The scope of artificial intelligence in shaping the technology and its functionalities for the next generation is dynamic. It’s an exciting time to explore the widespread enterprise use cases. The transformations that contextualize its ethical use and its broader promises to drive innovation require optimal management strategies. How are tech leaders driving AI transformation? What is their large-scale perspective for managing their dynamic use cases? What are some key AI technology trends that businesses need to adopt in the modern landscape?
To understand and deepen the answers to these questions, we spoke to Sanjay Srivastava, Chief Digital Officer, Genpact about the ethical role of AI, its evolving importance in the business landscape, its use cases in driving innovation, key challenges in Development of AI-based solutions and key AI trends to watch out for this year.
Deeply rooted in the innovation ecosystem, Sanjay is a board advisor to several AI startups and a tech incubator, as well as a limited partner in digitally-focused venture funds. Previously he was CDO, CEO, startup entrepreneur and engineer. As Chief Digital Officer at Genpact, he chairs the company’s Executive Technology Board. He works with senior technical leaders at F500 companies to drive digital transformation at the intersection of technology, data structures and operating models.
Here are a few interesting excerpts from the interview:
What are the large-scale transformations that contextualize artificial intelligence?
AI is an amazing prediction engine, arguably the best prediction engine we’ll ever have. AI is reaching sufficient maturity to power many serious enterprise applications. The actual models or algorithms will become a commodity, and the key to success will be the application of these broad-based models in the context of sensitive, nuanced use cases, and the role of the domain and subject matter expert will be key.
For example in pharmacovigilance – AI can easily be used to extract adverse events from doctor notes, phone records and social media posts to spot a pattern in a large body of health trend data; and it can automatically classify and report these adverse events to regulators. But making decisions based on this data is risky and could have significant public health implications. So running this process through a reinforcement loop with an AI engine is not enough to automatically promote the better model. Since the entire process must be compliant with the law, companies should use two instances – one to run the current model and another that looks at the data to further improve the model.
We believe that successful digital transformation requires the orchestration of four dimensions – technology, data, people and process.
The opportunity with AI is to combine the knowledge and experience of a subject matter expert or a business end user with the advances and technologies surrounding AI in a way that allows us to leverage the intersection of the two disciplines. When we do that, we consistently deliver great results.
Sanjay, how are Chief Digital Officers and Tech Leaders driving AI transformation?
The role that technology practice plays in organizations has evolved over time, as has the responsibility of technology leaders. From providing ‘technology as a service’ they now want to tap into ‘technology as a strategy’. As technology leaders increasingly integrate advanced technologies into more aspects of the business, we face several challenges, including maintaining extensive business controls and a lack of collaboration between IT and business functions.
In addition, we also need to think holistically around people, processes and data to increase transformation value. Data has evolved from a by-product of automation to a first-class citizen in leading companies, which has become a key priority for technology leaders today. Furthermore, digital transformation is about fundamentally transforming an entire end-to-end process. Technology leaders need to understand that this requires a deep understanding of industry nuances and industry-leading process metrics – and reinventing and transforming the processes – long before they are automated. Likewise, in the people dimension, it is critical for technology leaders to think through the new operating model, resource acquisition, and reskilling as critical enablers for the transformation of the business.
Explain the role of ethical AI and its importance. What are some optimal strategies for organizations to implement the ethical use of AI?
Artificial intelligence is now an integral part of our everyday lives. And while its use is widespread—from robo-advisors making investment recommendations to predictive maintenance that improves machine utilization—it also poses risks and difficulties that can arise from improper use of technology or widening gaps and disparities . There have been concerns about how organizations are using it to make decisions, including bias in AI systems and a lack of skills to design AI solutions. There seems to be a marked increase in board level recognition of the pitfalls and the need for ethical frameworks. With these frameworks, companies can build trust with consumers, which supports AI adoption and brand reputation.
Companies that succeed in building industrialized AI systems over the long term will not achieve this by accident, but by focusing on building digital ethics and governance into their platforms from the start. For organizations that don’t, it’s not just about missed opportunities; They expose themselves to significant reputational, regulatory and legal consequences.
Before we say goodbye, Sanjay, consider some key AI trends that tech leaders should be aware of and adopt.
Based on our work with Fortune 500 companies that we serve globally, we see the following AI trends that leaders need to adopt:
- Rapid adoption of AI: AI adoption is ramping up across the board, and organizations are embarking on a “responsible AI” journey.
- Development of a data culture: Data is key to enabling AI and is now becoming the biggest driver of transformative value for businesses. But data is often scattered across multiple entities and lines of business, without a common model for ownership, use, and storage, and often lacks central control of master data, hierarchy, and lineage.
- Human in the loop: Perhaps the biggest challenge in applying AI to the enterprise is that no matter how good their computer vision, text extraction, or pattern recognition algorithms are, humans ultimately need to make the most of the information AI provides and make decisions, often in split seconds
- Digital ethics as a basis: As AI use cases expand, it becomes increasingly important to have strong governance to proactively oversee associated digital ethics. With the abundance of personal information, business leaders need to think strategically about how to integrate ethics into their AI programs and underlying datasets.
