Today, India finds itself standing on a diabolical point towards two roads, one leading to its fall  against Artificial Intelligence(AI) and another leading to its sustainable coexistence with the  changing times. In the status quo, India lacks a robust legal framework to oversee how  algorithms make decisions and to match the unparalleled growth of AI in the last decade. It is  pertinent to understand that “technology should empower people, not entrench them”. 

The issue that arises is the security of data of such a populous country with people from diverse  socio-economic and religious backgrounds. 

Al and accountability: The two sides of the same coin 

Al is increasingly used in today’s time leading to critics state that how it is a mirror that  mirrors that imperfections and biases of the society. Most AI systems work as "black box"  where the decisions taken by AI are not known even to the creators of the technology. This  risks security and accountability in a world were times are fast, steady and complex. The 

black box problem often becomes the reason of decisions taken which negatively affects  people from diverse backgrounds. 

What the systems are fed through public and private key infrastructure is what they perceive  and apply. They are embedded in algorithms and often are deep rooted with human and gender  biases. An AI fed algorithm would carry this same notion forward but in hidden and automated  manner. This invisibility makes decisions harder to detect and much substantial to comprehend. 

As, AI becomes extremely important in different fields to apply and boost the economic situations establishing globalized ties along with increased efficiency and putting no limit to its  danger in India. Where predominantly caste, class, religion, and gender disparities are already heavily pertinent. Without accountability, algorithmic connections could often reach out if hand and can infringe on the rights of public related to expression and privacy. Unregulated data often  marginalizes and deepens the word based on its pre-decided history and fed data that could  possibly erode faith in public systems and digitization. 

Certain global practices can help in providing a breakthrough, for AI legal jurisprudence in India  especially the Algorithmic impact assessments adopted by the Organization for Economic Co operation and wherein there is accountability prospect involved through the Federal Trade  commission. India can draw a parallel from the European Unions “AI Act, 2024” which is comprehensive regulation providing different AI stakeholders a limit to which  information can be procured and it divides the procurement through government and private  stakeholders. 

AI regulation India relies on existing judgements, guidelines, and advisories from various  regulatory bodies and courts. Recently the Supreme court of India flagged lengthy pleadings &  use of ai-generated statements in courtroom arguments and asked differential courts across India 

to strike out unnecessary pleadings in civil suits and use of computer generated statements. Furthermore, an issue regarding AI generated avatar’s came into hurdle in India and the courts  down poured upon the mis-use and mala-fide intention in respect to deepfakes. In this case AI  was used to create generative images and deepfakes of famous celebrities to sell merchandise,  false courses using fake enforcement by deceiving people.

Justice Prathiba M. Singh of Delhi High court said in her order: 

“The Court cannot turn a blind eye to such misuse of a personality's name and other  elements of his persona.” This order showcases the need for laws to adapt and evolve with the  emerging challenges caused by AI use and abuse. 

India can ensure AI is safe and equitable for all, rather than merely addressing issues after they  arise:  

First and foremost, there is a requirement for explicit guidelines for the application of Al,  particularly when it comes to making critical human decisions like loan or employment  decisions. According to these guidelines, AI must be transparent; we must comprehend its  decision-making process rather than relying solely on a "black box" to provide us with an  answer. 

Second, we must constantly monitor AI for the problem of bias. This entails ensuring AI doesn't  unfairly treat some groups of people worse than others, which can occur if the data it uses to  learn isn't balanced.  

Third, people ought to have the ability to contest AI's decisions if they believe they are incorrect,  as well as the right to know when the system is making decisions about them through  educational means. This is about safeguarding our holy trifecta of fundamental, digital and  human rights in an increasingly Al-driven world, not just about protecting our data. 

Conclusion 

It is fundamentally important to understand accountability refers to the consequences of real  world problems influenced by algorithms used in the decision making process. The use of artificial  intelligence is spreading from sector to sector like the Health and Aadhar sector. Differential  algorithms that are now controlling different sectors and there is no such legal framework in  India that provides an organisation that holds accountability for these algorithms. That’s why a problem arises between ethnicity and effectiveness and that effectiveness can be portrayed only  through lumbering and garnering the governance models adopted in Global institutions with the  pre-existing models in our country.