Lack of expertise, complex data hindering AI adoption in India – Digital Transformation News

As Indian enterprises increasingly adopt artificial intelligence (AI), significant barriers like a lack of expertise, insufficient tools, and data complexity remain. IBM India & South Asia managing director, Sandip Patel, in an interview with Jatin Grover, discusses how small AI models are driving cost efficiency, IBM’s role in chip design, and the evolution of AI adoption in India. Excerpts:

What areas is IBM focusing on in India to drive revenue growth?

IBM is dedicated to helping organisations leverage data and AI for scaling operations, boosting growth, and enhancing competitiveness. Our enterprise-ready AI and data platform, watsonx, integrates seamlessly with Red Hat OpenShift, IBM Infrastructure platforms, and IBM Consulting services to support digital transformation. By delivering integrated hybrid cloud solutions, we are empowering businesses to stay ahead in a competitive market.

IBM’s consulting business has faced global challenges as enterprises cut spending. How does the scenario differ in India?

Contrary to global trends, our consulting business in India continues to grow, thanks to our deep industry, technological, and domain expertise. Indian businesses are actively seeking robust technology solutions, skilled resources, and guidance to execute their digital strategies successfully. IBM’s consulting team is well-positioned to meet these needs by delivering tailored solutions.

Indian businesses ahead of global counterparts in AI adoption: Report

Gen AI fuels surge in cyber threats, advance strategies needed: Palo Alto

Wage ceiling under EPFO to be raised steeply to Rs 30,000/month

India’s path to semiconductor competitiveness: Building a resilient chip manufacturing ecosystem

Compute costs for training AI models are high. How can companies reduce these expenses when building large language models (LLMs)?

Adopting smaller, purpose-built AI models can significantly reduce compute costs while maintaining high performance. For instance, tools like InstructLab (an open-source project by IBM and Red Hat) enable fine-tuning smaller models with proprietary enterprise data. This approach delivers task-specific performance comparable to state-of-the-art large models while achieving cost savings ranging from 3x to 23x, particularly in domains like finance, HR, and customer support.

IBM has strong expertise in semiconductor R&D. How is the company contributing to this ecosystem in India?

IBM’s India-based server development team specialises in designing advanced processors and enterprise-class systems. We collaborate with L&T Semiconductor Technologies to design edge processors and hybrid cloud systems for diverse applications, including mobility, industrial technology, and energy. Moreover, our partnerships with premier institutions like IIT Bombay and IIT Madras focus on workforce development and fostering innovation in semiconductor design.

India’s semiconductor design-linked incentive scheme has faced challenges. What can improve its outcomes?

The design-linked scheme should prioritise incentives for costs related to intellectual property (IP) and design tools. IBM is open to sharing its expertise and serving on committees evaluating applications to enhance product quality. With 20% of the global chip design talent, India is poised to lead in this field. Collaboration between design houses, established semiconductor firms, and academic institutions can drive innovative projects and accelerate local product development.

How are Indian enterprises adopting AI, and what challenges do they face?

IBM’s 2023 Global AI Adoption Index found that 59% of large Indian enterprises have integrated AI into their operations, the highest among surveyed countries. Factors such as improved AI tools and the need for cost reduction drive this adoption. However, barriers like a lack of AI skills, inadequate tools, data complexity, and ethical concerns persist. Many businesses feel unprepared and unsure about deriving tangible value from AI investments.

With generative AI booming, do B2B models offer better return on investment compared to B2C?

Value creation from AI is use-case dependent rather than model-specific. While B2C adoption of generative AI tends to be faster, long-term return on investment for B2B applications is substantial. Enterprises gain multi-fold value by integrating foundation models into their workflows. This allows businesses to tailor models for their specific needs rather than relying on generic consumer-facing solutions.

Do LLMs face challenges with hallucination, and how can this be mitigated?

Hallucinations often arise from biased or unrepresentative training data. To mitigate this, models must be trained on high-quality data, tested rigorously, and equipped with clear limitations. Developing guardrails to prevent adversarial attacks is also crucial. Properly defining an LLM’s purpose and responsibilities helps reduce such vulnerabilities.

How is hybrid cloud adoption evolving in India, and are there data security concerns?

Hybrid cloud is gaining traction as businesses manage distributed applications across public, private, and on-premises environments. However, this expansion increases security risks, including shadow data. IBM’s 2024 Cost of a Data Breach report noted that 29% of data breaches in India involve data stored in multiple environments. To address this, organisations must adopt a hybrid-by-design approach, ensuring secure and agile infrastructure while investing in skilled talent to manage these systems effectively.

_{Images are for reference only.Images and contents gathered automatic from google or 3rd party sources.All rights on the images and contents are with their legal original owners.}