Featured Story
Vodafone and Three merger will further decimate UK telco jobs
The UK's fast-shrinking market for telecom jobs looks set to get even smaller after the combination of Vodafone with Three.
A top technologist at AWS outlines a few ways generative AI will transform the telco industry and offers some considerations for network operators evaluating AI use.
June 27, 2023
It's hard to look anywhere in the news and not come across headlines on generative artificial intelligence (AI) – a type of AI that can create new content, including conversations, stories, images, videos, music, code and more. We believe this form of AI has the potential to bring about sweeping changes to the global economy. Goldman Sachs estimates it could drive a $7 trillion increase in global GDP and lift productivity growth by 1.5% over 10 years. And this opportunity extends to the telco industry.
(Source: Pitinan Piyavatin/Alamy Stock Photo)
Generative AI's potential for telcos
Here are a few ways generative AI will transform the telco industry and some key considerations in evaluating the technology.
Like other industries, generative AI can help telcos enhance efficiencies in many functions: for example, populating RFPs, deploying chatbots to aid in sales and personalizing marketing to individuals at scale. We expect to see huge growth in this area. However, we believe a few industry-specific applications are truly transformative.
Enhance the customer experience
Already, many telcos leverage AI to augment human interactions and improve the consistency of experience and resolution speed. Generative AI can take these activities one step further with interactive voice response – an evolution of early chatbot deployments to help customers resolve issues or get answers to questions. In addition, generative AI can help analyze real-time call discussions to provide prompts and resources to help agents resolve customer inquiries. Customer service agents will play a key role in the process, but we believe generative AI can reinvent and improve every customer experience and application.
Simplify network planning, installation, configuration and operations
Generative AI can play a key role in all aspects of the network lifecycle. Engineers rely on manuals and documented processes when installing network elements. Generative AI can ingest this data and provide interactive guidance and prompts to speed up and simplify installation tasks. Foundational models can also be trained on network topology and configuration data to suggest the configuration of network elements. Generative AI-based applications can recommend troubleshooting actions and procedures to networking operating engineers when there is a network failure.
Optimize business performance
Generative AI can help telcos identify areas where they are losing revenue or incurring revenue leakage. Deployed across business processes, generative AI can examine profits, revenues, various consumer plans, expenses and customer charges to recommend how to evolve offerings to optimize profits.
Figure 2:
(Source: Sipa US/Alamy Stock Photo)
Key considerations
Only some instances of AI or ML today will require generative AI. There are plenty of examples where traditional forms of AI, based on supervised and unsupervised learning techniques, are more than sufficient for telco processes. For example, traditional AI is great at helping telcos predict churn, detect anomalies in the network or track metrics such as net promoter score. Beyond that, there are other important considerations to keep in mind.
The cost of developing or training foundational models
Much of the capital spent on foundational models – the large AI models trained on enormous quantities of data that power generative AI applications – goes into training them. Public models are trained on large amounts of publicly available data, but are more general purpose and may not work well on specialized tasks. On the other hand, custom models may be trained on a combination of public and company-specific data, providing more specific applications to the industry or organization. Building a model from scratch is time-consuming, expensive and requires specialized expertise. Still, for organizations with significant data, resources and a use case requiring specific domain knowledge, developing a bespoke foundational model may make sense. There are also efforts to democratize access to this technology by letting customers take an existing model as a starting point, and privately train it using the customer's proprietary data, making the model better suited for a specific task.
Data quality and responsible AI
Generative AI is only as good as the data it's trained on, and there's always the risk of bias or inaccuracies. Before considering generative AI (or any form of AI), starting with high-quality, unified data is important. Generative AI requires extensive data sets, training and oversights to make inferences and answers. Sometimes, public and private foundational models can suffer from "hallucinations," creating inaccurate responses that may look believable but are incorrect. For this reason, generative AI is not recommended for tasks that require complete certainty, which might not be possible due to the nature of the problem or lack of sufficiently large and high-quality data.
Equally important is ensuring that this technology is implemented responsibly. Some new generative AI tools and services have responsible AI features built into the product, such as surfacing when generated code resembles existing open source code or detecting and removing harmful content in training data sets and filtering outposts that contain harmful content (e.g., hate speech, profanity and violence).
Data security
For enterprises to leverage generative AI for company purposes, it requires large sets of proprietary data. While there are public options in the market, these approaches introduce new considerations around security and privacy, including intellectual property. Business and IT leaders should work closely with security, compliance and legal teams to identify and mitigate these risks, ensuring that generative AI is deployed securely and responsibly. Further, scope plans around compliance and regulations and think carefully about who owns the data used.
Whether it is generative AI or natural language processing, it's important to consider the applications, discuss a data organization strategy and evaluate the ROI before commercially deploying AI. That said, we believe AI to be the single most transformational technology of our time, and generative AI is unlocking exciting new possibilities that every business should look to explore and experiment with.
— Ishwar Parulkar, Chief Technologist, Telecom & Edge Cloud, AWS
You May Also Like