Innovation

Experian is honored to be recognized as a winner of the Artificial Intelligence AI Excellence Award by Business Intelligence Group. Experian was recognized for its credit and collections decisioning solution, PowerCurve, which features intelligent agent-customer assist that processes complex, regulated, and subjective interactions with customers, revolutionizing how they digitally interact, on their terms, with lenders. This AI virtual assistant offers customers 24/7 access to support from their credit provider - on a financially sensitive transaction such as collections. We embedded over 30 years of collections knowledge and experience delivered to major clients across the globe into a credit risk decision management solution that delivers this domain expertise, data, analytics, decisioning, and workflow to clients of all sizes via cloud technology. Harnessing the power of AI has enabled lenders to easily connect with customers on their terms in a fair and transparent way, and to recover losses in an operationally efficient way but also improve customer satisfaction. When’s the last time you heard a customer share a positive story about collections?  This digital self-service channel allows customers to engage with a lender on their terms and allows the lender to fulfill key objectives for quickly recovering losses, using the company’s strategic parameters for segmentation and treatment paths. The intelligent agent customer-assist feature within PowerCurve continuously learns which distinguishes it from the traditional decision-tree structure of a chatbot. It remembers interactions and learns from them, has short-term and long-term conversation goals, and recognizes small talk. It treats customers fairly and transparently. The result feels more empathetic and allows for an always-on and real-time consumer interaction. It can also offer a future-forward benefit to lenders, by allowing their employees to strategically focus on other areas of the business, potentially increasing the capacity for more credit products and services innovation. This Business Intelligence Group awards program sets out to recognize those organizations, products and people who bring Artificial Intelligence (AI) to life and apply it to solve real problems. Related stories: Going the last mile: Improving the End-to-End Digital Customer Journey Technology today: Focus on innovation drives award-winning, AI-powered consumer lending Podcast: Driving product vision with the customer front and center, a software architect’s view

At a time when the digital transition is in full swing, accelerated by the rapid change brought by the pandemic, the role of technology architecture seems to gain a renewed importance. This discipline is all about finding solutions to the challenge at hand by bridging the gap between business needs and technological enablement. In the latest Insights in Action podcast series ‘Women Making Waves in Tech’, we hear from Aruna Chalasani, senior software architect at Experian Global Decision Analytics. In this 20-minute conversation, she talks us through her experience driving innovation in highly diversified, geo-distributed teams and shares key learnings and takeaways about driving product vision with the customer front and center. It’s all about customer enablement Aruna highlights how, over the past year, the lifespan of innovation has continued to shorten. That means a shrinking timeframe to enact innovation from a technology perspective, whether that’s migrating to the cloud, leveraging machine learning, refining the customer experience, boosting data integration, or infusing automation at scale into the business. This fast-tracked drive into the digital world has also brought an increased digital risk. Security remains paramount, and they must look at it from a holistic perspective. These new innovation parameters call for robust customer enablement – self-service keeps gaining traction as a proven way to enable organizations to meet consumer demands at the speed of change, while more and more organizations seek the benefits of cloud-based solutions and automated decision management. For technology architects, driving innovation these days is about responding to customers’ needs even quicker, yet without losing sight of security, scalability, usability, and efficiency gains. Listen to the full podcast Access all episodes of Insights in Action on Soundcloud, Spotify, Google Podcasts

