Group Of Businesspeople On A Meeting At Their Company

Five “no regret” actions for TMT companies to unlock generative AI’s potential

Authors

15 minute read 30 Nov. 2023
Related topics
AI

TMT companies already have a lead in GenAI adoption. We’re exploring how they can maintain that lead and unlock value, now and in the future.

In brief

  • With over half (52%) of TMT companies already using GenAI – well ahead of other industries – the sector is uniquely well-placed to drive GenAI adoption.1
  • As early adopters, TMT companies will also be among the first to tackle GenAI’s uncertainties and risks. 
  • To maximize GenAI’s value, TMT companies must pursue innovation and integrity through select actions. 

As the world of generative AI (GenAI) takes shape, technology, media and entertainment and telecommunications (TMT) companies will play a crucial role by injecting GenAI into their service portfolios and deploying it in their internal digitalization roadmaps. GenAI’s transformative potential and rapid acceleration also pose challenges, from organizational constraints to regulatory uncertainties. Mindful of these complex forces, we’ve identified five ”no regret” actions that TMT companies should pursue to convert GenAI’s promise into long-term value.

TMT companies will be at the forefront of the GenAI revolution. Many technology providers will enable GenAI solutions for their customers, with telcos enabling the key infrastructure. For media and entertainment companies, GenAI will trigger both innovation and disruption of business models.

Organizations across all three sectors can also harness GenAI as they transform internal systems and processes, building on the benefits of “traditional” AI. Here, TMT companies find themselves in a promising position relative to other industries. This is evident in higher levels of GenAI adoption, led by technology companies. It’s only in media and entertainment companies where employee usage falls below the cross-sector average, a possible sign of greater employee resistance. But taken together, TMT companies are well-placed to both deploy GenAI internally and build it into their products and services.


Encouraging usage levels are supported by ongoing investment: 45% of TMT companies are investing in AI-driven innovation – with AI product and service changes already integrated into capital allocation processes – and a further 46% plan significant capital investments in the next 12 months.2 This puts TMT companies in prime position to leverage GenAI to benefit the wider economy. And while GenAI is potentially disruptive, TMT companies have previously adapted to changing technology cycles and the pressures they bring. Looking ahead, TMT companies should focus on how best to harness GenAI as part of the AI capabilities they already possess.

Shaping new opportunities for customers, platforms and collaboration

As TMT companies explore how they can benefit from GenAI – both in the near- and long term – they should consider use cases that deliver both growth and efficiency. Use case clusters that stand out include:

  • Driving customer experience excellence: In the near term, GenAI can revolutionize customer interactions by creating smarter chatbots, and by blending human and digital assistance in new ways. This is particularly relevant to telcos with extensive customer support operations. GenAI can also help personalize services, ensuring that investments – whether in content recommendation engines or fiber network upgrades – better fit specific customer needs.
  • Unlocking platform business models: In the longer-term, Gen AI can unlock the power of new business models, specifically service platforms made available across ecosystem partners and intermediaries. GenAI’s potential to help enhance product development and distribution, curate B2B2X customer journeys and improve “ecosystem satisfaction” means it can play a pervasive role in business model innovation.
  • Becoming an AI-augmented organization: Knowledge management, productivity and automation all stand to gain in a GenAI-centric organization with more empowered employees. By improving information accessibility and sharing, TMT companies can finally overcome enduring silos – between IT and product development functions, for example – automate low-value tasks, and free up time to focus on value creation.

Use case needs will vary by sector. Technology companies will be keen to enhance their existing platform offerings with GenAI, looking for the most promising adjacency with other emerging technologies in their product portfolio. Telcos will focus on how GenAI can either augment or substitute existing customer care channels, systems and processes. And media providers will assess GenAI’s specific role at every stage of the content lifecycle, from creation through to curation and distribution.

