Background Circle Background Circle

Governance, Learning, and Ethics Frameworks: Dan Power Identifies Key Challenges

Nearly four decades into his career, Power is no stranger to the complexities of the data world. After 24 years as a consultant and 12 years in corporate America focused on data strategy, governance, quality, and analytics as well as master data management in financial services, high tech, and healthcare industries — he’s ready for an increased focus on AI across industries. 


DATA LEADERS NETWORK: Pertaining to the data space, briefly describe where you’ve been, what you’re currently working on, and what you envision for the space in the next 1-3 years.

POWER: I’ve just accepted a position as Head of Data Governance at a financial firm. My biggest interest now is developing people and teams, passing on what I’ve learned, and staying current by continually learning new things. 

Formerly, I was a Managing Director and Head of Data Governance at State Street Global Markets, where I helped launch the Data Governance program in 2017. Before that, I ran a boutique consulting firm called Hub Designs from 2007 to 2017; My clients included Apple, AIG, LPL Financial, Red Hat, and EMC. 

In the next 1-3 years, we’ll continue to see increased focus on AI in corporate America, and I expect a growing recognition that successful AI rests on a Data Governance foundation. 


DATA LEADERS NETWORK: What emerging trends or technologies do you see impacting data governance practices, and how do you foresee yourself contributing to this future?

POWER: I think it will continue to be all about AI, which is great in that it further justifies and validates what we’ve been saying about the importance of high-quality, well-governed data. The downside is that it has been crowding out everything else for a while now. It’s gotten hard to get some types of projects approved if they don’t involve AI, critical infrastructure, or mandated regulatory reporting. I plan on continuing to read, research, and learn, and to write, speak, and share my thoughts broadly with the data community. 


DATA LEADERS NETWORK: What are the most compelling discussions happening around data strategy and governance that are essential for leveraging AI effectively?

POWER: This is a fascinating area. I attend a fair number of conferences and talk with my CDO and data colleagues quite a bit. People are taking a look at their data strategies, and their governance efforts, with an eye towards recasting them to better support their company’s AI strategies. At the same time, some companies are taking a fairly slow and experimental approach to AI, keeping it mostly inward facing for now before they build their skill levels and confidence to take on customer-facing use cases for AI. 


DATA LEADERS NETWORK: In your presentation “The Role of Privacy and Ethics in AI/ML & Alternative Data” at FIMA US 2023, you identified clear indicators that the U.S. is heading toward regulation. What do you think about the current frameworks used to inform risk management? What questions aren’t being asked that deserve consideration?

POWER: I do think there’s some new legislation around privacy and AI ethics coming in the U.S. 

The Generative AI tools that have attracted so much attention in the last 18 months are trained on broad and deep datasets from a wide variety of sources. This has led to some copyright, privacy, and intellectual property questions. 

The New York Times is suing OpenAI and Microsoft for copyright infringement, for using millions of their articles to train ChatGPT. There’s also a class-action lawsuit by a California law firm alleging that OpenAI violated the copyrights and privacy of millions of users by scraping data from the Internet to train its models. There’s another class-action lawsuit from the Authors Guild on behalf of dozens of best-selling writers. 

I don’t think Congress and the various large state legislatures will ignore this situation. I think the U.S. wants to be a leader in AI, and political leaders realize you can’t have the gold rush / Wild West situation we have right now without corresponding meaningful regulations. 


DATA LEADERS NETWORK: In your view, what are the critical elements and considerations that must be at the root of all ethical AI initiatives? Is data quality more fundamental than ethics?

POWER: We have to be conscious about embedded bias in the training data, and also of the “creativity” of AI systems to “work around” the efforts of the development team, and to recognize the bias embedded in historical data. There’s a serious risk of “digital redlining” or “baking our prejudices into silicon.” That embedded bias is a type of data quality issue. There is bias and flawed reasoning reflected in the historical data that the AI system will be only too happy to replicate. 


DATA LEADERS NETWORK: You’ve made insightful predictions about the rising demand for Chief Artificial Intelligence Officers, a role that would evolve from current CDOs through reskilling and upskilling. Please expand on what reskilling and upskilling mean for data leaders. Specifically, what will enable CDOs to transition into these AI leadership roles?

POWER: CDOs today are the best group of people to pick up the mantle of being Chief AI Officers. They understand the data space very well, are used to managing super smart people, and can manage and communicate effectively with stakeholders across the enterprise, and outside of it. 


DATA LEADERS NETWORK: How are conferences, continuing education, and current networking platforms failing today’s business leaders who want to solve real data governance and management issues?

POWER: Conferences are a great way to meet colleagues, to network, and to be exposed to new thinking and vendors’ products. But they’re short and fairly superficial. Continuing education has lagged the data space significantly. By the time continuing education courses are created, the field has changed and evolved, so it’s hard to keep the courses in sync with the data space. There are various professional networking platforms out there like LinkedIn and Insight Jam, and a few consulting firms are doing good work helping people to network and mentor or be mentored. But solving today’s real data governance and management issues is hard to do using any of these methods. 


DATA LEADERS NETWORK: You actively shape the industry through advisory roles, conference appearances, and prolific authorship. What actionable advice can you share with data leaders who want to be active thought leaders in the industry?

POWER: As much as possible, share what you’ve learned. Use well-known, easily accessible vehicles such as LinkedIn, industry groups, writing, and speaking at conferences. We have to be courageous and put ourselves out there with new ideas and approaches, even when it’s a little nerve wracking. 


DATA LEADERS NETWORK: Given your expertise in data governance and strategy, how do you envision enhancing mentorship or leadership programs to ensure they effectively support the growth and development of emerging data leaders?

POWER: Enhancing mentorship and leadership programs can be done both formally and informally. Within the corporation, I’ve tended to be an informal mentor (i.e. when people reach out and ask for my help in developing to the next level). I love to teach people, and I find that, ultimately, I get more out of mentoring and developing people than I put into it. 


DATA LEADERS NETWORK: What leadership principles do you leverage when building a strong team that will actualize a successful data initiative?

POWER: Good question. My first principle is “servant leadership,” a philosophy in which the goal of the leader is to serve. A servant leader shares power, puts the needs of the employees first, and helps people develop and perform as highly as possible. This has worked really well for me over the past 20 years. Successful data initiatives are delivered by people, and I always want to bring out the best in the people that work for me, or with me. 


DATA LEADERS NETWORK: How do you help data leaders shift their focus from technical adjustments and tasks that take years of work to more impactful strategies that drive success?

POWER: Data leaders need to realize that the organizational and cultural change aspects of data initiatives are the toughest areas. The technology isn’t necessarily easy, but it’s easier than the “soft stuff.” People need to work on their listening and communication skills, learning to be empathetic, and practicing their skills at influencing, driving collaboration and alignment, and managing conflict. 


DATA LEADERS NETWORK: Beyond data-driven decisions, what excites you most about empowering organizations through data?

POWER: Empowering organizations through data is really about empowering people, like helping them learn new “soft” skills, learn new data skills, and become a version of themselves they weren’t really sure they could be. It’s a great feeling to mentor and develop someone, and to see them coming along behind you, growing and changing and developing. As people grow and change, organizations evolve and develop too. Helping companies and their data be better governed and managed is a great feeling as well.  


DATA LEADERS NETWORK: Is there anything else you’d like to share on the topic of data that hasn’t been captured in these questions?

POWER: At its best, data can illuminate and reflect our understanding of the world, and give us ways to solve business problems with processes, technology, and data. I’m fascinated by the political and cultural aspects of data, and get the most pleasure from helping people have their own “aha” moments about data. 

Join the conversation on LinkedIn.