By SP Kochhar
Artificial intelligence, or AI, is dominating discussions today in both tech and non-tech forums alike. The customer-facing application, generative AI (GenAI), has especially become a buzzword, drawing attention and leading to policy discussions and regulatory actions across the world. While these are progressive developments and definitely required to propel India’s aspirations for leadership in yet another technology vertical, there are other integral parts to this AI ecosystem which seem to be missing in conversations but need to be focused upon.
While GenAI seems to be the area of focus for all today, its efficacy is contingent upon the entirety of the ecosystem that facilitates its operations. The crux of GenAI lies in its ability to parse through data, organised in specific templates within data bins, to generate meaningful outputs. The collection of data based on criteria and sorting them into specific bins is what I mean by a ‘templated’ approach in this regard. Without such a structured input process, these data repositories risk becoming mere data dumps, significantly diminishing the quality of AI-generated responses. This underscores the critical role of embedded AI systems in preprocessing data to ensure its utility.
Also Read Inside track by Coomi Kapoor: Straight shooting The MV Ruen Episode: Payoffs from Investing in Naval and Air Power The elephant in RBI’s room Ringside view by Tushar Bhaduri: Mumbai Indians captain Hardik Pandya getting booed at home shows IPL fans want their voice heard Also Read
Google may charge you money to use AI based search soon; here’s what we know so far
Understanding an AI system’s workings elucidates the importance of templated data management. Data, in its raw form, is akin to unmined gold — valuable but not immediately useful. The various sectors/sources produce large data sets, which essentially is ‘raw data’. The embedded AI systems take the requirements from the GenAI system at one end and sift through raw data to make templatised trend-based data sets, which are then used as inputs for GenAI to produce the required results. This is akin to producing intelligible information or ‘knowledge’ by embedded AI systems. Thereafter, the GenAI system absorbs this data, analysing it to read the trends and inferring from these trends to produce results, or what can be termed as ‘wisdom’, which cater to the requirements of the users.
The template is a critical part of this process, and needs to be formulated by the government to ensure uniformity of the data being collected. It would also help provide a semblance of sanity and sanitation checks on the diverse data being collected from across multiple sectors and from multiple sources. Unless a specified template is implemented for the purpose, owing to the diversity of sources, the data collected may be fragmented or cluttered, which would constrain the production of fruitful outcomes from the eventual GenAI application used.
It is equally important to understand that the analysis and subsequent outcomes produced by the GenAI depend fundamentally on the nature of the data present in the system. Relying on data sets tailored to the nuances of the Indian environment is crucial for AI systems to generate accurate and relevant outcomes. The diversity of India’s geography, languages, consumer behaviour and industrial landscape demands a localised approach to data compilation. This necessitates governmental oversight to ensure the representation of India’s unique attributes in AI anonymised data sets, a cornerstone in leveraging AI technology for national advancement by making these data sets available for experiment and utilisation without impinging on privacy and data security.
For example, consider the scenario of implementing a national digital education initiative designed to cater to the diverse needs of India’s student population, aiming to address various educational requirements across regions, accounting for local languages, varying curricula and access to resources. For the initiative to be effective, it is crucial to gather and organise detailed information about these variables and feed them into an embedded AI system. The system then analyses the data, identifies patterns, and passes them on to GenAI, thus allowing the latter to tailor educational content and delivery methods to meet the specific needs of different student groups. If the system were to rely solely on global education models/data, it might not fully capture the unique educational landscapes of India’s regions, potentially leading to less effective or irrelevant content. Localised data ensures the AI system can generate more accurate and impactful solutions, showcasing the importance of tailored AI applications in national development projects.
India’s ambition to become a digital economy leader necessitates a robust foundation in AI technologies. As we race forward on the digital highway, it’s crucial to not just celebrate the milestones but also lay down the tracks — by focusing on the underlying AI ecosystem’s components. Ensuring the comprehensive development of AI technologies, backed by localised, structured data sets, is paramount for India’s journey towards technological pre-eminence. This strategic focus will not only fortify our position in the global tech arena but also ensure the sustainable and inclusive growth of our nation in the digital age.
