.png){kind=logo}
Generative AI emerges as the kind of technological force that can produce realistic images and videos, and also human-like texts. The innovation it brings revolutionizes business models, freeing hitherto unseen creativity in sectors across content creation, design, customer service, and indeed healthcare.
Indian businesses are increasingly learning the need for governance frameworks to ensure that generative AI is used ethically and responsibly. Companies are committed to being open to societal expectations, utilizing this opportunity in ways that can mitigate risks.
According to NASSCOM, generative AI startups have seen a rise of more than 100% since 2021, with nearly 70% of private investments happening in 2022. Meanwhile, SAP India has recently stressed that a staggering 77% of startups are investing in technologies such as advanced AI, which is indicative of a considerable thrust on innovation.
1. Ethical Issues: The quicker uptake of GenAI also raises a number of issues that require frameworks of robust governance. Some of the other ones include ethics: such output of GenAI can be misleading or misleading-harmful deepfakes or biased. The ethical questions related to AI systems navigating them away from perpetuation of harmful stereotypes and spreading misinformation will demand careful consideration.
The faster the adoption of GenAI takes place, the more urgent its governance becomes. Misleading outputs, harmful deepfakes, and biases will be only a few problems that will then need to be critically addressed in order to prevent entrenched harm to stereotypes and misinformation. According to KPMG reports, India CEOs are worried about ethical challenges facing the effective implementation of generative AI that will be seriously constrained for 64 percent of cases by 2024.
2. Security Risks: The GenAI systems can be attacked in the forms of data poisoning or adversarial inputs and compromise the integrity and reliability of the systems. As a consequence, it is really essential to protect such systems against the above threats for establishing trust.
GenAI systems are prone to poisoning of data and adversarial inputs, and this can dent their integrity and reliability. Hence, securing these systems is so important that trust can be developed. According to a Telecom report, more than 73% of Indian organizations will deploy generative artificial intelligence in the next year to advance security measures and better integrate IT objectives with more related business objectives.
3. Compliance with the Law: In most countries, regulations governing the diminution of AI use exist. Thus, the company needs to be in a position to determine that its GenAI is perfectly compliant with these regulations, from data protection to an explanation of how the decisions the AI system will make are being made.
As for AI systems, with the progressing regulation of the technology, the use of such generative AI systems by the companies must be ensured to be law compliant with the requirements regarding data protection and transparency in decision-making. The new advisory released by the Ministry of Electronics and Information Technology, India, on March 15, 2024, supersedes the previous advisory dated March 1, 2024. The new advisory puts emphasis on intermediaries and platforms, pointing out frequent failure to meet due diligence obligations under the IT Rules 2021.
4. Transparency and Accountability: Users and stakeholders would want to know how GenAI makes its decisions. The transparent AI system that gives an account of the rationale for any output will have earned trust and accountability.
For generative AI systems, transparency and accountability are key. It will only through understanding how these decisions are taken that users and stakeholders may trust the integrity of the decision-making process. A transparent AI, which gives explanations about the outputs, builds trust and accountability. In December 2023, a study carried out by TELUS Digital showed that 71% of respondents believed that brands need to explain explicitly the products and services that use generative AI.
Indian companies are developing comprehensive GenAI governance frameworks that include the following components:
Ethical guidelines will have to be the bedrock of any governance framework of GenAI. They will define principles on responsible usage of AI with the prime connotation of fairness, accountability, and transparency. Companies like Tata Consultancy Services (TCS) have well-established guidelines for the kind of ethical standards which their AI systems adhere to so that every engagement of all their systems will alight well with the values of the society and to its ethical considerations.
There is a clear identification, assessment, and subsequently mitigation of the risks involved with GenAI. Some of these are periodic auditing and assessments that would determine areas of possible vulnerability and how such vulnerabilities could be mitigated. Infosys, for instance, has been very aggressive with the development of a robust risk management framework that assures security over its AI systems, furthered by regulatory requirements .
Transparency and explainability can help in winning the trust in GenAI systems. Companies are, in this regard, exerting their efforts by making investments in different technologies and methodologies that contribute to transparency in AI-decision making. These encompass development tools explaining the reasons behind the outputs created by AI and discussing the same with concerned users and stakeholders.
Data is the lifeblood of GenAI and proper data governance shall only mean the quality and integrity of the outputs by AI. Among the frameworks used in most companies today in terms of data governance are standards for data quality, policy on data privacy, and methods of data provenance or traceability.
Effective governance of GenAI requires communication with the stakeholder. This includes customers, employees, and regulators. Companies are designing mechanisms through which it can seek stakeholder feedback, and ensure that such feedback does find a place in its AI governance practices. In this way, AI systems satisfy stakeholders' needs and expectations.
As the AI system requires an adaptive governance framework, as much as that shift and fluctuations, the companies have begun the continuous processes of monitoring and improvement processes for GenAI systems that have to keep them abreast with the best ethical standards and requirements of regulatory needs. Governance policies and practices always relate with the changes seen in emerging trends and best practices.
The education level of employees has been the basis of building this responsible AI culture at the individual level. Big companies have been implementing resourceful programs to train their workforce in the ethical and practical dimensions of GenAI, like data privacy, security, and the ethics of AI technologies.
Indian companies are focused on building robust governance frameworks that connect to the responsible and ethical use of technologies. So, such companies look forward to infusing their AI systems with trust and accountability in order to fulfill their ethical concerns appropriately and to manage risks to build transparency and engage with the stakeholders.
This can unlock full generative AI potential while helping to mitigate the risks by having comprehensive governance frameworks developed and implemented for GenAI.
