Page 20 - Connecting the Future How Connectivity and AI Unlock New Potential

Connecting the Future: How Connectivity and AI Unlock New Potential

Furthermore, research indicates that AI-driven spectrum allocation can significantly enhance spec-
trum utilization:
• A study published in the European Journal of Science, Innovation and Technology demonstrated
that an AI-based dynamic spectrum allocation model achieved 88% efficiency in managing
available spectrum, surpassing traditional static allocation methods. 53
• A report by the Smith Institute highlights AI’s potential to revolutionize spectrum management
by optimizing resource usage and reducing interference. 54
While specific figures may vary across studies, these findings collectively underscore AI’’ capacity
to substantially improve spectrum utilization and network efficiency.

2�1�3 AI Connectivity in the Last Mile

As generative AI applications become more sophisticated and bandwidth-intensive, capable of
producing 4K, 8K, and even 12K video content in real time, the importance of high-performance
connectivity at the network edge has grown exponentially. This evolution marks a shift in how
service providers must approach the last mile of connectivity: the final segment of the network
that delivers services to end users. Once a persistent bottleneck in digital infrastructure, the last
mile is now a focal point for innovation, with AI emerging as a transformative force in bridging
performance, scalability, and user experience.
New generative AI models like OpenAI’s GPT-4o and Sora, Cohere’s Command, and Mistral’s Large
empower users to generate complex, data-rich content on-demand – raising the stakes for latency,
throughput, and compute proximity. The challenge lies in meeting these performance expectations
without overburdening core networks or centralized data centers. This has led to a growing need
for AI-assisted edge computing, where service providers move compute and storage resources

14