75% of industry leaders strongly believe responsible AI can significantly improve decision-making and governance – Primus Partners Report

0
376
75% of industry leaders strongly believe responsible AI can significantly improve decision-making and governance – Primus Partners Report
75% of industry leaders strongly believe responsible AI can significantly improve decision-making and governance – Primus Partners Report
  • 73% support establishing a dedicated AI Governance and Ethics Board within the corporate structure to oversee responsible AI practices
  • 78% emphasize the need for ongoing skill development and training to keep pace with AI advancements

 

Bengaluru: Primus Partners today released an influential report titled ‘Responsible by Design: Industry’s Perspective on India’s AI Framework’, presenting a detailed study of the responsible and ethical AI ecosystem in India from the industry lens. The report advocates ethical, transparent, and accountable AI practices while outlining possible enforcement mechanisms for AI regulations in India. 

 

The report offers an in-depth analysis of India’s current AI landscape and focuses on critical aspects of responsible AI development and governance. The study comes at a crucial time as India positions itself at the forefront of global AI innovation while addressing the ethical challenges posed by rapid technological advancement.

 

Devroop Dhar, Co-Founder and Managing Director, Primus Partners, stressed the importance of responsible AI development: “India’s current AI landscape reflects a segment full of opportunities. As we journey towards becoming a trillion-dollar digital economy, we must adopt a balanced approach to AI governance. This involves leveraging AI for its economic benefits while ensuring robust regulatory frameworks that can prevent or curb risks such as deepfakes by promoting ethical standards. As AI’s influence is driving significant advancements and efficiencies across diverse industries worldwide, responsible use of AI should be a shared global priority.” 

 

Based on insights from industry leaders and domain experts across India, the report stresses the potential for AI to contribute up to 10% to India’s GDP by 2025 if governed responsibly. It highlights the challenges of data quality and availability (which include data scarcity, biases in datasets, and the complexities of data preparation and cleaning), noting that 76% of the industry views these as their top concerns in AI deployment. It also emphasizes that 61% of the industry prioritizes privacy and security, including threats posed by deepfakes, as a crucial principle-based element in safe AI deployment.

 

Other key findings from the report include:

  • 53% of respondents believe that AI systems should be designed to be transparent and explainable to ensure trust and accountability
  • Over 78% emphasize the need for ongoing skill development and training to keep pace with AI advancements
  • 51% of surveyed industry leaders prefer a hybrid regulatory approach combining traditional frameworks with adaptive guidelines
  • 47.3% believe that over-regulation and over-compliance if put in place, can become a deterrent to AI’s growth
  • 67.6% of the industry believes that more PPP engagements are required in critical sectors such as health and education
  • 78% of the industry views that accelerating the Ethical AI Ecosystem in India necessitates operational requisites like continuous skill development and training and Structural Realignment in core AI organizations
  • 89% of the industry believes that specific regulations are essential to ensure ethical practices, accountability, and transparency in AI development and deployment

 

The report encompasses strategic actionable recommendations for establishing a robust framework that supports innovation while addressing the ethical challenges posed by AI. It calls for a collaborative approach between the government and the private sector to develop standards and practices that ensure AI benefits all segments of society. Additionally, the report underscores the importance of continuous learning, capacity building, voluntary product iterations and responsible adoption, and educating consumers or end-users in shaping an AI landscape that is both innovative and responsible.