AI adoption surges as 88% of global firms fund dedicated AI budgets: Nasscom Report Highlights Major Shift in Global AI Strategy

The global business landscape is undergoing a profound transformation, driven by an unprecedented surge in Artificial Intelligence (AI) adoption. A groundbreaking new report from Nasscom, titled ‘Enterprise Experiments with AI Agents – 2025 Global Trends,’ reveals that a staggering 88 percent of global enterprises now boast dedicated budgets for AI initiatives. This significant financial commitment underscores a pivotal shift in corporate strategy: moving beyond initial experimentation with Generative AI (GenAI) towards the development and deployment of sophisticated, goal-oriented systems known as AI agents.

This comprehensive study, drawing insights from over 100 companies across diverse industries and global regions, paints a clear picture of an enterprise world preparing for the next frontier of AI. With nearly two-thirds of these firms allocating more than 15 percent of their total technology budgets to AI projects, the message is clear: AI is no longer a peripheral experiment but a core strategic imperative.

A New Era of AI Investment: Beyond Experimentation

For years, Artificial Intelligence has been a buzzword, but its practical application in the enterprise has often been characterized by cautious pilot programs and exploratory projects. The Nasscom report signals a definitive departure from this tentative approach. The 88% figure for dedicated AI budgets is not just a statistic; it represents a fundamental re-prioritization of resources, reflecting a widespread recognition that AI is indispensable for future competitiveness and operational efficiency.

This substantial investment push is particularly noteworthy given the economic uncertainties that have characterized recent years. It suggests that businesses view AI not as a discretionary spend, but as a critical enabler for resilience, innovation, and growth. The allocation of over 15% of tech budgets by a majority of these firms indicates a deep-seated commitment, moving AI from a departmental experiment to a C-suite priority. This level of funding allows for the development of robust, scalable AI solutions rather than fragmented, siloed initiatives. It also points to a growing maturity in how enterprises perceive AI’s role – from a cost center to a value driver.

The report’s findings align with broader industry trends indicating that AI, particularly advanced forms like Generative AI and now AI agents, is poised to unlock significant productivity gains and create new revenue streams. Companies are no longer just looking to automate simple tasks; they are aiming to augment human capabilities, enhance decision-making, and fundamentally reshape business processes.

Understanding the Shift: From Generative AI to Autonomous Agents

A key highlight of the Nasscom report is the discernible shift in focus from mere Generative AI tools to the more complex and autonomous realm of AI agents. While Generative AI, exemplified by large language models (LLMs) like those powering sophisticated chatbots and content creation tools, has captivated public imagination and corporate interest, the report suggests enterprises are now looking to leverage these foundational technologies for more active, goal-driven applications.

Generative AI primarily focuses on creating new content – text, images, code, or even music – based on learned patterns from vast datasets. It has been instrumental in automating creative tasks, enhancing customer interactions through advanced chatbots, and accelerating content generation. However, Generative AI, by itself, often requires human prompting and oversight to guide its output and ensure its relevance. You can explore more about Generative AI use cases in 2025 across various industries.

AI agents, on the other hand, represent the “next evolution” that Sangeeta Gupta, Senior Vice President and Chief Strategy Officer at Nasscom, aptly describes. Unlike passive analytical tools or generative models that wait for prompts, AI agents are designed to act independently, pursue specific goals, and make decisions within defined parameters. They are essentially autonomous software entities that can perceive their environment, reason about their actions, and execute tasks to achieve a desired outcome, often interacting with other systems or even humans. Think of them as digital employees capable of understanding complex instructions, breaking them down into sub-tasks, and executing them sequentially, learning and adapting along the way. IoT Analytics also notes the rising discussions among CEOs about Agentic AI.

This transition signifies a philosophical shift in how businesses view work, intelligence, and autonomy. It moves from AI as a tool for analysis or content generation to AI as an active participant in operational workflows. For instance, instead of an LLM generating a draft email, an AI agent could be tasked with managing an entire customer support ticket lifecycle, from initial query to resolution, involving multiple steps like data retrieval, communication, and system updates. This demands a higher level of trust, robust data governance, and, crucially, continuous human oversight.

Building the Foundation: Data, Tools, and Specialized Teams

The successful deployment of AI agents is not a plug-and-play operation. The Nasscom report emphasizes that businesses are actively “strengthening their AI foundation” to support this next phase. This involves a multi-pronged approach:

