The first step is to enact robust data protection
It is welcome that India is hosting a global summit on artificial intelligence (AI) and that the Prime Minister has addressed the gathering and expressed commitment at the highest level of government to wholesome development and regulation of AI. AI will fast become not just a major component of economic competitiveness but also a force multiplier in strategic capacity. It also poses serious challenges in itself and in the way it is put to use. Therefore, control and regulation of AI are global concerns of mounting importance, on which the G20 grouping of the world’s 20 largest economies have adopted guidelines and principles.
For India to offer something more than lip service to developing AI, the first thing to do is to put in place a robust data protection framework. Data is oxygen for AI and how data is used to train AI has implications for the data subjects whose data is utilised for the purpose and for the kind of algorithm that is produced. In the US, racial bias has been built into facial recognition software that makes use of AI. That embarrassment has led to some principles being formulated for AI development, as well. Transparency and explainability, for example. If someone adversely affected by AI decisions wants to challenge a decision, the AI in use must be able to explain how and why it reached the conclusion it did. Robustness, safety and security must be ensured, for which traceability of the data sets used for creating or training the algorithms involved is essential. Accountability is another principle — AI actors must be accountable for the proper functioning of AI. Regulation of AI and of algorithms must emerge as a robust and active field of study and practice in India.
The Prime Minister said that AI should not be weaponised in the hands of non-State actors. How to translate this fine sentiment into action is the question. The Global Partnership on Artificial Intelligence excludes China, whose labs and companies operate at the cutting edge of AI. That makes global coordination to keep AI safe rather tough.
This piece appeared as an editorial opinion in the print edition of The Economic Times.