The year 2023 has truly been a defining one for artificial intelligence. All the major dictionaries including Cambridge, Oxford, and Merriam-Webster have named artificial intelligence (AI) or AI-related terms as their word of the year. The pace of innovation and adoption of AI in 2023 has been arguably unprecedented. India found itself emerging as a critical piece in the development and deployment of this technology. TechCircle explains what AI meant for India in 2023:
New Delhi Declaration on AI :
Taking over from last year’s host France, India assumed the Chief Chair position of the Global Partnership on Artificial Intelligence (GPAI), hosting the prestigious GPAI Summit 2023 in December. The three-day summit, inaugurated by Prime Minister Narendra Modi, attracted active participation from stakeholders representing 29 countries. During the inauguration, PM Modi announced an AI mission, aimed at fostering AI compute power for startups and innovators.
One of the most significant outcomes of the summit was the adoption of the New Delhi Declaration on AI by the 29 participating countries. This declaration recognises the advancement of AI technology and its role in generating economic growth, innovation, and jobs across sectors, while also acknowledging risks pertaining to misinformation, unemployment, lack of transparency, and a threat to democratic values. “We underscore GPAI’s pivotal role in addressing contemporary AI issues, including generative AI, through applied AI projects aimed at addressing societal problems and global challenges, maximising the benefits and mitigating associated risks,” the declaration noted.
Bletchley Declaration :
In November, India became one of the 30 countries and regions (including EU) to sign the Bletchley Declaration, which is an agreement on AI. The agreement is based on five principles – the need for safe and responsible AI; associated opportunities and risks; commitment to inclusive cooperation; measures for AI safety; and support to international research. India was represented by MoS IT Rajeev Chandrasekhar who said that AI should be guided by safety and trust for users, and accountability for platforms.
Indian IT and generative AI :
The Indian IT and ITeS industry caught on to the AI and generative AI fever this year. All the top five IT services companies announced generative AI practices and solutions, along with a commitment to training a major chunk of their workforce in relevant skills. For instance, Infosys rebranded itself as an AI-first company, TCS carved out a separate business unit focused on AI and Cloud, and Wipro announced a $1 billion in AI investment, and Tech Mahindra launched Generative AI Studio.
As evident from the last quarter's earnings (Q2FY24), companies have been prompt in rolling out pilot programs and proof of concepts (PoC), but the pace of obtaining return on investment from generative AI was sluggish. “The adoption rate is overall very low, but gradually it will increase over time. Further, our analysis shows that by 2025, 80% of enterprise applications such as productivity software, instant messaging software, or business intelligence applications will have generative AI incorporated into them,” Biswajit Maity, senior principal analyst, Gartner told TechCircle in an earlier interaction.
Impact on jobs :
Fintech giant Paytm's parent company One97 Communications axed over 1,000 jobs in December. A company spokesperson told Reuters that job costs happened as the company expects to save up to 15% in employee costs due to AI. Further, Bengaluru-based Dukaan said in June that it laid off 90% of its support staff, replacing them with an AI bot. These examples are indicative of AI’s impact on jobs. IBM’s chief Arvind Krishna said that it plans to replace 8,000 jobs with AI in the next five years. To be sure, IBM India employs 100,000 people. These anticipations have caused apprehensions among Indian workers. A Randstad study in September found that one in two are concerned about impact of AI on jobs.
It is not entirely a lost cause. Over 85% of Indian employers are confident that AI will create new jobs in the next five years, an October study by the job portal Indeed found. Corroborating this sentiment is a World Economic Forum’s May 2023 report which stated that Indian job market would grow by 22% in five years, on the back of AI and machine learning.
Indic AI models :
Shortly before Bhavish Aggarwal filed IPO for Ola Electric, his team launched Krutrim, an Indic language-focused foundational model. While less is known of Krutrim’s technical specifications, the team said that it is trained on 2 trillion tokens and can understand queries in 22 languages and generate responses in 10. Krutrim finds itself in the company of several other India-focused language models and bots, which Sarvam AI’s OpenHathi, Corover.ai’s BharatGPT, and TechMahindra’s Project Indus. In addition, the Indian government launched Bhashini, an initiative for building a national public digital platform for languages using artificial intelligence
AI-based deepfake :
The menace of deepfake intensified this year thanks to AI-powered hyperealistic fake videos. Deepfake videos of public personalities, especially women, spread widely on social media platform. Union Minister for IT Ashwini Vaishnaw called it a threat to democracy, while PM Modi also flagged the issue.
This week, the Ministry of Electronics and Information Technology (MeitY) issued an advisory to all social media platforms. It mandates the platforms to clearly and precisely communicate to users the implications of publishing, sharing and storing deepfakes.