SP Kochhar, Director-general, COAI (Cellular Operators Association of India). Views are personal
By SP Kochhar
Artificial intelligence, or AI, is dominating discussions today in both tech and non-tech forums alike. The customer-facing application, generative AI (GenAI), has especially become a buzzword, drawing attention and leading to policy discussions and regulatory actions across the world. While these are progressive developments and definitely required to propel India’s aspirations for leadership in yet another technology vertical, there are other integral parts to this AI ecosystem which seem to be missing in conversations but need to be focused upon.
While GenAI seems to be the area of focus for all today, its efficacy is contingent upon the entirety of the ecosystem that facilitates its operations. The crux of GenAI lies in its ability to parse through data, organised in specific templates within data bins, to generate meaningful outputs. The collection of data based on criteria and sorting them into specific bins is what I mean by a ‘templated’ approach in this regard. Without such a structured input process, these data repositories risk becoming mere data dumps, significantly diminishing the quality of AI-generated responses. This underscores the critical role of embedded AI systems in preprocessing data to ensure its utility.
Understanding an AI system’s workings elucidates the importance of templated data management. Data, in its raw form, is akin to unmined gold — valuable but not immediately useful. The various sectors/sources produce large data sets, which essentially is ‘raw data’. The embedded AI systems take the requirements from the GenAI system at one end and sift through raw data to make templatised trend-based data sets, which are then used as inputs for GenAI to produce the required results. This is akin to producing intelligible information or ‘knowledge’ by embedded AI systems. Thereafter, the GenAI system absorbs this data, analysing it to read the trends and inferring from these trends to produce results, or what can be termed as ‘wisdom’, which cater to the requirements of the users.
The template is a critical part of this process, and needs to be formulated by the government to ensure uniformity of the data being collected. It would also help provide a semblance of sanity and sanitation checks on the diverse data being collected from across multiple sectors and from multiple sources. Unless a specified template is implemented for the purpose, owing to the diversity of sources, the data collected may be fragmented or cluttered, which would constrain the production of fruitful outcomes from the eventual GenAI application used.
It is equally important to understand that the analysis and subsequent outcomes produced by the GenAI depend fundamentally on the nature of the data present in the system. Relying on data sets tailored to the nuances of the Indian environment is crucial for AI systems to generate accurate and relevant outcomes. The diversity of India’s geography, languages, consumer behaviour and industrial landscape demands a localised approach to data compilation. This necessitates governmental oversight to ensure the representation of India’s unique attributes in AI anonymised data sets, a cornerstone in leveraging AI technology for national advancement by making these data sets available for experiment and utilisation without impinging on privacy and data security.
For example, consider the scenario of implementing a national digital education initiative designed to cater to the diverse needs of India’s student population, aiming to address various educational requirements across regions, accounting for local languages, varying curricula and access to resources. For the initiative to be effective, it is crucial to gather and organise detailed information about these variables and feed them into an embedded AI system. The system then analyses the data, identifies patterns, and passes them on to GenAI, thus allowing the latter to tailor educational content and delivery methods to meet the specific needs of different student groups. If the system were to rely solely on global education models/data, it might not fully capture the unique educational landscapes of India’s regions, potentially leading to less effective or irrelevant content. Localised data ensures the AI system can generate more accurate and impactful solutions, showcasing the importance of tailored AI applications in national development projects.
India’s ambition to become a digital economy leader necessitates a robust foundation in AI technologies. As we race forward on the digital highway, it’s crucial to not just celebrate the milestones but also lay down the tracks — by focusing on the underlying AI ecosystem’s components. Ensuring the comprehensive development of AI technologies, backed by localised, structured data sets, is paramount for India’s journey towards technological pre-eminence. This strategic focus will not only fortify our position in the global tech arena but also ensure the sustainable and inclusive growth of our nation in the digital age.
SP Kochhar, Director-general, COAI (Cellular Operators Association of India). Views are personal
Get live Share Market updates, Stock Market Quotes, and the latest India News and business news on Financial Express. Download the Financial Express App for the latest finance news.
login
who are we?
contact us
qosheapp
Toi Staff
Gideon Levy
Belen Fernandez
Rami G Khouri
Mort Laitner
Donald Low
Ali Fathollah-Nejad
Nikkei Editorial
Michael Kwet
visit website
