As with large corporations, small and self-funded organisations are also migrating to AI and cloud. However, there are challenges to be met.
It’s a known fact that many big companies have leveraged artificial intelligence (AI) solutions and enjoyed a significant return on investment (ROI). The power of automating and streamlining processes is so strong that AI strategies are increasingly aligned with business goals. Customer satisfaction, minimal errors and better performance levels are some of its outcomes. Cloud migration has also helped cut costs.
There appears to be the notion that only large corporations need to automate processes and that smaller ones needn’t necessarily adopt frontier technologies. To that effect, many fledgling companies treat AI with skepticism. Moreover, many of them don’t have the wherewithal or skill sets to handle AI-cloud solutions.
Other hurdles present themselves. The cost of scaling-up operations to adopt AI solutions tends to pull people down. This fear is reinforced by the fact that Covid-19 has put a financial strain on many balance sheets. “Holdbacks could also come in the form of siloed functioning. Then there’s a lack of skill sets to handle AI-cloud-data based solutions. This is likely to impact the future prospects of professionals,” said Rishu Sharma, associate research director-cloud and AI, IDC India, speaking at the recent Nasscom ‘XperienceAI Virtual Summit 2021’.
AI makes for a viable business model, but self-funded organisations as well as beginners are still trying to figure out the possibilities of its adoption. As of now, they seem to make do with robotic process automation (RPA).
Probably one of the means of making everyone familiar with AI is to open up the market for AI knowledge-based jobs. Such a market could be created through a bottom-up approach. Technical experts can be roped in to fine-tune the skill sets of the workforce. It may take well over one fiscal year to train the entire workforce. Technical educational institutions may chalk out a curriculum to make aspirants ready for jobs such as data scientist, business intelligence developer or machine learning engineer, among others. Building human capital could be a top priority to address requirements such as data engineering and security.
Another thought that comes to mind is that of collaboration between organisations and state governments to establish ‘Centres of Excellence’ (CoE). A data matrix could be built at such a centre, which could then be leveraged to understand the requirements of the vendors or partners. Based on this, researchers could be encouraged to develop APIs or Application Programming Interface. Deep-learning algorithms can be incorporated as a value-add to the process. CoEs may possibly package options such that they are in line with the current requirements or else the technology can become obsolete.
Once the groundwork is done, AI could become accessible to smaller organisations as well. That’s because AI packages machine learning and voice-based services. AI chatbots can perform mundane tasks as well as scheduling appointments. Predictive analysis helps in lowering production errors and the cost of production is a common concern for any organisation, regardless of size.
Another perceived need for AI migration across organisations is because of the increasing proliferation of intelligent devices. AI solutions, APIs and DL will form the crux of intelligent devices.
“We believe that the future of intelligent devices is crucial. Within the resilient framework, a support system as well as automated decisions needs to be factored in. Auto-augmenting and auto-deploying at scale will be the future of smart factories,” observed Sharma. It could mean that right from design to execution, everything will be automated. This will give scope for corrections as well as any number of trial and error processes based on AI algorithms. In short, AI is being used as an enterprise tool. If the company has a limited budget it can be included in enterprise software programmes. They can make a beginning by migrating to the cloud.
To begin with, AI can happen on a budget. Analytics dashboards and chatbots can become platform options. In some situations, reusable code can also be a choice. AI software can be integrated at the operator’s level in a user-friendly manner. Perhaps it can be used to get something as simple as customer feedback before being integrated across the tech stack. AI solutions can be repurposed or tweaked depending on the user requirements. Consequently, organisations both big and small may consolidate the best AI practices.
AI can make work more productive and reduce repetitive tasks. In this respect, AI has the potential to be a transformer. It urges legacy brands to shed old practices and shift to AI tools. Along with AI, organisations need data scientists. Data readiness becomes realigned as part of AI strategy. The human-machine interface requires a software platform and data readiness.
Going back in time, from the 1950s, technology purists have felt that AI will solve all human problems. Several futuristic projections did the rounds. But it has not been the case so far. We will have to wait and watch what’s in store for the future.
Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.