Skip to content
Fast Company Executive Board

The Fast Company Executive Board is a private, fee-based network of influential leaders, experts, executives, and entrepreneurs who share their insights with our audience.

Why companies need to build products with global AI compliance in mind

Fast Company Executive Board member Deepti Kunupudi highlights five key strategies for making AI governance an innovation enabler.

[Courtesy of Deepti Kunupudi]

Profile photo of Fast Company Executive Board

AI development is moving at an unprecedented pace. In a few short years, generative AI has become an integral part of individual and organizational workflows. As AI is evolving quickly, so are the global regulations that govern how businesses use customer data. 

AI runs on data, and companies are constantly collecting it—including personally identifiable information (email address, date of birth, credit card number, etc.), purchase history, and customer service interactions—and using it to make decisions. Companies with an international footprint must comply with data protection requirements in each country and region they operate in or face substantial penalties. A brand selling to customers in Europe, North America, and South America, for example, must ensure that it is meeting the rules outlined by GDPR in the EU, HIPAA in the U.S., and LGPD in Brazil. 

According to Deepti Kunupudi, former chief data and analytics officer and current advisory board member for multiple companies, leaders need to build global AI compliance into their products from the beginning, not as an afterthought. When they are proactive and intentional about compliance, they can improve their workflows, products, and business outcomes and avoid expensive, time-consuming replications. 

“Compliance is at the crux of how you make decisions for your business,” says Kunupudi. “You have to understand how you’re using AI, what kind of data your AI is being trained on, and what the rules and regulations are in each country or state. That said, people often see AI compliance as an innovation roadblock—but it is actually an innovation enabler when you keep strategic design principles in mind. Building a system and then putting compliance on top of it doesn’t work. You need to build compliance in from day one.” 

5 STRATEGIES FOR ENHANCING COMPLIANCE AND INNOVATION

1. Create a cross-functional compliance team.

AI isn’t only the domain of the data science team. Different departments in your organization are likely already using AI, which means they should be part of the compliance conversation. Identify your current AI use cases, such as résumé screening, content brainstorming, or invoice processing, and create a cross-functional monitoring team. Secure early buy-in on governance controls from key stakeholders, and schedule regular bi-weekly meetings to reduce resistance and stay on track.  

“Typically, when you’re building products around AI, whether you’re using it for reducing costs or increasing your top-line revenue, you need to meet all security and compliance requirements,” says Kunupudi. “A cross-functional monitoring team with representatives from legal, data science, product, and cybersecurity departments, for example, will oversee the product development from a holistic point of view. They will monitor how you’re using data, if you’re using the right information, and if there are regulations around it.” 

2. Stay plugged into your industry. 

Don’t tackle AI compliance in a vacuum. Connect with others in your industry around shared challenges and best practices. 

“Participate in industry conferences, especially around AI, to see how other organizations are handling AI adoption and regulation,” says Kunupudi. “Certain industries are more highly regulated than others, and it can be valuable to learn from different approaches.” 

3. Treat compliance as an accelerator. 

As you create new products that use AI, integrate compliance into each stage of development: ideation, prototyping, testing, deployment, and monitoring. Think of regulations as opportunities to strengthen your solution, not just meet legal requirements. 

“Verify that the AI systems you are enabling for your organization or your customer are using data in an ethical way,” explains Kunupudi. “Implement checks to detect and reduce bias, minimize data usage, and complete documentation so you can move to the next phase. This will make your product more resilient to legal changes and help you gain customer trust.”

4. Develop continuous compliance monitoring. 

Once an AI product is launched, data will flow in continuously. Establish an ongoing compliance monitoring plan to analyze data, identify data or model drift, evaluate the quality of the AI model, and align all elements with the regulations of each country where you are deploying. 

“Put your compliance hat on in every phase, and make sure that all of your integration points are robust,” says Kunupudi. “And if somebody asks for documentation, you can go back and show your work. For example, if you have an application developed and deployed in one country, and the data is biased toward that particular country, you need continuous compliance monitoring to flag and resolve any model drift.” 

5. Deploy, iterate, and deploy again. 

Launch a product in one region at a time. Fulfill the governance rules in one country, understanding how you need to handle data, build an AI product, and deploy and monitor it. Then focus on another country and adapt your existing product to that location’s compliance laws. 

“Test and deploy a product with complete compliance with each location’s rules,” says Kunupudi. “This gives you transparency if somebody has a legal issue in that particular country. In the EU, for instance, it’s primarily a risk-based model. You identify whether AI products are low, medium, or high risk and categorize the governance rules accordingly. If you put in the effort early on, you reduce your risk in one region and make it easier to adapt a product for another region.” 

AI governance is still changing at a rapid pace from country to country. Multinational companies with a proactive AI compliance strategy set themselves apart from competitors by fostering a culture of agility, resilience, and innovation.

“Governance has to be the first priority in your AI strategy,” says Kunupudi. “Start early, understand the regulatory requirements, and be resilient to legal changes because the laws are fluid right now. Keep continuous compliance and risk assessment in mind at each phase. Modular frameworks will help you overcome roadblocks and achieve your goals faster.”