Generative Enterprise Services 2024: Our half-term report with lessons for every enterprise leader

HFS Research has long recognized that generative AI (GenAI) was developing so rapidly that we would need to follow up on our market-defining Generative Enterprise Services 2023 report with a second one in 2024. Midway through this latest research—which will ultimately involve nearly 50 service providers, consulting firms, and a range of supporting data points—fresh trends and implications for every enterprise leader have already emerged from ongoing conversations with the industry.

The adoption of GenAI has accelerated beyond experimental phases, transforming into large-scale, enterprise-wide initiatives led by major service providers. As service providers respond to enterprise demand, they move beyond simple AI-driven efficiencies to position GenAI as a catalyst for growth, new revenue models, and improved customer engagement.

But this shift is also unearthing new complexities. Enterprises face opportunities and challenges in the GenAI landscape, from agentic AI workflows to responsible AI frameworks. Here’s a first look at what enterprise leaders should know based on the initial themes from our briefings—our half-term report if you will. Our full report will be published toward the end of the year, taking in what we learn from the remaining briefings, meetings, and exposure to technology updates as we progress.

Those firms and examples mentioned in this report are illustrative examples of the many more we are discovering as part of this research.

Exhibit 1: Service providers are embracing growth, ecosystems, agentic AI, and more—moving beyond an initial emphasis on cost-cutting and productivity

Source: HFS Research Generative Enterprise Services Horizons, 2024 briefings (at midway point)

Growth-focused AI services are shifting the narrative from cost-cutting

Based on our conversations to date, 70% of service providers are moving away from the familiar narrative of AI-driven cost efficiencies, positioning GenAI as a driver of business growth. This group includes Infosys, Mphasis, and Tech Mahindra (see Exhibit 1). They are among a growing group of service providers shifting their value propositions to emphasize GenAI’s ability to enable new revenue streams, enhance customer experiences, and unlock strategic insights.

For example, Mphasis has re-engineered its mortgage processing service, using GenAI to deliver more personalized customer journeys, reducing average processing times and raising customer satisfaction. This focus on customer experience represents a significant evolution from earlier AI implementations focused solely on back-office automation. Providers are touting their growth potential by leveraging advanced customer personalization, increased engagement, and value-adding services that respond dynamically to real-time needs.

This pivot should encourage you. The strategic focus on growth aligns AI initiatives more closely with top-line objectives, making GenAI an enabler of business agility and revenue generation rather than just an efficiency tool. However, they demand clear metrics and KPIs from providers to ensure growth targets are achievable and not overly ambitious.

Agentic AI and autonomous workflows are a step-change in enterprise AI capabilities

Agentic AI—already much discussed by 60% of service providers—represents a step-change in AI’s capabilities, enabling GenAI systems to act autonomously within defined parameters. Unlike traditional reactive AI, agentic AI can initiate actions, adjust workflows, and make autonomous decisions based on real-time inputs, driving highly adaptive “agentic workflows.”

This embrace of agentic AI aligns with an accelerating journey toward leg four of our Tech Services Vision 2030 (see Exhibit 2)—AI-led Agentic Services.

Exhibit 2: 60% of service providers that have engaged to date with our 2024 Generative Enterprise Services Horizon report are talking up agentic services

Source: HFS Research, 2024

Providers such as Firstsource and Virtusa are shaping use cases that illustrate the possibilities. Firstsource, for instance, applies agentic AI to claims processing in healthcare, leveraging “human-in-the-loop” (HITL) models to ensure compliance while enabling autonomous claims adjudication.

Agentic AI holds transformative potential for enterprises, especially those with complex, data-heavy workflows requiring fast response times. However, accountability and trust remain crucial concerns. Leaders should examine governance frameworks closely, balancing the potential for speed and autonomy with robust oversight mechanisms to mitigate risks of errors and bias.

