NVIDIA Report: Telco Transformation Requires Artificial Intelligence

This syndicated post originally appeared at Zeus Kerravala, Author at eWEEK.

AI is the missing piece in helping telcos complete their transformation puzzle.

Mobile World Congress (MWC) kicks off this week in Barcelona. The theme of the event is telco transformation – as it has been for some time.

Clearly, the evolution of telcos is long overdue. Most communication and network service providers have networks that rely too heavily on people, spend too much on infrastructure and deliver services too slowly to compete with the cloud companies. This leads to their net promotor and customer service scores being abhorrently low. This is why this group of companies is often relegated to being the “plumbing” of company infrastructure.

Modernization Critical to Long Term Success

Digital transformation has given telcos an excellent opportunity to change the way they are perceived. The building blocks of digitization are technologies like cloud, IoT and mobility and these all have one thing in common – they are network-centric, making the network more valuable.

In a recent survey of mine, I asked business leaders their opinion of the value of the network today versus pre-pandemic and two-third believe the network to have increased in business value, putting the telcos in a strong position to help their customers transform.

To accomplish this though, telco modernization must happen. At MWC, this opinion was shared with everyone I spoke with. However, it is a fact that we have been talking about the need to evolve for over a decade. This large a shift requires new hardware, software and a completely different operating model. While the vendor community has done a nice job of evolving equipment, it’s also true that labor and skill shortage has remained a problem. But the rise of artificial intelligence (AI) could change that.

AI: Core Component of Telco Modernization

AI can revolutionize the telecom industry by transforming all aspects of network planning and deployment, operations, and customer engagement. Although the technology is still in the early implementation stage, telcos are starting to invest significantly in AI, according to a new report released by NVIDIA titled State of AI in Telecommunications: 2023 Trends.

NVIDIA surveyed more than 400 telecom professionals globally to evaluate the opportunities and challenges of implementing AI in this industry. The respondents included C-suite leaders, managers, developers, and IT architects from mobile, fixed, and cable companies. The survey was conducted between mid-November 2022 and mid-January 2023.

The report uncovered that most (95 percent) telecom professionals are engaged with AI, which is even higher among industry decision-makers (100 percent). Despite the high interest in AI, however, many companies still haven’t widely adopted AI. Only 34 percent of the respondents said they’re using AI for more than six months. Meanwhile, 23 percent are trying to figure out different options for AI, and 18 percent are in the pilot phase.

AI’s Impact on Optimizing Operational Costs

Operational optimization could be the biggest opportunity for AI in the telecom industry. Approximately 60 percent of the respondents believe AI automation that replaces or augments human effort will drive efficiency. Additionally, 44 percent expect AI to reduce the cost of operations and 35 percent expect AI to enhance customer engagement.

For companies that are already implementing AI, it has had a positive impact on both revenue and cost.

AI Leads to a Boost in Revenue

In the last year, 73 percent of the respondents saw their revenue increase after implementing AI and 17 percent reported revenue gains of more than 10 percent in specific parts of their business.

Furthermore, 80 percent of the respondents said their annual costs decreased in the last year and a smaller number of the respondents (15 percent) reported a reduction above 10 percent in parts of their business.

A leading challenge for widespread adoption of AI in the telecom industry is the inability to quantify its return on investment (ROI), according to 44 percent of the respondents. Despite AI having a positive impact on revenue and cost, there seems to be a disparity between expectations and reality when implementing AI-based solutions.

Other challenges cited by telecom professionals include a lack of skilled personnel, poor infrastructure, and not enough funds for AI. For instance, 34 percent of the respondents said their company has an insufficient number of data scientists, while 33 percent said they don’t have a budget for AI projects.

AI Investment Needs to Increase

Given the challenges, investment in AI appears to be low, where 50 percent of telcos are spending less than $1 million annually. In comparison, telcos typically spend at least a billion dollars annually on capex.

The report’s findings show that the level of investment in AI doesn’t match the level of enthusiasm and engagement. Only 42 percent of the respondents agreed that their current investment is at the right level.

However, the outlook for 2023 is positive, with 47 percent of telecom professionals planning to increase their AI investments.

Although nearly half (49 percent) are still in the learning, assessment, or trial/pilot phase, telcos are determined to move toward implementing AI. Data shows telcos are testing various AI-enabled solutions for network operations, base station site planning, truck-routing optimization, and machine learning data analytics. Many are also focused on improving the customer experience by implementing engines, virtual assistants, and digital avatars.

In the next six to 12 months, telcos plan to increase investment in predictive maintenance and use AI to tackle problems in their hardware and software infrastructure, including customer on-premises equipment like mobile devices and set-top boxes. Machine learning (ML) and high-performance computing platforms seem to be the leading AI tools of choice. Telcos are also investing in deep learning, digital twins, and simulation for the purpose of testing infrastructure upgrades in the virtual world before rolling them to customers.

Today, AI is deployed mostly in hybrid environments (cloud/on-premises) and telcos are happy with this model. But there is a shift happening to all-cloud for companies that are seeking improved performance of cloud solutions. The biggest concerns for companies that are worried about moving to all-cloud include information security (35 percent) and network stability (32 percent). Cost wasn’t named as a priority when choosing a deployment environment for AI, suggesting that telcos are taking more of a strategic approach rather than a cost-driven one.

Going forward, partnerships will play a critical role for telcos implementing AI. Telcos will need reliable, knowledgeable partners for supplies, support, and expertise to accelerate adoption. In fact, 83 percent of the respondents said they use third-party consultants or software vendors for AI projects. Additionally, partners will help develop new services, and as a result, allow telcos to deliver better customer experiences.

Author: Zeus Kerravala

Zeus Kerravala is the founder and principal analyst with ZK Research. Kerravala provides a mix of tactical advice to help his clients in the current business climate and long term strategic advice.