The rapid evolution of Artificial Intelligence (AI) has fundamentally transformed how businesses operate, making it an imperative for organizations to adopt and integrate these technologies into their workflows. In an exclusive discussion with Tech Achieve Media, Sandeep Gupta, Managing Director at Protiviti Member Firm in India, highlights why businesses can no longer afford to delay their AI journey. With boardrooms increasingly prioritizing AI strategies and business leaders from diverse functions advocating for AI adoption, the shift towards an “AI-first” approach is accelerating. Gupta explores the drivers behind this transformation, the challenges of ROI, and the lessons organizations can glean from early adopters to future-proof their investments in AI. He also shared exclusive insights into the recently released Protiviti India and CII report titled “AI Trends and Future Impact: Industry Adoption and Insights”, in which he played an instrumental role.
TAM: Like many have cloud first approach while deciding Infra now, what is the probability leaders will have AI first approach to solve problems? Or will companies hold back due to the ROI aspect?
Sandeep Gupta: A year or two ago, there was significant hesitancy about adopting AI. However, last year completely changed the game. Today, boardrooms are saying, “We’ve experienced AI in our personal lives, and it’s time to integrate it into our businesses.” Peer pressure also plays a role—companies are observing their competitors adopting AI and realizing that if they don’t follow suit or prioritize technology, they risk falling behind. No one wants to become the next Xerox or Kodak—companies that failed to adapt in time.
Also read – Embracing the AI Revolution: Sachin Tayal, MD, Protiviti
Market forces are driving this transformation, and businesses now understand the realities of AI adoption. They’re actively building governance structures around it to ensure a systematic approach. What’s notable is that technology adoption has been democratized. It’s no longer just the CIO driving these initiatives. Now, business leaders—such as Chief Marketing Officers or HR heads—are pushing for AI integration. They’re saying, “We need this.” And if their IT teams don’t act, employees themselves are finding ways to implement AI tools to enhance workflows.
AI has become pervasive—it’s reaching every desktop and becoming integral to modern business operations. Take automobiles, for instance: advanced driver-assistance systems (ADAS) are now a standard feature in smart cars. Similarly, enterprise applications like SAP, Microsoft, Oracle, and CRM systems are embedding AI capabilities. Companies that fail to leverage these technologies risk losing out.
Even the most traditional organizations are exploring how to enable their workforce with AI. The focus is on boosting productivity and achieving a multiplier effect on business revenues—not just cost savings. While AI adoption often starts with cost optimization, it quickly evolves into a driver of growth. Limited resources can now be maximized to enhance throughput, visibility, and overall business outcomes. Ignoring AI is no longer an option, and boardroom discussions reflect this shift. Even the largest and most established corporations are embracing AI to remain competitive and future-ready.
TAM: As AI evolves, what strategies can organisations adopt to future-proof AI investments?
Sandeep Gupta: Organizations today manage both structured and unstructured data within their enterprise applications. They are utilizing tools like OCR (Optical Character Recognition) to process scanned documents and digitize vast libraries of information. Beyond this, there is a wealth of data generated from shop floors and other ecosystems where significant investments have already been made. The next logical step is adding an AI layer to enhance and leverage these existing systems.
Also read: India Stands at the Cusp of AI-driven Transformation – Protiviti and CII Report
This approach represents an interim phase, likely to dominate over the next four to five years. Companies looking to adopt AI now will invest in such technologies. However, in the long run, ecosystems will come equipped with their own integrated AI capabilities. This means businesses won’t need to reinvent the wheel—most advancements will arrive through software upgrades.
For example, SAP and Microsoft are already embedding AI into their platforms. Microsoft’s Copilot and agent-based solutions are now a reality, with licenses being offered to users to initiate adoption seamlessly. These innovations make AI adoption accessible and relatively straightforward, signaling that it’s no longer a question of “if” but “when” organizations will embrace these advancements.
TAM: What lessons can organizations learn from early adopters of AI?
Sandeep Gupta: Other businesses are learning from these trends. For instance, some customers who initially shifted to digital banks are now returning to legacy banks. Meanwhile, traditional brick-and-mortar banks are striving to catch up with the digital banks.
Digital banks had the advantage of starting from scratch, enabling them to build their ecosystems around cutting-edge technologies like AI. In contrast, legacy banks are now working hard to match their agility—whether in reducing customer acquisition costs or transforming processes like loan approvals. For example, moving from a lead to loan approval within 10 minutes to an hour has become a key focus. This transformation is becoming increasingly widespread across industries.
It’s not just banking—this shift is happening in every sector. Take media and journalism, for example. I was speaking to an editor who shared that journalists today don’t necessarily need to excel at writing long-form content. Instead, they can focus on capturing key points, while AI tools handle the task of drafting detailed reports. This evolution is driven by the need to save time and process large volumes of information.