1. Investments in GenAI Tools: While the focus is shifting to agents, the underlying Generative AI capabilities remain crucial. Companies are investing in advanced GenAI platforms and tools that can serve as the “brains” for their agents, providing capabilities like natural language understanding, complex reasoning, and content generation. This includes licensing proprietary models, building custom models, or leveraging open-source frameworks. The choice often depends on the specific use case, data sensitivity, and required level of customization.
2. Robust Data Infrastructure: AI agents are inherently data-hungry. Their ability to perceive, reason, and act depends entirely on access to clean, reliable, and relevant data. Enterprises are therefore pouring resources into upgrading their data infrastructure. This involves:
   • Data Lakes and Warehouses: Consolidating disparate data sources into centralized repositories.
   • Data Governance: Establishing clear policies and procedures for data quality, security, privacy, and access. This is paramount, especially when agents are making decisions based on sensitive information.
   • Real-time Data Pipelines: Ensuring that agents have access to up-to-the-minute information to make timely and accurate decisions.
   • Vector Databases: These are becoming increasingly important for storing and retrieving embeddings (numerical representations of data), which are crucial for LLMs and AI agents to understand context and relationships within data.
3. Flexible Processes: Traditional, rigid business processes can hinder AI adoption. Companies are realizing the need for more agile and adaptable workflows that can accommodate the dynamic nature of AI agents. This involves re-evaluating existing processes, identifying bottlenecks, and redesigning them to integrate AI seamlessly. This often means embracing concepts like process automation, low-code/no-code platforms, and microservices architectures that allow for modular and flexible deployment of AI capabilities.
4. Specialized AI Teams: The report highlights that “many companies have already formed specialised AI teams.” This is a critical organizational development. These teams typically comprise AI engineers, data scientists, machine learning operations (MLOps) specialists, prompt engineers, and AI ethicists. Their role extends beyond just building models; they are responsible for the entire lifecycle of AI systems, from conceptualization and development to deployment, monitoring, and maintenance. The demand for such specialized talent is soaring, leading to intense competition for skilled professionals.
5. Upgrading Tech Setups: Beyond data infrastructure, general IT infrastructure upgrades are essential. This includes investing in more powerful computing resources (GPUs, TPUs), cloud-based AI services, and scalable deployment environments that can handle the computational demands of running multiple AI agents simultaneously. Edge computing is also gaining traction for scenarios where low latency and local processing are critical for agent performance.

The Rise of Agentic AI: Internal Focus, Cautious External Steps

One of the most significant revelations from the Nasscom report is the burgeoning interest in Agentic AI. The study found that a substantial 62 percent of companies are actively experimenting with such AI agents. However, their initial deployment is predominantly focused on internal tasks, where the risks are lower and the benefits can be more easily controlled and measured.

Internal Applications:

• IT Operations: AI agents can automate routine IT tasks like network monitoring, incident response, ticket routing, and system diagnostics. For example, an agent could detect a server anomaly, diagnose the root cause, and even initiate a fix, all with human oversight. This frees up IT staff for more complex problem-solving and strategic initiatives.
• Human Resources (HR): In HR, agents can streamline processes such as onboarding, answering employee FAQs, managing leave requests, and even assisting with recruitment by screening resumes and scheduling interviews. This enhances employee experience and reduces administrative burden.
• Finance: Financial operations can benefit from agents handling tasks like invoice processing, expense reconciliation, fraud detection, and even basic financial reporting. An agent could analyze transaction data to flag suspicious activities or automate the generation of routine financial summaries.

External Applications:

In contrast, external uses of Agentic AI, particularly in areas like customer service, are still limited, with only 31 percent of enterprises using them in these client-facing roles. This caution is understandable. Customer-facing agents require a higher degree of accuracy, empathy, and the ability to handle complex, nuanced interactions. Errors in external-facing agents can directly impact brand reputation and customer satisfaction. However, the potential benefits are immense, including 24/7 availability, faster response times, and personalized customer experiences. As the technology matures and trust builds, this number is expected to rise significantly.

Looking ahead, the report highlights a strong future commitment: a remarkable 88 percent of companies plan to set aside budgets specifically for Agentic AI systems in 2025. This forward-looking investment signals a clear intent to scale these capabilities beyond initial experiments, recognizing their potential to redefine operational paradigms and competitive advantage.

Navigating the Future: Human Oversight and Responsible Scaling

Despite the enthusiasm for Agentic AI, the Nasscom report also reveals a healthy degree of caution among enterprises. The emphasis on responsible scaling is evident in the prevailing deployment model:

• Human-in-the-Loop (HITL): Around 77 percent of companies are designing Agentic AI systems with a ‘human-in-the-loop’ model. This approach ensures that critical decisions or actions taken by AI agents are reviewed, validated, or overridden by human operators. This is crucial for maintaining control, ensuring ethical behavior, and adapting to unforeseen circumstances. For example, an AI agent might propose a solution to a customer issue, but a human agent would approve and send the final response. This model builds trust, allows for continuous learning and refinement of the AI, and mitigates risks associated with fully autonomous systems.
• Fully Autonomous Agents: Only 46 percent of surveyed companies are currently testing fully autonomous agents. These are systems that operate without direct human intervention once deployed, making decisions and taking actions independently. While offering the highest levels of efficiency and speed, they also carry the highest risks, particularly concerning accountability, bias propagation, and unintended consequences. The cautious approach here reflects a recognition of these complexities and the need for robust testing, fail-safes, and clear ethical guidelines before widespread deployment.