Figure 2: Indicative use cases by TMT sector

Source: EY Knowledge analysis

GenAI sets new challenges for TMT

The fact that TMT companies are further along the GenAI adoption path than other industries means they will be the first to experience its challenges. Indeed, 68% of TMT companies (pdf) believe they’re not doing enough to manage the unintended consequences of AI, underlining the responsibilities that come with ”first mover” advantages. We see the following as major issues that require management attention:

  1. Uncertain regulatory and policy environment: Regulators recognize the ethical challenges posed by AI, alongside the need for new forms of data governance and protection. While new rules should bring more certainty, they will take time to embed and may differ by geographic region, adding to complexity and potentially limiting innovation. Leading TMT companies have already voiced such fears regarding the prospective EU AI Act3.

  2. Ecosystem limitations: While many TMT companies have supplier and value chain ecosystems in place, it may be difficult to absorb a new wave of GenAI-based partnerships within those existing ecosystem structures. New and existing partners’ rationales and priorities may differ when it comes to GenAI implementation, or they may face contrasting regulatory pressures.

  3. Budget and investment constraints: Challenging macroeconomic environments could constrain TMT companies’ ability to invest in AI. This, in turn, may force greater reliance on partners to bring in GenAI technology capabilities and expertise. Likewise, budget constraints may also limit enterprise customers’ ability to invest in both AI technology and skills.

  4. Employee and customer resistance: GenAI’s potential to accelerate automation and reduce human involvement is already unnerving employees, notably in the US media and entertainment industry.4 At the same time, customer concerns about data privacy and quality may create confusion, or hinder acceptance of AI-based interactions. EY research shows that 48% of consumers are worried about algorithms used by apps and websites, and how they impact what they see online.5

  5. Inadequate data and intellectual property (IP) governance and protection: TMT companies are data-rich organizations, but this creates complexities. Clean, curated datasets are crucial to train GenAI algorithms. Organizations where data still remains fragmented and hard to manage, will need to adapt their data governance faster and further. Meanwhile, new risks can emerge from combining proprietary and public data, and from increased data sharing between TMT companies, partners and customers. What’s more, the risk of IP infringement will likely grow in a GenAI world.

Organizations also need to consider sector-specific challenges. Telcos, for example, should consider how GenAI adoption will impact future network loads and associated investment commitments. Media companies may experience business model disruption ahead of other industry sectors – as with previous technology cycles – while technology companies may be the most exposed to regulatory uncertainties as they look to globalize GenAI platform solutions.

These nuances aside, TMT companies realize they need to collaborate further with other industry stakeholders to address ethical risks: 74% of CEOs believe that the business community needs to focus much more on the ethical implications of AI and how its use could impact key areas of our lives.6 This sense of shared commitment has important implications for ecosystem strategies.

Five key actions for TMT companies

1. Establish an AI control tower to centralize innovation, knowledge and skills.

Thirty percent of TMT companies already have a group dedicated to AI adoption and use.7 GenAI should act as the trigger for all companies to go further and establish an AI control tower. This moves beyond use-case experimentation to help reimagine business models, improve governance and centralize skills. The control tower group can comprise a mix of business-unit heads and other relevant executive roles – chief digital and data officers, for example – to identify and prioritize GenAI opportunities, while assessing disruptive risks, talent requirements and data governance needs.

Where to act now

Designate someone from the C-suite as an AI leader, to plan and coordinate the control tower’s activities with other parts of the business, including pre-existing digital business units or centers of excellence. Ensure that the activities of the control tower are aligned to the organization’s overall business and technology strategies.

Identify relevant skills required and immediate skills gaps, taking care to consider new roles that report to AI leadership. Meanwhile, train teams in the business and technical aspects of GenAI, leveraging core principles that can inform longer-term reskilling needs.

Develop a portfolio of targeted GenAI opportunities. As part of this, revisit your existing catalog of AI use cases and identify opportunities to incorporate GenAI into them where feasible. Prioritize GenAI use cases based on metrics such as impact, complexity, scalability and time to market. Pick a healthy mix of “quick wins” and more complex use cases.

Meanwhile, technology companies providing GenAI-based offerings to enterprises should consider commercial principles upfront. These include whether to offer GenAI as a standalone product or incorporate it into existing service bundles, and the best pricing model to begin with – from free trials and tiered pricing, through to value-based pricing.

