(US & Canada) VIDEO | AI Is Not a Substitute for Business Strategy — Brown-Forman, Director of Advanced Analytics and AI Strategy

Amit Parulekar, Director of Advanced Analytics and AI strategy at Brown-Forman, speaks with Nayla Sidhoum, Client Director at AHEAD, in a video interview about his role, collaborating with business teams and leveraging AI, challenges in harnessing AI power, need for effective change management, have a solid data foundation, choosing the right model to do AI, and establishing AI governance through the AI Center of Excellence.

The Brown–Forman Corporation is a major spirits and wine business based in the U.S.

As the Director of Advanced Analytics and AI strategy, Parulekar leads and mentors a global data science team. His role requires working on high-impact analytic problems, finding solutions, and driving analytics culture across the company.

Adding on, Parulekar affirms working with the business teams, to identify opportunities and leverage AI to enhance operations and streamline processes. He also leads the development and integration of AI applications, models, and algorithms to ensure they are robust and scalable adding another feather to his hat.

When asked about the challenges in harnessing AI, Parulekar lists three challenges:

1. AI awareness and knowing what to invest in

2. The data challenge

3. AI governance

Regarding AI awareness he maintains that there is a misconception and fear around AI among most business and technology leaders. Parulekar advocates that effective change management is a must to demystify the apprehension around AI. He asserts that AI is not a substitute for business strategy.

Adding on, Parulekar states that effective change management ensures the availability of the right kind of skill sets and the right awareness amongst stakeholders about what AI is.

In terms of data challenges, he stresses the need for a solid data foundation and urges companies to ensure good data quality, data management techniques, and the right data governance principles. It is imperative to invest in the prerequisites to implement a successful AI strategy, says Parulekar.

About AI governance, he maintains that a well-planned AI governance framework is critical and will allow organizations to navigate through the AI evolution.

Moving forward, Parulekar states that the choice for a centralized, decentralized, or hybrid model to pursue AI depends on the company. A multitude of factors go into building the models, including business goals, maturity level of data culture, organizational complexity, and diversity, he notes.

Furthermore, Parulekar affirms that having a centralized model is advantageous to the diverse and heavily regulated healthcare sector. He adds that it provides a unified and standardized approach to managing data.

On the other hand, for a smaller organization, a decentralized framework may be more beneficial, Parulekar asserts. Since Brown-Forman is a medium-sized company, it has adopted a hybrid model of AI governance, as it envelopes the best of centralized and decentralized AI governance, he says.

According to Parulekar, the hybrid model has created a balance in the aspects of risk management. Also, it enables assessing new initiatives with greater flexibility to further adapt to the unique business unit challenges.

Brown-Forman has also established an AI center of excellence driven by the commitment to ensure that selected use cases have merit and will aid in standardizing responsible AI management life cycle, he shares.

The COE prioritizes establishing AI governance and has designated the role of an AI product manager, which is critical to the development of those use cases, says Parulekar. The AI product manager ensures that framework principles are followed and AI development happens within guardrails, he adds.

 When asked about fitting internal data quality into the equation of trust, Parulekar reiterates that data is a key challenge and the biggest precursor for AI technologies. As data fuels AI usage, feeding good-quality data into the models is non-negotiable.

Thereafter, Parulekar states that companies should commit to advancing AI by investing substantial energy and resources to build an AI governance framework and the right data foundation. It is pivotal to have a blueprint that serves as the guiding principle for the data lifecycle.

In conclusion, Parulekar states that the data that organizations create must be collected in a well-defined way, with a good data harmonization policy, good quality data, and data stewardship.

CDO Magazine appreciates Amit Parulekar for sharing his data insights with our global community.