Industry-specific solutions are on the rise, tailored to vertical needs

More than half (55%) of service providers offer industry-specific AI solutions, building bespoke models and tools that address various sectors’ distinct regulatory and operational demands. Others are in the process of building out their offerings in this direction. Infosys’ industry blueprints are just one example—emphasizing solutions tailored to vertical-specific needs.

In theory, these tailored models bring clear benefits regarding relevance and impact, particularly in highly regulated sectors such as finance and healthcare. However, they risk vendor lock-in as enterprises increasingly depend on proprietary models, infrastructures, and partnerships. As these models gain traction, enterprises risk being tethered to a single vendor’s ecosystem, potentially limiting flexibility and increasing long-term costs.

Leaders should seek assurances around interoperability and the ease of transitioning to other models if necessary. Additionally, they should confirm that these specialized AI tools can scale to meet changing business requirements without needing constant customization.

Responsible AI and governance frameworks are becoming table stakes—but they don’t have all the answers

Providers are actively embedding governance into their GenAI offerings with frameworks that aim to enforce ethical standards and regulatory compliance across AI deployments. KPMG’s “Trusted AI” model is one example of targeting compliance challenges head-on in regulated environments. These frameworks, emphasized by approximately half of service providers to date, offer enterprises built-in protections against data bias, security vulnerabilities, and lack of transparency.

Governance frameworks offer a strong foundation, but true transparency and bias management remain a work in progress across most provider offerings.

Such initiatives are critical to risk-averse sectors such as finance, healthcare, and government. However, leaders should be wary of taking provider assurances at face value. Enterprises must thoroughly assess each provider’s framework to ensure they align with internal governance and compliance needs, especially around data use and privacy.

We need to strike a human-AI balance between risk and innovation

Human-in-the-loop (HITL) models are gaining traction (40% highlighting them) as providers seek to enable complex GenAI applications that require human judgment. In financial advisory or claims processing scenarios, HITL workflows balance AI’s speed with human oversight to manage risks, adapt to nuances, and maintain accountability.

However, while HITL provides a safeguard, it limits the potential for full automation. Service providers are advocating HITL to mitigate risks, yet the long-term goal should be to reduce human intervention as AI matures progressively.

HITL can bridge the trust gap in early GenAI deployments. However, enterprises should see it as an interim step, working with providers to gradually phase out human oversight as AI models demonstrate increased accuracy and reliability. Leaders must carefully plan this progression to avoid getting stuck with workflows that can’t fully scale due to HITL dependencies.

Versatility and scalability demand strategic partnerships and ecosystem collaboration

In response to enterprise demand for versatile, scalable AI capabilities, service providers are forming alliances with hyperscalers and niche AI firms. Partnerships with major players such as AWS, Microsoft, and Google enable providers to leverage advanced AI platforms, while partnerships with specialized AI firms (e.g., Hugging Face) bring additional expertise. Two-thirds of the firms we have spoken with are prioritizing such relationships.

For enterprises, these partnerships mean faster access to advanced tools and resources without significant infrastructure investments. However, they also increase the risk of over-dependency on a limited number of tech ecosystems, which could restrict interoperability and flexibility over time.

While strategic partnerships expedite time-to-market for new solutions, enterprises must retain control over their AI roadmap. Leaders should emphasize multicloud compatibility, ensuring solutions integrate across existing enterprise systems without locking them into one vendor’s environment.

The Bottom Line: Proceed with enthusiasm and some caution: Understand agentic, establish strong governance, and stay flexible

The halfway mark in our 2024 Generative Enterprise Horizons research reveals a market rapidly evolving to meet enterprise needs for growth, agility, and compliance. Providers are moving toward comprehensive, autonomous AI solutions tailored to sector-specific challenges. They also embed governance frameworks that aim to ensure responsible use while supporting complex workflows.

As you prepare to integrate GenAI more fully, move forward with both enthusiasm and caution. Focus on setting clear objectives, understanding the full implications of agentic AI, and establishing strong internal governance to complement provider-led controls. By doing so, you will make the most of GenAI’s capabilities while guarding against vendor lock-in, compliance risks, and scalability limitations.