Who Will Emerge As The Winner In The AI Race

Forbes

24-03-2025

Rajat Sharma, Founder and lead Advisor, ngnadvisory.com
The "Fourth Industrial Revolution" is steering us from an API-driven economy to an AI-driven one. While significant investments, predictions and innovations have shaped this transition, the impact has yet to fully meet expectations.
This shift gained momentum when Microsoft invested $10 billion in OpenAI, triggering a wave of investments by hyperscalers into AI infrastructure. These companies positioned AI not only as a co-pilot for software engineering but also as a transformative force in business productivity, customer service and technology operations.
AI is evolving beyond automation—moving toward intelligent agents that can act (automate processes), learn (derive insights from data) and think (determine what, why, when and how to automate).
Agentic AI is emerging as the next paradigm in AI evolution, though its definition varies based on how different companies market their AI-related products and services. This mirrors the early days of cloud computing in the 2010s. Analysts and industry experts predict that agentic AI will revolutionize business operations.
Put simply, agentic AI is interconnected intelligent AI agents autonomously running an enterprise—from corporate functions (supply chain, customer relationships) to IT operations and product engineering—leading to unprecedented productivity, speed to market and profitability (EBITDA).
To harness agentic AI, industries and enterprises must undergo transformation across six key areas:
1. Process Automation: The evolution from robotic process automation (RPA) to intelligent process automation (IPA) has now led to AI-powered agents.
2. Integration: Business connectivity has advanced from complex enterprise integration (EAI) to API-based integration across business-to-employee (B2E), business-to-consumer (B2C) and business-to-business (B2B) ecosystems.
3. Cloud: Cloud providers have shifted from pure-play infrastructure as a service (IaaS) to offering AI as a service, enabled by PaaS, containers and serverless computing.
4. Data: Data management has progressed from legacy flat files and spreadsheets to big data, AI-driven data warehouses (lakehouse) and master data management (MDM), handling both structured and unstructured data.
5. Security: Cybersecurity has evolved from simple malware protection to advanced threat detection and response, securing enterprise-wide operations.
6. Applications: Software architecture has transitioned from mainframes to microservices and APIs, passing through three-tier web applications and service-oriented architecture (SOA).
The AI ecosystem today is largely driven by four major types of players:
1. AI Infrastructure Firms: These include companies like Nvidia (GPUs), integrated CPU/GPU providers, energy suppliers (e.g., Constellation Energy) and enterprise telecom firms (e.g., Lumen) that support AI model training at scale.
2. Hyperscalers: Cloud giants like AWS, Microsoft Azure and Google Cloud Platform (GCP) that offer AI as a service.
3. AI Software Firms: Large SaaS providers integrating AI into their platforms (Salesforce, ServiceNow, SAP) and AI-native firms specializing in big data (Palantir, Databricks, Snowflake).
4. AI Service Providers: IT outsourcing and consulting firms (Accenture, TCS, HCL) pivoting toward AI transformation. Some analysts call them "services as a software" companies.
Additionally, IBM spans multiple categories, offering AI infrastructure, software and services. Recent market shifts, such as disruptions from new AI players like DeepSeek, are also forcing industry leaders to rethink their strategies. Hyperscalers now suggest that agentic AI could disrupt the SaaS market, as enterprises face growing fatigue from rising licensing and consumption costs. This has fueled the rapid adoption of FinOps—a discipline aimed at optimizing and reducing AI-related spending.
Among the four key categories of AI players, I believe AI service providers are best positioned to drive enterprise AI adoption. Their advantage stems from:
• Deep Integration Across The Enterprise: They are embedded in IT operations, engineering and corporate functions.
• Industry-Specific Expertise: Their vertical service strategies are mature and well-defined.
• Platform Agnosticism: Unlike infrastructure providers or hyperscalers, they do not need to invest heavily in AI infrastructure. Instead, they can select and implement the best AI models for each use case.
• Enterprise AI Modernization Leadership: As companies seek to build agentic AI strategies, AI service providers will play a key role in assessment and modernization, something infrastructure providers, hyperscalers and AI software vendors cannot achieve alone.
• Expertise In Agentic AI: Successfully deploying agentic AI requires mastery across six key transformation levers, a capability that resides primarily with AI service providers and system integrators.
While AI service providers lead the charge, other key players—including hyperscalers, infrastructure providers and AI software vendors—can improve their competitive stance by:
• Embracing Open AI Architectures: Moving beyond their proprietary LLMs to recommend and deploy the best-fit AI models for enterprises. AWS Bedrock has already taken steps in this direction, with other hyperscalers following suit.
• Focusing On Business Value Over Consumption Metrics: Instead of prioritizing instance or license consumption, hyperscalers and AI software vendors must shift their focus toward enterprise-wide transformation that drives real business outcomes.
• Strengthening Partnerships With System Integrators: Large system integrators possess deep industry expertise and enterprise-level integration capabilities. Building strong alliances with them will be critical to scaling AI transformation efforts.
The DNA of agentic AI requires expertise across all six transformation levers. In the AI race, I believe the key leaders will be the firms that:
• Pivot quickly to AI transformation.
• Build robust AI capabilities and partner ecosystems.
• Move beyond traditional application development and maintenance (ADMS) and IT/cloud-managed services.
Ultimately, success in AI will belong to those who innovate, integrate and transform—not those who rely on past models of success. The future of AI-powered business will be shaped by those who fully embrace AI-driven enterprise evolution.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?