What to decide later
  • Design a comprehensive roadmap for scaling GenAI solutions across the business. As organizations learn more about use cases’ outcomes and feasibility, teams can decide whether to allocate more resources to GenAI projects.
  • Explore more transformational business models and service portfolios that take advantage of GenAI – these include platform or B2B2X services that involve joint go-to-market with partners. Define the commercial terms that underpin “sell with” approaches, paying particular attention to revenue-share models.
  • Create a long-term plan to acquire new talent with GenAI capabilities in areas relevant to priority use cases. Regularly review and prioritize specific GenAI roles, focusing on those that will differentiate your business over time, as opposed to those likely to become commoditized.
2. Reimagine business functions and ways of working

Realizing GenAI’s potential to increase productivity and overhaul business models hinges on new ways of working. GenAI will enable more seamless interaction between business functions, with roles and responsibilities evolving over time. Organizational structures and processes should reflect these new ways of working. New information flows between previously siloed teams become possible, with GenAI empowering employees rather than displacing jobs. Initial signs from TMT employees are positive: 51% expect a net positive impact on how work is done.8

Where to act now

To mitigate employee resistance, launch small-scale pilot projects using proprietary enterprise data to test GenAI solutions and gather feedback. Use the results to show how GenAI can enhance existing processes, improve employee efficiency and augment capabilities.

Ensure that your AI control tower works closely with other parts of the business to create the right internal feedback loops. Make sure that employees at all levels feel that they’re part of the journey by explaining how GenAI is being deployed and the underlying data sets it uses. This will help to build trust and confidence in AI-based outputs.

Make sure that leadership clearly communicates how workflows are changing, and that business unit leaders meet regularly to share progress and future plans, Highlighting AI’s role as a collaborative tool transforms it from a source of possible resistance into an opportunity for growth.

What to decide later
  • Re-evaluate the operating model in light of AI-driven improvements in data management, paying attention to new points of intersection between legacy business functions, and between those functions and the AI control tower.
  • Invest in function-specific GenAI training, as well as organization-wide upskilling and reskilling. Develop a talent development plan in sync with your technology roadmap and business function transformation. Consider creating an internal talent marketplace to enable employees to shift to new, emerging roles and build multi-function AI competencies.
  • Introduce continuous monitoring mechanisms and develop key performance indicators (KPIs) to demonstrate GenAI initiatives’ long-term value.
3. Put GenAI at the center of your ecosystem strategy

TMT companies, from technology giants and hyperscalers to network equipment vendors and telcos, are established ecosystem orchestrators. That experience could put them at an advantage, but it also means they need to consider how to adapt existing ecosystem structures. Begin by assessing capability gaps, ensuring that the ecosystem strategy caters to evolving AI opportunities, and look for new ways to tap into cutting-edge research and knowledge.

Where to act now

Prioritize AI discussions with the existing partner ecosystem, highlighting areas of mutual interest and potential cooperation in GenAI. Continual monitoring of the AI partner landscape for new opportunities is critical. Identify new partners – whether start-ups, immediate industry peers or academic institutions – that can enhance your GenAI initiatives. Meanwhile, technology companies can partner with companies in different industries to create customized, domain-specific large language models (LLMs) and proprietary knowledge graphs. These can be integrated with public models or offered as a service.

Assess your GenAI readiness across different layers, such as infrastructure (compute facilities, cloud, data), model, or applications development, identifying the role partnerships can play. Plug into existing pre-trained models and data ecosystems to explore use cases. All throughout, TMT companies should ensure that secure data sharing and integration protocols are adopted by ecosystem partners.

What to decide later
  • Scale the AI competencies within your ecosystem through select partners and deprioritize less relevant partnerships.
  • Develop closer relationships with start-ups for co-innovation and consider acquisitions and joint ventures that can extend skills and expertise.
  • Conduct regular reviews of your ecosystem strategy to ensure it’s fully aligned with your evolving AI objectives. And pay close attention to policy or regulatory factors that may influence partner choices and suitability.

