Artificial Intelligence (AI) holds immense potential in the telecommunications industry. AI can assist with operational efficiency and network management, enabling real-time adjustments and optimisations across the entire network infrastructure. It also has the power to revolutionise customer interactions and service delivery through automated and personalised marketing campaigns and sophisticated customer service solutions.
Despite a lot of the hype behind the AI features and use cases that telecom software vendors push being smoke and mirrors, offering little value, the transformative potential of AI is there. Carriers seem yet to realise its full potential, however. Although there is stated willingness to invest in AI, the telecom sector remains cautious about deployment.
At the root of this caution lies a lack of trust in AI, the complications of compliance with existing regulations and a lack of strategic capability regarding people and processes. Only fools rush in, of course, so this is not necessarily bad. However, to build trust and reap the rewards of AI adoption, telcos will need to invest in the IT and AI skills required to support rollout. Seeking industry collaboration and knowledge sharing to identify best practices and absolute clarity on the dos and don’ts will also lower perceived barriers to AI entry. Ultimately, however, more robust regulation will be needed to put the proper guardrails in place for confident AI deployment.
The current regulatory landscape
The European Union’s AI Act allows for the deployment of AI systems in various aspects of telecom operations, including network management, fraud detection, and customer service. Despite this, many telecom service providers are hesitant to adopt AI technologies due to the stringent compliance requirements and the potential for hefty fines similar to those imposed under the General Data Protection Regulation (GDPR).
The EU AI Act mandates several critical requirements for telecom companies, from transparency to data management and human oversight. Telecom companies must ensure that AI systems are transparent to users, providing clear information on how AI-driven decisions are made. Proper data management practices must be in place to ensure that the data used by AI systems is of high quality and free from biases, and adequate human oversight is necessary to intervene in or override AI decisions if required.
Balancing security and transparency
Although transparency and user opt-in are already crucial to align with GDPR and other regulatory frameworks, ensuring data security and privacy while maintaining transparency is a delicate balance – particularly considering the sheer volume of data points and attributes telecom service providers manage. Telecom companies must safeguard sensitive customer information while providing the necessary transparency about their AI systems and data usage. This dual obligation can be daunting and often deters companies from fully leveraging AI’s capabilities.
Therefore, AI models prioritising data protection and compliance will be essential for telecom operators managing vast amounts of sensitive data. This is also leading to a shift towards hybrid cloud strategies and the use of targeted AI models trained on specific internal datasets to mitigate risks associated with data leakage and compliance issues. The Business Support System (BSS) behind the telecom’s operations must be equipped to handle and automate according to the complexity of scenarios created by Artificial Intelligence.
Moreover, the business case is easier to make by focusing on AI solutions that drive tangible returns on investment (ROI) and prioritising AI applications that solve specific business problems and deliver measurable outcomes, such as improved customer experience and reduced operational costs.
Human input and pragmatic oversight
Integrating AI into existing systems and workflows requires careful planning and strategic investment – particularly in the skilled AI talent needed to ensure successful implementation. However, while the human oversight identified by the EU AI Act does add a layer of safety, it can become counterproductive in scenarios where millions of events could trigger changes and workflows. The requirement for human approval in every instance could hinder the efficiency and benefits AI offers. Therefore, a more pragmatic approach to oversight is necessary, allowing AI systems to operate autonomously within predefined safety and ethical boundaries.
The role of industry collaboration
Telco industry collaboration will play a significant role in lowering these barriers to market faced in AI implementation. Initiatives like the UKTIN’s AI Expert Working Group are fostering cooperation among industry experts to explore AI’s potential and address emerging challenges. Such efforts to identify common standards, best practices, technological advancements, and strategic opportunities for AI adoption in telecoms will help to establish the absolute clarity required for both large and small providers to move forward on their AI journey.
Supporting standardisation efforts will also be crucial to ensuring interoperability and reducing risks associated with proprietary systems. Standardisation can help create a level playing field, fostering innovation and competition while ensuring that AI systems from different providers can work together seamlessly.
Unlocking potential
Though the transformative potential of AI for telecoms is enormous, the cautious approach of telecom companies, driven by trust issues, regulatory compliance challenges, and strategic capability gaps, underscores the complexity of AI adoption. To unlock the full potential of AI, further regulation is necessary to build trust among service providers.
Clear, pragmatic guidelines and industry collaboration can help address compliance challenges, ensuring data security, transparency, and the seamless integration of AI technologies. By fostering a regulatory environment that balances innovation with oversight, the telecom sector can confidently adopt AI to reap the rewards. As telecom companies navigate AI’s challenges, the focus should remain on deploying AI solutions that deliver tangible business outcomes and drive innovation, paving the way for a more efficient and customer-centric future in telecommunications.
Article by Liz Parry, the CCO at Lifecycle Software
Comment on this article via X: @IoTNow_ and visit our homepage IoT Now
