From Anita’s desk
Anita Patel
Minerva Tec Group
Every business leader I speak to is implementing AI tools or planning to, but is the network actually ready? In this July 2026 issue, I'm looking at the bandwidth, latency, and infrastructure AI needs, because AI doesn't run on enthusiasm.
3x
Projected enterprise traffic within three years
85%
Organisations lacking AI-ready network flexibility
50%+
Traffic increase when scaling AI within two years
Your AI investment depends on your network
This issue projects that AI will triple enterprise network traffic within three years and reports that 85% of organisations lack networks flexible enough to support AI at scale. Businesses scaling AI tools experience traffic increases of more than 50% within two years.
Every prompt, response, and real-time analysis moves data across your network. Slow responses, incomplete outputs, and tools timing out can be network problems rather than AI problems.
Market watch: treat networking as business-critical
A well-connected head office isn't enough if other sites still run on 2018 infrastructure. Network as a Service (NaaS), SD-WAN, and managed connectivity make enterprise-grade networking accessible to mid-size businesses without the enterprise price tag.
Ask your IT team whether bandwidth has been reviewed since cloud and AI tools were introduced. If it hasn't been reviewed in two years, start there. A network that was right for 2024 may be throttling your business today.
- Enough bandwidth for AI traffic alongside other workloads.
- Low latency for real-time responses.
- Resilient infrastructure without a single point of failure.
- Visibility into network traffic and performance.
Partner spotlight: Meter Network as a Service
Meter brings wired, wireless, SD-WAN, and cellular networking together as one managed service for a predictable monthly fee, with no capital outlay. Its CBRS cellular solution addresses coverage gaps with faster deployment and lower costs than traditional antenna systems.
Hundreds of businesses rely on Meter to manage their networks, freeing IT teams to focus on business priorities rather than juggling vendors and troubleshooting reactively.
Ask Anita: my team is using ChatGPT without telling me
Be concerned rather than panicked, but do act. The risk isn't simply the tool; it's the client data or internal documents being entered without visibility over where that information goes.
Start with a conversation, not a crackdown. Find out which tools are being used and with what data, then build a clear picture and appropriate AI governance. Get in touch if you'd like help. I read every question.
Key takeaways
- Assess network capacity before scaling AI.
- Review bandwidth, latency, resilience, and visibility across all sites.
- Consider managed networking options that fit your business.
- Address unapproved AI use through visibility and governance.
View the original document (2 pages)
