microsoft among them,” he said, emphasizing the significance of these companies’ involvement.

Regnier also highlighted the importance of collaboration in the early stages of AI regulation, stating that the AI Pact provides an opportunity for companies to share best practices and align with future legal requirements.

Challenges for AI regulations

The AI Act itself represents a pioneering effort by the EU to establish a comprehensive regulatory framework for AI. With provisions already in force and more becoming applicable over the next two years, the Act sets out specific rules for high-risk AI systems, general-purpose AI models, and AI systems embedded in regulated products.

The signatories of the AI Pact have pledged to meet these regulatory standards ahead of the legal deadline, signaling a proactive approach to compliance. However, as the Act’s full implementation draws closer, the question remains whether other major AI companies will join the pact or continue to operate outside its voluntary framework.

The Commission hopes that by offering companies a chance to prepare early, the transition to the AI Act’s requirements will be smoother. Those who have already signed the pact are expected to have a head start in complying with the EU’s regulatory framework, potentially avoiding some of the challenges that may arise for companies that delay preparation.

List of companies that signed the pact