Sangeeta Gupta’s emphasis on “strong trust, data readiness, and continuous human oversight” resonates deeply with these findings. Building trust in AI agents requires transparency in their operation, explainability of their decisions, and a clear understanding of their limitations. Data readiness, as discussed, ensures that agents have access to high-quality, unbiased data. Continuous human oversight serves as the ultimate safeguard, allowing for intervention when necessary and ensuring that AI systems remain aligned with organizational values and objectives.

Industry Leadership: Manufacturing Paves the Way

The report identifies manufacturing companies as being particularly advanced in their adoption of AI agents. This sector’s inherent need for precision, efficiency, and automation makes it a fertile ground for AI agent deployment. Examples include:

• Robotics and Automation: AI agents can control and optimize robotic systems on assembly lines, enabling more flexible manufacturing, predictive maintenance, and quality control.
• Quality Control: Agents equipped with computer vision can inspect products for defects with higher accuracy and speed than human inspectors, reducing waste and improving product quality.
• Supply Chain Optimization: Agents can analyze vast amounts of data to predict demand, optimize logistics, manage inventory, and respond dynamically to supply chain disruptions.
• Predictive Maintenance: By monitoring sensor data from machinery, AI agents can predict equipment failures before they occur, scheduling maintenance proactively and minimizing downtime.

While manufacturing leads, other industries are also making strides. Healthcare is exploring agents for diagnostic assistance and personalized treatment plans, finance for algorithmic trading and risk assessment, and retail for personalized recommendations and inventory management. The specific applications vary, but the underlying principle of goal-oriented, autonomous action remains consistent.

The Broader Implications: Ethics, Regulation, and the Future Workforce

The rise of AI agents brings with it a host of broader implications that enterprises must navigate carefully.

• Ethical AI: As AI agents gain more autonomy, ethical considerations become paramount. Questions around accountability for decisions made by agents, potential biases embedded in their training data, and the transparency of their operations are critical. Companies are increasingly investing in AI ethics frameworks, responsible AI principles, and diverse AI development teams to mitigate these risks. The concept of “AI safety” – ensuring AI systems do not cause unintended harm – is gaining significant traction.
• Regulatory Environment: Governments worldwide are grappling with how to regulate AI. The European Union’s AI Act, for example, categorizes AI systems by risk level and imposes strict requirements on high-risk applications. While full enforcement begins in August 2026, certain provisions take effect earlier, starting in February 2025. Similar legislative efforts are underway in the United States, the UK, and other regions. Enterprises adopting AI agents must stay abreast of these evolving regulations to ensure compliance and avoid legal pitfalls. This regulatory landscape will undoubtedly influence the pace and nature of AI agent deployment, particularly for external-facing or high-impact applications.
• Impact on the Workforce: The deployment of AI agents will inevitably reshape job roles and require a re-evaluation of human-AI collaboration. While some routine tasks may be automated, new roles will emerge in AI development, oversight, maintenance, and strategic application. The emphasis will shift from repetitive tasks to skills like critical thinking, creativity, problem-solving, and emotional intelligence. Companies will need to invest heavily in upskilling and reskilling their workforce to prepare for this future, fostering a culture of continuous learning and adaptability. The “philosophical shifts in how we view work” that Sangeeta Gupta mentioned are not just theoretical; they will manifest in tangible changes to organizational structures and talent strategies.
• Competitive Advantage: Early and effective adoption of AI agents can confer a significant competitive advantage. Businesses that successfully integrate these systems will likely see enhanced operational efficiency, faster time-to-market for new products and services, superior customer experiences, and more agile responses to market changes. This creates a strong incentive for investment, driving the trends observed in the Nasscom report.

Conclusion: A Transformative Path Ahead

The ‘Enterprise Experiments with AI Agents – 2025 Global Trends‘ report from Nasscom paints a compelling picture of an enterprise world on the cusp of a major AI-driven transformation. The substantial increase in dedicated AI budgets and the strategic pivot towards AI agents signal a mature and aggressive approach to leveraging artificial intelligence.

While the journey towards fully autonomous AI agents is being undertaken with necessary caution, particularly with a strong emphasis on human oversight, the direction is clear. Businesses are investing in the foundational elements – data infrastructure, GenAI tools, and specialized teams – to support this next wave of innovation. The anticipated benefits of faster decision-making, improved market responsiveness, and enhanced operational efficiency are powerful drivers.

As AI agents become more sophisticated and pervasive, they will not only automate tasks but also fundamentally redefine how businesses operate, interact with customers, and compete in the global marketplace. The coming years will undoubtedly witness an acceleration in the deployment of these intelligent systems, making responsible innovation, ethical considerations, and continuous learning paramount for enterprises navigating this transformative path. The future of enterprise is increasingly agentic, and the groundwork is being laid today.

Ready to take your career to the next level? Join our dynamic courses: ACCA, HESI A2, ATI TEAS 7 , HESI EXIT  , NCLEX – RN and NCLEX – PN, Financial Literacy!🌟 Dive into a world of opportunities and empower yourself for success. Explore more at Serrari Ed and start your exciting journey today! ✨

photo source: Google

By: Montel Kamau

Serrari Financial Analyst

1st July, 2025