Key Takeaways
- The AI Pact focuses on ethical AI development, AI literacy, and identifying high-risk systems.
- 100+ companies, including Microsoft, Google, and OpenAI, have signed the EU AI Pact, committing to early compliance with the act.
- Apple and Meta have declined to sign, citing concerns over regulatory issues.
Google, OpenAI, Microsoft, and Amazon are among 100 companies who are the first to sign up to a voluntary alliance aiming to help usher in new AI legislation, the European Commission confirmed.
While the majority of Big Tech companies, including Microsoft and Google, have pledged their support for the EU’s AI Pact, both Apple and Meta have notably declined to join. Both companies face ongoing challenges related to their AI strategies in Europe.
Why Did Apple and Meta decline to sign AI pact?
Both companies’ decision to hold out may stem from their reservations about European AI regulations.
- Apple has delayed the launch of AI-powered features in the European Union due to concerns over compliance with the Digital Markets Act. These new features, set to debut in the U.S. next month, will not be available in Europe.
- Meta, on the other hand, is grappling with General Data Protection Regulation (GDPR) compliance and has postponed the release of its multimodal LLaMA-3 AI model in Europe. Meta has expressed frustration with the unpredictable nature of EU regulations, warning that ongoing regulatory uncertainty could stifle innovation and set European AI development back.
What is the EU AI Pact?
The EU AI Pact is the European Union’s ambitious effort to regulate artificial intelligence (AI). It is a voluntary framework designed to help companies begin aligning with the EU’s upcoming AI Act, which introduces stringent rules and regulations for AI systems, particularly those deemed high-risk. The pact focuses on ethical AI development, promoting transparency, and improving AI literacy across various industries. Companies that sign up are committing to taking proactive steps ahead of formal regulatory deadlines.
The Pact is built on two pillars:
Pillar II: Encourages AI providers and deployers to proactively work towards compliance with the AI Act’s obligations.
Pillar I: Engages the AI Pact network, promotes the exchange of best practices and provides guidance on implementing the AI Act.
Which firms have signed the AI pact?
On September 25, 2024, the European Commission gathered key industry stakeholders in Brussels to celebrate the first signatories of the AI Pact. Over 100 companies have signed the voluntary pledges, representing a wide range of sectors including IT, telecoms, healthcare, banking, automotive, and aeronautics. Signatories range from multinational corporations like Microsoft and Google to European small and medium enterprises (SMEs).
Alongside Google, Microsoft, OpenAI, and Amazon other household names include Adobe, IBM, Mastercard, and Samsung. Others on the list include SAP, Salesforce, Snap, Airbus, Porsche, Lenovo, Qualcomm, Accenture, Atlassian, Cisco, and Palantir.
Here is the full list of early AI Pact sign-ups.
AI Pact core actions
Participating companies are asked to commit to three core actions aimed at promoting ethical and responsible AI development:
- Adopting an AI governance strategy: Companies are expected to implement AI governance frameworks that align with the AI Act and promote the responsible adoption of AI technologies.
- Identifying high-risk AI systems: Organizations are required to map and identify AI systems likely to be categorized as high-risk under the AI Act to ensure compliance with future regulations.
- Promoting AI awareness and literacy: Signatories must foster AI literacy and awareness within their organizations, ensuring that staff are educated on the ethical and responsible use of AI.
These commitments are designed to encourage companies to build ethical AI frameworks and prepare for the stringent requirements of the AI Act, which will come into full effect over the next few years.
What’s next?
The AI Pact is a strategic move by the EU to speed up industry adaptation to upcoming regulations and ensure responsible AI development. Although voluntary, the AI Pact serves as a crucial preparatory step for companies before the AI Act’s binding rules take effect.
Companies that fail to comply with the Act could face fines of up to 7% of their global annual revenue, up to 3% for violations of other obligations, and up to 1.5% for supplying incorrect information, making early compliance a strategic move for many organizations.
