We are ready to experiment – but the OEMs are reluctant We have identified the way in which AI can help us. But our customers, that is, the OEMs, do not want to experiment with a process that has been working well for a long time. Without their support, we wouldn’t be able to tap into AI’s benefits. Availability of the right knowledge can help prevent sticking points Knowledge-sharing is critical for us. If a database can be created that can map issues to potential AI solutions, it can play the role of an expert for us who can diagnose the problem and recommend targeted solutions. Hiring such an expert is crucial but it is in no way financially feasible for us. Workers need to see AI as their ally, not their replacement There is a general reluctance among workers to trust AI systems. They view it as a threat to their jobs. If you can help us show them that AI can assist, not replace, them – by simplifying tasks and allowing them to upskill and remain relevant – we will find it easier to get them onboard. Role models and success stories encourage us, can you show us more? We learn a lot among ourselves. For instance, here in the Peenya Industrial Cluster, many of us owners get together and discuss ways to improve our processes and discuss challenges. In this regard, with a new technology such as AI, where there is general reluctance on how useful it can actually prove, I think if some heroes or role models can be showcased who have been able to improve their processes, that will be very helpful and encouraging. Perspectives from start-ups and industry We can’t jump to AI when most of us haven’t walked the digital basics The challenge with Indian SMEs is that they are very diverse with respect to their AI-readiness. We need to create AI solutions that can be easily integrated into existing systems, especially for those enterprises that are digitally less mature. Many SMEs are still operating with very minimal digital infrastructure. Without these basics in place, it’s going to be very difficult to introduce advanced AI solutions. We will need to meet each SME where it is SMEs need a phased approach to bring them up to speed; we will have to meet each where they are in their digital journey. Currently, the absence of reliable data that can help an enterprise achieve an overall view of the factory floor is a major bottleneck in deploying AI. Our role then is to first help these businesses build a digital foundation so they can fully benefit from AI. Ecosystem power propels small firms For AI to be truly effective, SMEs, OEMs and the government need to collaborate closely. Large companies should assist their smaller counterparts by sharing knowledge and offering digital infrastructure – AI implementation is a shared effort within the ecosystem. We need AI built for Indian realities, not global wallets The problem is that most of the solutions produced in India are developed for deployment abroad because that commands a hefty price tag. We urgently need to find a way to create India- specific solutions that can be offered to Indian enterprises at a reasonable cost. Help SMEs find their target then the right tool When deploying AI solutions, simply saying that you want to improve your process isn’t enough for us. You must know exactly what improvement you are looking for. For instance, you could say you want to reduce your quality rejection rate from 20% to 5% or that you want to reduce the on-belt time from one minute to 30 seconds. We can also help you clearly outline these goals, but in the absence of these, there is no point in moving ahead. Process needs production-ready AI, not prototypes When we work with clients, we usually see more risk aversion among process industries. Experimentation, if gone wrong, can prove to be very costly there – a single mistake could shut down the entire operation for hours. So most of these manufacturers just want to make sure that the current process is running as smoothly as it can. But, let me assure you, the opportunity is equal in both types of industries. Let AI translate export standards while your best minds solve real challenges When you’re dealing with dozens of different export standards, having experts spend time on repeatedly decoding, translating, understanding or synthesizing compliance requirements and related documentation is just wasteful. AI can take all those buyer requirements and convert them directly into the language your teams can understand – let AI handle this routine work and free up your skilled people to tackle the really complex challenges that actually need human expertise. Transforming Small Businesses: An AI Playbook for India’s SMEs 17

Transforming Small Businesses 2025