Understanding AI: Myths and Applications in Financial Services

AI is no longer for data scientists only. Most businesses have made AI tools part of their workflow. For example, we’ve seen chatbots, automated emails, and the like. As we embrace artificial intelligence, we have to discuss it. What is AI? What is it not? Should we use it for everything?

In this episode, Sarah Hoffman, VP of AI and Machine Learning Research in Fidelity Investments, defines AI and its impact on our lives. She also describes the ethical challenges of data bias and what people are doing to overcome it. Finally, Sarah explains why diversity and prejudice are significant concerns in AI development.

Tune in to this episode to learn how AI pushes for innovation. 

Here are three reasons why you should listen to this episode:

1. Discover artificial intelligence as a new approach to learning and training. 
2. Learn how AI is a reflection of our own beliefs and understanding. 
3. Understand the relevance of diversity in the field of artificial intelligence.

Resources

• Connect with Sarah on LinkedIn and Twitter 
• Know more about FCAT
• Be part of Random Hacks of Kindness

Episode Highlights

[00:47] AI and ML Research at FCAT

• The Fidelity Center for Applied Technology (FCAT) has a long history of innovation and commitment to technology.

Sarah: "We invest deeply in technology. But we've also always recognized that technology is just a tool. It's really how we apply it that matters." 

• FCAT develops platforms and products to empower the next generation.
• Sarah is a part of FCAT's research team. They explore the future of artificial intelligence (AI).

[02:29] Defining AI

• Sarah uses AI and machine learning (ML) interchangeably.
• ML refers to code learning from data.  It produces answers based on stored information to make predictions.
• AI is math, not magic. 

[03:43] AI in the Finance World 

• FCAT provides services that harness AI’s true potential.
• Financial services use AI tools often. Many people use chatbots, robo-advisers, and automated email responders.
• Several companies have adopted personalization and sentiment analysis.
• Models need to adjust when something changes in the world.

[06:46] Data Ethics Concerns 

• AI learns biases through data. 
• Ethics boards address ethical issues regarding AI projects and decide whether a problem needs AI.
• Fairness and explainability tools are available to protect against inadvertent biases. 
• Using AI can enhance how we train people and use fairness and explainability tools. 
• AI tools can help with issues regarding biases. 

[10:25] Automatic Writing and Coding 

• Multiple tools are now available to help you automatically write and code. Some tools can write code for you.
• The system is not perfect, but it's a good way for new developers to begin.
• Artificial intelligence can speed up coding for experienced programmers.
• AI tools are limited in terms of fact-checking. It can help with brainstorming, but it’s essential to be critical regarding the data an AI provides.

[16:04] Democratizing Innovation  

• The field of no-code/low-code has improved over the years. 
• We must democratize innovation to share ideas with everyone.

[19:57] Education and Machine Learning

• Each individual has their preferred learning method. For instance, some people don't learn well in classroom settings. 
• Artificial intelligence offers another approach to learning.  
• AI can simplify jargon and make knowledge more accessible to a broader range of people.

[22:20] Inclusivity in AI 

• AI tools provide feedback on how to use inclusive terms when speaking or writing. 
• AI biases reflect our assumptions and prejudices. 

Sarah: “AI tools are showing us the AI bias. Maybe we can start thinking about it for ourselves and think about 'Do we have this bias?' And I think that's another way that AI could help us be more inclusive."

[26:23] A Prediction Ahead of its Time 

• Sarah predicted that companies must prepare for a more flexible work environment by 2025. 
• Her prediction happened earlier than expected due to the pandemic. 

[27:39] Diversity in AI

• People with similar backgrounds might not see concerns that affect people outside their demographic.
• Diversity in the field of AI helps make its tools more inclusive.   

Jay: “[AI] is democratizing because it brings people into the tech. It's inspiring people–young people. Maybe folks that wouldn't otherwise be taught are encouraged to pursue these things. It's so promising for STEM education opportunities, and it'll be fun to watch as it evolves.”

[29:48] Jay’s Key Takeaways

• AI is math first, not magic. It requires colossal amounts of data to create a model.
• AI is everywhere and impacts humanity in many ways.
• AI democratizes innovation and can create things that previously were only doable by humans.

About the Speaker

Sarah Hoffman is the Vice President for AI and Machine Learning Research at Fidelity Investments for over four years. Her research foci include the future of AI, enabling data-driven enterprises, the future of work, and digital ethics. She started as an Information Technology Analyst and gradually built her way into AI and Machine Learning over the past years. Sarah is passionate about enhancing AI to democratize innovation and creativity for everyone. 

If you want to reach out, contact Sarah via LinkedIn or Twitter. 

Enjoyed this Episode?

If you did, subscribe and share it with your friends! 

Post a review and share it! If you enjoyed tuning in, then leave us a review. You can also share this episode with your friends and colleagues. AI and machine learning are a big part of the future, so help your community find out more about it!

Have any questions? You can connect with us on LinkedIn. 

Thank you for tuning in! For more updates, please visit our website. You may also tune in on Apple Podcasts or Spotify.