Assistive AI can help in augmenting the decision making process in healthcare, agriculture, security and education
The role artificial intelligence (AI) is playing in transforming the business and the industry is beginning to be understood and experienced by us as consumers and providers of products and services. Today, there are several initiatives such as Smart Cities to take advantage of AI for the societal good. The question often debated in this context is whether autonomous AI is feasible and what is the right option. Even businesses are not yet ready to let AI carry out independent analysis from multiple sources and completely take over the key operations.
Therefore, while AI has the potential to lend support for the agenda of transformation of the society in multifarious ways, as of now, it is the assistive or augmentative dimension of AI which would be most ideal for making a difference to the citizens. Through a combination of ways such as analysis of voluminous data in a reasonable time to evaluate and present options, queries to arrive at answers to their problems, measuring potential outcomes and impact at various stages and across various projects, assistive AI could help in augmenting the decision making process in various fields—healthcare, agriculture, security, education and water, to name a few.
In the context of agriculture for instance, AI should be capable of making recommendations on cultivation of crops based on the combination of data pertaining to farm sizes, soil quality and costs of fertilisers along with the dynamic data of market prices, logistics, inventory, storage and weather patterns. Hence in order for AI to be able to live up to its promise, it is important to work towards building AI decision models built around the ability to draw upon the data from several applications and systems.
In the context of education, moving away from the focus on mass education to mass but customised attention based on individual needs would be most relevant for country like India in order to make learning- teaching process effective. Student centric learning can be supported by AI models with adaptive assessment techniques, customised learning pathways, predictive analytics of student performance and content aligned with learning styles of students. Similarly, appropriate counselling advice regarding career pathways could be provided to the youth based on algorithms that can indicate options around capabilities, interests, job opportunities, skill gaps for aspirational roles and learning efforts.
There is a need for employment exchanges to not only build databases to profile candidates and their background in order to share with them the ongoing information of jobs available from the formal sector but they should also be in a position to take advantage of the increasing demand for gig jobs. By building AI models, they should be able to tap into the job markets and through predictive and prescriptive analytics, present the potential for new sets of skills in a dynamic fashion, connect with the ecosystem for micro skilling as well as with the worldwide opportunities for such skills.
In order for AI to deliver the best social good, it is extremely important to move away from static IT applications and analytics derived from specific trained models to dynamic models that interact with several systems and datasets emanating from multiple sources working together to provide recommendations of higher order. AI can deliver, only if there is clarity on what we want it to deliver.
Originally appeared in Financial Express