AI is not yet a silver bullet for channel efficiency
Jonathan Wright, director of products and operations at GCX explains why the sooner the channel accepts AI for what it is right now, the sooner it stands to benefit
2024 was the year of artificial intelligence (AI) hype, as the technology moved firmly beyond theory and businesses dashed to deploy AI-powered tools and applications in the pursuit of efficiency gains – with IT and telecoms channel partners following suit.
However, the hype bubble meant expectations were sky-high, which is now causing roadblocks to effective uptake and implementation. It’s even fair to say the channel has prematurely entered its period of disillusionment, following what some believe to be a lack of tangible results, and others believe to be a total failure to deliver against albeit inflated expectations.
This calls for a more balanced conversation. Has it really failed? Or did the channel charge in without first assessing what use cases were genuinely achievable, and what could only be considered speculative as far as the near future was concerned.
For now, the disparity between hopes and realities, and significant barriers to implementation remain. Pressing on with inflated expectations and overexposure could leave the channel in an increasingly precarious position.
AI is not yet groundbreaking
It’s easy to see why AI gathered the hype it has, and at its best, certainly could be transformational – but it has a way to go before then. At present, channel partners fall into the trap of making sweeping transformation decisions without adequately considering how AI fits into their overall plans. Let’s consider the genuine, current use cases and take networking as an example.
Broadly speaking, currently, the most consistent use cases for AI are all quite straightforward and rather unglamorous. This includes predictive analytics and simple controlled automation routines like optimising data flow efficiency or running playbooks against scenarios. While these are important for networking and security, you wouldn’t necessarily consider them groundbreaking.
The high costs associated with AI infrastructure spending and the process of model training mean that a very careful return on investment (ROI) analysis is required. This is especially needed to ensure channel partners aren’t rushing to use AI just because it’s the shiny new technology on the block.
Going further, the high associated costs mean channel partners are unlikely to be able to reasonably pass on the costs to customers, especially where the deliverable may be a “narrow AI” deployment with outcomes that may be more akin to automation. Importantly, this limits the value that can be gained, as most of this value may be internal to the channel and thus monetisation may be restricted.
This hasn’t stopped the pursuit of transformative use cases though. We have recently seen a shift towards more autonomous AI, in the form of agents. However, again, this remains unreliable and requires significant human oversight as it stands – further hindering adoption in the networking space, especially as networks expand in size.
Automation and the customer experience play
The short-term focus for channel partners should be to use AI for enhanced insight and trend analysis. This enables better network monitoring and preventive maintenance. Partners with larger scale operations, such as telecom providers can also leverage AI for crucial use cases like traffic predictability, dynamic routing and - arguably - most importantly, the planning of next-generation infrastructure, which requires a more long-term view on the ROI for AI.
Deloitte has additionally suggested a key use case for AI to be customer experience, something becoming increasingly important for MSPs to stand out against competition as reliable technology partners. Therefore, focusing on developing these areas will allow partners to successfully tailor solutions to customer needs – this involves a much higher level of transparency not only to provide the customers with a suite of analysis, but to ensure that supporting processes and people deliver a similarly transparency to enhance the customer experience.
In addition, AI will help build the automation case in networking and security as it makes solutions like Security Orchestration and Response (SOAR) more effective. And, importantly, despite the name, SOAR solutions go beyond simple security automation use cases. For example, SOAR will help bring networking and security management under a “single pane of glass” by automating ticketing, tapping into threat intelligence feeds and speeding up response times for a smoother user experience.
Networking and security are becoming increasingly intertwined amid a convergent framework in SASE and the market direction toward single stack technology solutions, so this will be key as the channel turns its attention towards next-generation network planning. This will be especially true as the market structure evolves to support more adaptive networks. AI’s ability to analyse datasets and predict trends will therefore shape its growing relevance in the channel’s future.
Patience and realism will be key
AI’s potential is clear, but it’s not yet a silver bullet, and it is currently a very noisy marketplace. The sooner the channel accepts this technology for what it is right now, the sooner it stands to benefit.
Focusing on predictive analysis, automation as an efficiency driver and the customer experience will be key going forward, but perhaps most important of all, is the genuine role AI stands to play in helping lay the groundwork for more dynamic and adaptive networks in the future.
If we take things slow and focus on the foundations now, we will be sure to reap the benefits in the years to come.