The recent podcast episode I participated in, hosted by the gracious Mr Prem Kumar, was an engaging and thought-provoking session on the theme, “Ethics at the Speed of Innovation: Is AI Innovation leaving Humanity and Ethics behind?” The discussion revolved around the critical role of ethics in AI, especially in the context of rapid innovation and real-world challenges.
Here’s a detailed recap of the key points and examples I shared during the talk.
Why do you want to use AI, can you do it without AI – this hard question brings the focus to the real objectives of the solution we are building. We need to keep the user in the centre. Empowering the user and that is really important: We somehow got a notion that when we do digitization, everything has to be decided by the system, No. I gave an example of a bank manager who needs the power to give small loans in an emergency immediately. Systems need to collaborate with one another.
AI is progressing at an unprecedented pace, but it often outpaces the ethical considerations essential for its responsible deployment. I began by emphasizing the importance of balancing innovation with ethics, coining the concept of a “Minimum Ethical Product” (MEP) as an alternative to the popular “Minimum Viable Product” (MVP). This framework stresses the need to meet ethical standards, even if it means cutting features to prioritize fairness, transparency, and equity.
Companies like Microsoft, Google and IBM are placing a lot of importance on what they call as Responsible AI. Many papers talk about placing the responsibility of AI on the people who are building it.
Use AI systems, training them on data based on frequencies that match the needs of your business.
The discussion then moved to the pillars of human-centric AI, focusing on user-focused design, empowerment, collaboration, and ethical integrity. Real-world examples illustrated these principles. Platforms like Khan Academy and ChatGPT demonstrate how AI can enhance education and productivity through personalized, user-centric solutions. Microsoft’s Seeing AI app, designed for visually impaired users, demonstrates how technology can profoundly improve lives by prioritizing human needs. IBM Watson for Oncology is an example where AI assists doctors by providing evidence-based treatment options. This empowers healthcare professionals to make informed decisions, improving patient outcomes. AI chatbots like those used by HDFC Bank in India prioritize customer needs by providing 24/7 support, answering queries, and resolving issues efficiently.
One of the most compelling segments of the discussion was around the ethical dilemmas AI introduces. Drawing from the infamous “Trolley Problem,” I explained how AI systems, like those used in healthcare for resource allocation during the COVID-19 pandemic, face complex ethical choices. These systems, while efficient, often lack the empathy and cultural context needed for equitable decisions.
Bias was another critical topic. I referred to the case of Amazon’s AI-powered recruitment tool in 2014, which exhibited gender bias due to its reliance on historical data dominated by male applicants. This highlighted the need for diverse teams in AI development to counteract biases not just from data but also from developers, algorithms, and cultural contexts.
Ethical frameworks across cultures were also discussed. I noted how ethics can vary significantly by region and culture, with norms in India differing from those in Afghanistan or South America. This cultural diversity makes the creation of universal ethical guidelines challenging but essential.
Practical challenges in ethical AI were addressed with examples like price discrimination based on consumer profiling. Taking this further, I raised a hypothetical scenario where businesses might adjust product prices based on whether a customer is browsing on an iPhone or walking into a store wearing an Apple Watch. Such practices, while technically feasible, highlight the fine line between innovation and exploitation.
The role of leadership in fostering ethical AI was another key topic. I stressed the importance of CEOs and business leaders actively using and understanding AI tools. They need to spend more money training their users on how to use and take advantage of the new AI tools. Ethical innovation doesn’t have to hinder progress; it can serve as a competitive advantage, especially for smaller companies. Establishing cross-functional ethical committees and adopting global-local strategies for governance were proposed as practical steps.
I highlighted the increasing legislation around AI ethics globally, pointing to frameworks in countries like India and Singapore. India’s Safe and Trusted AI Pillar under the IndiaAI Mission is a commendable initiative fostering ethical AI development. Such governance structures ensure that businesses balance innovation with compliance while remaining competitive.
As we wrapped up the podcast, I reflected on the unintended consequences of AI and the importance of global thinking with local action. Ethical AI requires leadership, transparency, and inclusivity. Education and healthcare emerged as pivotal areas for ethical AI integration. I also urged leaders to explore tools like ChatGPT to familiarize themselves with AI’s potential and recommended Mustafa Suleyman’s book, “The Coming Wave,” for further insights into AI’s future trajectory.
In conclusion, I sincerely thank Mr Prem Kumar for hosting this enlightening session and the team at Power of Knowing Forum for the opportunity to share my views. It was a pleasure to engage with such a vibrant and curious audience. To summarize, ethical AI must balance innovation with humanity, requiring global perspectives, local action, and unwavering commitment to fairness and transparency. While challenges persist, the possibilities for positive impact are immense. Let’s innovate responsibly and ensure we keep humanity at the centre of it all.
