India is rushing toward artificial intelligence, but the people most affected may not be the people deciding how it is used. This study asks a practical question: how can India spot the risks before they become everyday harm? Artificial intelligence is at a turning point in India.

The country is keen to adopt new technologies, but it also needs to think carefully about how they are used. AI is usually deployed with good intentions, to unlock value and create opportunities for people; however, it does not come without challenges.

Those challenges include privacy, data protection, job displacement, historical bias and discrimination. The study asks whether responsible research and innovation can help identify and deal with these issues. Its results point toward education, public policy, laws, regulation, and industry practices as places where improvement is needed.

The paper uses a simple four-part way to think about responsible innovation: anticipate possible effects, reflect critically on current practice, engage people through communication, and act on what those people reveal. In everyday life, it is like planning a road before opening it: look ahead at implications, question the design, ask stakeholders, and act on what you learn.

Because the study focused on the challenges of AI at the level of society, it used informal group discussions to investigate those challenges from the ground up. These discussions explored people’s attitudes, opinions, knowledge and beliefs.

They also helped develop shared understandings of common problems, and often of possible solutions. One suggestion was to have a diverse workforce, which could help reduce bias. But the concern is not only whether different people take part in designing systems; there may also be hidden bias that the people writing the systems do not see.

Privacy was another major concern. During the launch of the COVID-19 app, patient data was hacked and leaked, and misuse of data worried people across healthcare, climate and education. Participants also said that India lacked an adequate legal framework for data privacy.

As a result, it was not clear or transparent how data was generated, accessed, used and stored, or whether people had consented to those decisions. Privacy, data protection and data breaches came up repeatedly in the group discussions.

The Indian COVID-19 app was used as an example of the impact these problems can have. Similar problems were identified in applications used for climate, water, and education, showing that these concerns extend across several fields.

Participants suggested that an Indian law protecting people’s personal data could help build trust in how data is collected and shared. In India, people saw artificial intelligence as both an opportunity and a risk: progress depends on stronger education, research, and locally made tools, while privacy, unfair bias, unequal access, and weak public understanding could deepen existing problems.

The group discussions highlighted the need for responsible practices across several areas, so India can gain the benefits of AI while reducing its risks in a diverse society. Those areas include education, policy and governance, legislation and regulation, and innovation and industry practices.

The four-part framework can help identify the challenges and develop ways to address them. The approach connects public discussion with action: identify and consider the social and ethical challenges raised by focus groups, then develop ways to address those challenges.

The study also has an important limit. Because it used a small group of people with existing knowledge and interest in AI, its views do not necessarily represent the whole Indian population. The study offers a broad overview, but it gives more weight to academic and industry viewpoints.

Future research could include a wider range of stakeholders so that more experiences and views are represented. For people outside the field, the practical lesson is direct: AI decisions should not be made behind closed doors. Wider public involvement can help determine which systems deserve trust, which safeguards are needed, and how workers and communities are supported.

The study’s central message is that safer AI cannot be built by engineers and officials alone. People who use and live with these systems must help shape the rules, education, safeguards and jobs around them.