In recent years, there’s been a fervent belief that AI holds the key to solving numerous challenges across every industry – and telecoms is no exception. However, the truth is more nuanced. AI clearly has enormous potential, but also comes with risks and limitations.
In this two-part blog series, Openmind Networks gets to grips with the transformative potential of artificial intelligence, as well as the reasons why we shouldn’t get too excited just yet!
In Part I, our experts play devil’s advocate and outline why AI, in its current form is not the silver bullet for all operator challenges around growth and cost reduction.
(Stay tuned for Part II next week, where we’ll examine the flip side of the argument!)
Data Quality Issues
Telecom operators deal with vast amounts of data generated from a variety of sources such as network logs, customer interactions, and equipment sensors. However, there can be an overwhelming amount of data coming from these data sources and it is not something that is easy to capture for usage for AI programs. AI algorithms, large language models (LLM) and other potential use-cases rely on high-quality data for accurate predictions and insights. If, for example, network data on text messages sent, call detail records, or even message content is not accurately captured and surfaced for the AI software, then the outputs of AI cannot be fully trusted. In the absence of clean and dependable data, AI solutions may produce flawed outcomes, leading to erroneous decision-making, and passive, non value-add use cases like simple chatbots.
Complexity of the Networks
Telco networks are incredibly complex, comprising diverse technologies, protocols and equipment. AI algorithms struggle to comprehend and optimize such intricate systems fully. While AI can automate certain tasks and processes within telco networks, addressing fundamental issues like network optimization, resource allocation and fault prediction requires a deep understanding of network architecture and domain expertise that AI alone may not possess. As is the case with many AI applications currently, they are a significant time saver and assistant to humans with expert knowledge. They are not capable of operating in complex environments with much autonomy as of yet. Reducing overheads in manpower is still possible with AI assistants, but their impact is still limited.
Regulatory and Compliance Challenges
The telecommunications industry is subject to stringent regulations regarding data privacy, security and consumer protection. In the EU, for example, GDPR does not allow for the reading of message content, presenting issues for firewall operators who are seeking to protect mobile subscribers from malicious content or links. AI systems operating in telco environments must adhere to these data protection regulations, which often entail complex legal and ethical considerations. Ensuring compliance while leveraging AI for innovation poses a significant challenge for telcos, as any misstep could result in legal repercussions and damage to their reputation.
Upskilling the Workforce
Having the right foundation is only part of the issue facing operators. Integrating AI across operations will necessitate reskilling the workforce, a task not to be underestimated in an industry that his historically been slow to embrace new innovations, and workplace cultures that could be disrupted by the introduction of new technologies, adding further complexity.
Rolling out AI-powered technology, training and upskilling employees to deviate from traditional processes and procedures will likely be a protracted endeavor. While operators envision enhanced productivity, streamlined processes, and increased profitability through AI adoption, employees will be concerned about job security. Major operators like Ericsson, Vodafone and BT have already announced significant job cuts as they pivot towards emerging technologies.
CapEx is Being Spent Elsewhere
Realizing the full potential of AI demands telecom operators to redirect capital expenditure to the appropriate avenues and establish the foundations required to implement and harness AI-powered technologies across the organization.
Is this a feasible short-term proposition for an industry currently fixated on rolling out 5G networks and infrastructure, having already committed substantial CapEx to this endeavor?
To learn more about the topics covered in this article, or to discuss how Openmind Networks’ solutions might align with your messaging needs, please get in touch with our experienced messaging experts at [email protected] or contact our team of experts online here.
