On Artificial Intelligence
Essay in AINOWINSTITUTE titled: A New AI Lexicon: C is for Consent
For Artificial Intelligence or AI to function properly, vast amount of data is necessary and when the data is personal, then the idea of consent, and redundancy of consent have become a part of debates on technology and rights, writes Praavita Kashyap, Delhi-based lawyer and Trustee of Article 21 Trust.
She writes that the world over Governments are looking to the potential of AI and India also released a discussion paper in 2018 through Niti Ayog analysing the critical role that government needs to play in fostering innovation through research, providing access to infrastructure etc. to address the various government needs. She notes, that for building a marketplace surrounding AI, the collection of data becomes legitimised through a process, which she calls “bureaucratisation of consent” making consent meaningless. She explored the imbalance of power in the adoption of new technologies and the use of consent by the State as a regulatory body surrounding the Aadhaar scheme.
She noted that the Aadhaar project that was initiated in 2009 became the biggest project of personal data collection in India and mass enrolment was achieved through coercion and duress. Despite the Supreme Court’s interim orders stating that Aadhaar could not be made mandatory, the state Governments continued it for many welfare programmes, subsidies etc. “When the state owns, implements, and creates a massive data collection project, consent is eroded and turned into a meaningless bureaucratic step which unsuccessfully attempts to provide a cover for coercion,” she lamented.
She concluded by noting that the bureaucratisation of consent in the deployment of state-owned and operated technologies renders the moral and ethical power of consent itself meaningless and even dystopian. It also preys upon the vulnerable section of society and obscures the cost of harvesting data by the state.
Please click here to read the complete essay here
Essay in Quarterly Bar Review titled: Rule Of Law In The Age Of Technology-Assisted State Decision Making:
Writing for the Quarterly Bar Review published by Delhi High Court Bar Association, Prasanna S, Advocate on Record Supreme Court of India and Founding Trustee of Article 21 Trust argues in the essay that the decision-making that is now heavily based on technology or Artificial Intelligence must conform to the norm of the administrative law and the post-constitutional understanding to the ideas of the rule of law and the non-arbitrariness in administrative decision-making.
In several instances, State decision-making has been controlled through judicially evolved norms such as adherence to principles of natural justice, the requirement to act reasonably etc. Unlike the UK, where the rule of law and non-arbitrariness norms developed through law courts, these norms in India, are rooted in the constitution itself and are traced to Article 14 of the Constitution, notes Mr. Prasnna. Similarly, reasonableness and proportionality have been traced to the expressed requirements of Article (clauses) 19(2) through 19 (6) since the early 1950s, he highlighted.
He underlines that the ideal of non-arbitrariness is not merely a wordplay but an essential feature of the condition and a necessary severance from the ways of colonial government. He stressed that the ideals that have informed our constitutional contract and the administrative norms that have emerged and evolved from that constitutional promise, cannot be done away with by merely changing the mechanism by which state decision-making occurs and without a surer legal footing.
To control the state decision-making that relies on technology, Prasanna offers certain measures.
An impact assessment and an inquiry ought to be undertaken for the possible extent of compliance norms before the employment or adoption of technological solutions in decision-making.
The requirement of compliance with the norms must pass through as a contractual obligation to the providers of AI solutions.
A purely non-automated system must be given to persons involved where it is found that the decision-making system is not able to comply with the administrative legal norms.
No adverse presumption ought to be affixed either on the state actor or the decision subject for their failure to use the automated option.
And finally, the principles of the administrative law and rights flowing from there ought to be protected within the broad conception of the right to privacy in the digital age. The right to privacy has expanded from the primitive understanding of the mere right to be left alone to the right against digital automated processing under Article 22 of GDPR.
To read the complete essay published in the Quarterly Bar Review of Delhi High Court Bar Association, click here
To read the complete working paper please click here