The Center for the Fourth Industrial Revolution

Building a sustainable AI infrastructure

A successful AI economy relies on an integrated strategy and the right operating infrastructure, alongside an understanding of the challenges raised by GenAI’s ultra-high energy consumption. Building the appropriate models, while addressing key challenges, requires investment in infrastructure and collaboration with partners.

• Access to high-speed networks, key for AI connectivity, often requires collaboration between governments, telecom providers and tech firms.
• Protecting AI from the threat of disruption demands resilient, scalable and secure networking. Cybersecurity firms, infrastructure providers and governments must come together to build strong and robust AI systems. Trade corridors can be effective at providing resilient, AI-led supply chains and improving the connectivity between resource-rich regions and manufacturing hubs.
• AI is a highly energy-intensive technology, so working with energy providers and technology innovators is important for addressing its environmental impact. Implementing AI-driven grid optimization is one option, while the use of green energy run data centers with purchase power agreements, provides another.
• The high costs of AI infrastructure may prevent more wide-scale AI adoption. This could be overcome by working with private-sector cloud providers and investing in national cloud infrastructures to enable sovereign AI.
• Inclusion is important for ensuring equal access to AI-ready devices where private sector funding and community skills training can play an important role.
• In areas where network performance is poor, an alternative method of connectivity is required, such as low-earth orbit satellite services, as introduced by NTT DATA in June 2024, to overcome access issues including fiber cable breaks, power disruptions and off-grid locations.
  
  

Developing guardrails for ethics, safety and security

Capitalizing on the opportunities posed by AI must be carefully balanced against the importance of establishing robust ethical, regulatory and safety frameworks for the protection of individual privacy.

As AI deployment accelerates, governance becomes ever more important and is best served through a coordinated and integrated approach at local, regional and global levels. This requires stakeholders to come together and work in accordance with the five key guardrails that will determine the success of AI’s global adoption.

• Ethics are essential for the global adoption of AI and form the basis of societal trust.
• Regulation plays a key role in supporting a safe, fair and respectful society by balancing the demands of AI with the needs of maintaining data protection and personal privacy, and democratizing access.
• Safeguarding the world from the unintended consequences of AI requires appropriate policies and procedures around safety and security.
• The responsible use of AI must continue to highlight ethical, transparent and socially beneficial uses that extend to governmental regulations.
• Protecting intellectual property to fuel AI innovation must be balanced against the need for open access and the complexity of data ownership across international boundaries.

The Forum blueprint identifies critical decision points for all stakeholders and provides a pathway for building new intelligent economies. By advancing these missions within each layer simultaneously, we can work together to ensure AI becomes a true driver of inclusive growth and societal progress, creating a future where no region or community is left behind.