How EY can help

4. Build stakeholder confidence in AI

Many TMT companies have designed ethical frameworks for AI. But GenAI creates a whole new set of ethical dilemmas and security risks. TMT companies should address stakeholder concerns about AI-generated content such as IP and copyright issues, fake content, security and data privacy for training LLMs. Employee concerns are no less important. One interesting example in the creative industry is the Human Artistry Campaign, which advocates AI best practice for artists, performers, writers and athletes.9 Regular dialogue with policymakers is essential as the AI regulatory landscape continues to evolve. And in the absence of dedicated AI regulation, TMT companies should prize robust governance to build confidence in their AI applications.

Where to act now

Identify new risks emerging from GenAI and implement tools to mitigate them. Establish teams to implement and supervise ethical AI procedures, and ways to monitor and audit them. Before launching LLMs, stress-test them for model hallucination, jailbreaking, inappropriate content or other legal and reputational risks.

To encourage further adoption, address concerns about GenAI data privacy. Enterprises need assurance that technology companies won’t use their proprietary data to train general LLMs or inadvertently leak sensitive information during model training. Content publishers should prioritize solutions to track and control content such as deepfake audio, video or text.

Explore current solutions such as labelling AI-generated content and consider ethical AI training programs to educate and upskill employees. Take care to build controls that account for the probabilistic nature of the outputs and ways to verify the quality and robustness of results.

What to decide later
  • As GenAI goes mainstream in your organization, devise comprehensive ethical frameworks for data use and model implementations, educating users at all levels about risks and mitigation strategies.
  • Combine forces with industry players, regulators, non-profit and public bodies, academics and thought leaders to shape regulation of AI-generated content and address IP-related issues in regulation.
  • Create new feedback loops with customers around AI-based services, evaluating current levels of acceptance and future receptivity.
5. Inject GenAI into multi-year tech transformation

GenAI technology is evolving fast. While it’s important to keep your options open, understanding how best to harness GenAI within broader technology transformation programs will be crucial for long-term value creation. Many TMT companies are deploying a growing array of emerging technologies – cloud, edge computing, quantum computing – to accelerate transformation, and GenAI should not be treated in isolation, but as additive to other emerging technology investments.

Where to act now

Prioritize technology deployments based on your business strategy. Ensure your technology roadmap is in sync with the GenAI pilots you select. Prepare datasets for identified use cases in specific domains. Ensure availability of cloud and compute infrastructure to test smaller sets of GenAI solutions on proprietary data. Technology companies supplying GenAI solutions can offer synthetic datasets to demonstrate the power of GenAI solutions.

Consider how GenAI solutions will integrate with your existing technology stack, including content management, customer relationship management (CRM) and Enterprise Resource Planning (ERP) systems. In terms of infrastructure, ensure you build in flexibility about where you can run a full-stack foundation model.

 

What to decide later
  • Ensure enterprise architecture is ready for scaled GenAI deployment, including data, applications and infrastructure. Consider modernizing data and application platforms and upgrading compute infrastructure to support larger-scale models.
  • Evaluate points of adjacency and intersection between GenAI and other emerging technologies such as edge and quantum computing to maximize their combined impact and value creation.
  • Assess the impact of growing use of compute and datacenter facilities on your Environmental, Social and Governance (ESG) and sustainability commitments. Create metrics to track sustainability and measures to mitigate downside risks.

Summary

With a commanding lead in GenAI adoption, TMT companies are well positioned to drive value from this breakthrough technology. But to do that they must also be the first to overcome new and testing challenges that GenAI poses.

The five no-regret actions outlined here can help TMT companies do just that. As they develop GenAI capabilities, they’ll need to simultaneously focus on what they can do now and plan for later steps. GenAI will change the world. TMTs can make sure that they – and their stakeholders – maximize its value. 

About this article

Authors

Related topics
AI

EY refers to the global organization, and may refer to one or more, of the member firms of Ernst & Young Global Limited, each of which is a separate legal entity. Ernst & Young Global Limited, a UK company limited by guarantee, does not provide services to clients.