The same applies to research in sectors like BFSI (Banking, Financial Services, and Insurance). One of our BFSI clients highlighted the importance of timely research reports. If a report doesn’t reach investors’ desks first thing in the morning, it loses its value. That’s where AI steps in, ensuring information is processed and delivered quickly and efficiently.
TAM: Do you think leaders can ignore adoption of AI or GenAI? How will it affect their business according to you?
Sandeep Gupta: Generative AI is here to stay—simple as that. It’s not a passing fad or something people are exploring just out of fear of missing out. Instead, it’s finding meaningful use cases and becoming a cornerstone of technological adoption.
In fact, its adoption is likely to grow even further as the underlying technologies continue to evolve. The main challenge for generative AI initially was trust. People questioned its reliability, likening it to Wikipedia, where the source of information could be uncertain. However, that concern is being addressed.
Organizations are now focusing on controlling their data inputs by using trusted sources of information. They’re ensuring that generative AI processes only high-quality data from verified data warehouses and data lakes. This shift builds the trust that was previously missing and is making generative AI much more practical for businesses.
Previously, many assumed generative AI would function like ChatGPT, producing results from random or generic sources. Now, the technology is being integrated into organizational ecosystems. This tailored approach allows businesses to fully leverage generative AI while maintaining control over their data, ensuring trust and compliance within their systems.
TAM: The DPDP Draft Rules are out, and the law is soon going to be implemented. Do you see this having an impact on GenAI and its potential?
Sandeep Gupta: I believe the DPDP (Digital Personal Data Protection) Act is also enabling the use of AI. Why? Because it ties directly to trust—specifically trust in how customer data is handled. For instance, if customer data is ingested into a large language model (LLM) or any AI system, the question arises: Can that data be erased if the customer requests it, as mandated by the DPDP Act?
To address this, organizations are now investing in technologies that anonymize and summarize data. For example, if I want to analyze how many males or females are in a specific retail segment or other demographic information like race or geography, I can work with summarized and anonymized data that doesn’t link back to any individual. This is achieved through advanced encryption techniques and data-cleansing processes.
The data that feeds into AI models needs to go through a cycle of cleansing and refinement. Not all data is necessary—things like mobile numbers or email addresses can often be excluded. Instead, organizations focus on extracting preferences, geographic details, and similar insights.
If mechanisms like these are implemented alongside the DPDP’s legal framework, organizations will have greater trust in using AI. This trust will allow them to leverage data more effectively without compromising privacy. At the same time, companies must ensure compliance not just to meet legal requirements but also to use data responsibly and securely.
The DPDP report outlines these requirements in detail, including both compliance mandates and the technical safeguards organizations need to adopt. It’s not just about complying with the law; it’s about building systems that respect privacy while enabling the transformative power of AI.
TAM: What advice would you give to businesses navigating the complexities of AI adoption?
Sandeep Gupta: Let me give you a quick deep dive into how organizations should approach AI adoption. The first step is for the board and top leadership to establish a clear understanding of what specific challenges or goals they want AI to address. This involves identifying targeted use cases—whether it’s improving efficiency, maximizing revenue, enhancing production, or other priorities. Organizations need to focus on these objectives rather than taking a broad, unfocused approach.
Once the focus is clear, the next step is choosing the right technology platform—whether it’s machine learning, generative AI, or another solution. Stakeholder management is equally critical. As Seema pointed out earlier, there’s often concern at the ground level about how AI adoption might impact jobs. It’s essential to take employees into confidence, explaining how this transformation will change their day-to-day roles and potentially create new opportunities. Addressing concerns about job displacement upfront is key.
Governance is another crucial aspect. Any AI adoption must include a strong governance framework to oversee security, risk management, regulatory compliance (such as the DPDP Act), and adherence to ethical standards. This ensures that the organization’s AI strategy is both effective and responsible.
Organizations should start with a pilot program, following a structured approach similar to the software development lifecycle. This includes initial testing, reinforcement learning, and continuous improvement, as AI models won’t provide perfect results from the start. Human intelligence will still play a vital role in fine-tuning these systems, especially during the early stages.
This entire process is a journey, and there’s a right way to go about it. Unfortunately, many organizations approach AI in a fragmented manner—different departments experimenting independently without cohesive planning or governance. For example, HR might implement AI in one area, while another department does something entirely different. This lack of coordination leads to inefficiencies and missed opportunities.
To succeed, organizations must prioritize governance and ensure all AI efforts are aligned under a unified strategy. Only then can they fully realize AI’s potential.