Artificial Intelligence (AI) offers people and companies many advantages, and we interact with it every day. From the technology we use to do simple things like heating and cooling our homes, to more advanced tools that map potential disease outbreaks across the globe. AI is also being used more and more in the financial services sector – from matching new customers with the right loan and terms to assisting with transactions in real-time online. In a recent study, we found two-thirds of businesses surveyed globally are using AI to help manage their businesses today. More businesses are keen to use AI but are challenged to fulfill requirements for decision explainability – a must-do for ensuring consumers are treated fairly. The history of AI AI stems from the realization of the potential of computation. The father of theoretical computer science and AI, Alan Turing, introduced a theoretical mathematical model of computation – aptly named the Turing Machine – in 1936. He described this machine as being capable of computing anything computable. By 1950, his work posed the question “Can machines think?” He introduced the Turing Test, still in use today to subjectively evaluate whether a machine is intelligent based on its ability to have a conversation. Six years later, in 1956, prominent computer scientists proposed the famous Dartmouth Summer Project. Advanced concepts were introduced and discussed and the term "artificial intelligence" was first coined. Over the following two decades, AI flourished. Computers became not only faster, cheaper, and more accessible, but they were progressively able to store more information. Meanwhile, machine learning algorithms continued to improve, getting the interest of experts in different fields and industries and taking the realm of artificial intelligence to a tipping point in the early ’80s. Back then, John Hopfield and David Rumelhart popularized “deep learning” techniques which allowed computers to learn from experience. Meanwhile, Edward Feigenbaum introduced expert systems which mimicked the decision-making process of a human expert, allowing the program to ask an expert in a field how to respond in a given situation and to learn from it. How can AI benefit both businesses and consumers? Following these early milestones, the advanced analytics sector has experienced explosive growth – with AI impacting many aspects of our lives today. While most people have come to realize that AI can be beneficial, even since the early days, there have been many different views on how those involved in programming the algorithms must take the necessary steps to prevent AI from reinforcing stereotypes, widening wealth and educational gaps, or providing incorrect answers at critical junctures such as in a medical setting. As an example of what not to do: a famous language model was trained using 8 million pages sourced directly from the web. So, implicit in this model are the preconceptions and biases included in its training data. In this case, it led to a model with a trend towards greater male bias in more senior, higher-paying jobs. How to determine fairness in AI models So how can we ensure that the use of AI does not reinforce societal racism, sexism, or other stereotypes? That leads us to define fairness. It’s the impartial and just treatment of people without favoritism or discrimination; when no unjustified distinctions occur based on groups, classes, or other categories to which they are perceived to belong. But, within the world of AI, there are varying approaches to fairness associated with different metrics to evaluate and adopt this sought-after algorithmic fairness. Any solution requires defining dimensions of fairness, but realistically, it’s extremely hard to capture all these very sensitive variables and risky to store and process them. To truly determine if an AI system is fair requires an enormous amount of data and expertise. Additionally, promoting fairness requires an approach across the entire data science life cycle and modeling life cycle. All areas must be considered from the approach to data collection to ongoing evaluation of decisions. And, while fairness in AI is not ‘once and done’ or easily solved, the good news is that it is an area of great focus for regulators, academics, and data and analytics industry experts, like our peers at Experian. The growing importance of transparency and explainability Models generally compute calculations that are complex and involve more dimensions than we can directly comprehend. Given this processing step from model-input-to-model-output is unclear, it leads to questions around how a model has come to a decision. Importantly, how can one be sure that the model is behaving as expected? There are different ways to address explainability. One includes an understanding of how different inputs of a model affect its outputs. Shapley values, introduced by Nobel prize winner Lloyd Shapely, consider an aggregate of marginal contributions for all possible combinations. Another technique involves explaining the behavior of a decision by identifying model constants verse variables to extract what drove a decision and how. Yet another method uses counterfactual explanations, identifying the precise boundary where a decision changes. This method is easy to communicate since it involves statements such as if X had not occurred, Y would not have happened. As in the case of fairness, there’s an on-going dialogue around explainability, underpinned by current and yet to emerge new techniques that maintain model accuracy and improve explainability. Artificial intelligence is past its infancy stage. It’s already had an impact on our daily lives and is becoming increasingly ubiquitous. Fairness, along with a transparent and explainable approach are key ingredients to help this field continue its transition to maturity.

In a world drastically and constantly changing, industries and technologies are being transformed at a rate never before experienced. Recently, I had the opportunity to speak with Roy Schulte, Distinguished VP Analyst at Gartner, about trends in the evolving world of big data, advanced analytics, and decision intelligence — trends that are powered by advancements in cloud technology, machine learning, and real-time streaming. Below, I will share a summary of key highlights. The growth of decision management More businesses are implementing decision management. They want to make more automated decisions to accelerate outcomes while improving applicability. In tandem, decisions are becoming more complicated with regulators increasingly expecting decisions to be explainable and with audit trails built-in. Equally, the combination of machine learning and decision management has placed greater focus on the importance of avoiding bias. Bringing all of this together promises continuous decision improvement; updating models and strategies in days or even hours rather than weeks and months. Essential in responding to the rapidly changing environments, none more so than the impact of Covid-19. As businesses are implementing decision management, they are putting the new systems into the cloud. Based on a Gartner survey in mid-2020, 67% of respondents said their digital business platform will be cloud-native application architecture. It’s the primary criteria for the architecture of many of these new systems. Migrating to the cloud The reason decision management is going to the cloud is the same reason other areas of business are taking this step. Organizations are highly motivated not to run their own systems. There is no competitive advantage to doing so. They want to entrust it to others so they can focus on what the business does best. The migration will continue gradually to the cloud, with a current acceleration based on Covid-19. In a recent Gartner survey, 65% of respondents said the pandemic accelerated their plans and funding for doing digital business. Most models and strategies will be built in the cloud, and the actual runtime decisions will be distributed with some on the cloud, some on-premise in a data center, and some out in the edge in a mobile device. Real-time streaming In the past, traditional business information was done on static, snaps of data from the past. Today, much of this is going real-time, depending on the kind of decision that is being made. For example, when a customer is visiting a website, there are mere seconds to generate the next best offer. This is a real-time decision. However, some decisions do not require real-time, such as the strategic decision to acquire another company or not. That is why it’s important to align the decision speed and cadence with the actual business problem. If real-time data will be used, such as for an e-commerce situation, some of it must be streaming data. With the increase in factors taken into account when making decisions, you need data that is connected, contextual and continuous. The data must come for your entire ecosystem, not a single department, but across your company, business partners, your customers, or market data. Streaming examples for e-commerce might include location, what the person has been doing on the web recently (clickstreams), and records of contact with your business such as calls and emails. A real-time lookup is involved with inventory and external factors like credit rating through an API, and customer data will include historical and real-time. With these factors, real-time decisions for e-commerce will be more effective, with higher yield rates and lower fraud rates. Machine learning in model building and execution Machine learning (ML) is making predictions, not decisions. When a prediction is made and a score is provided, a rule must be applied to determine outcomes based on the score. If considering rules or analytics, the truth is that in most cases, both are needed. The goal of ML in decision management is to have applications that are easier to develop, faster to develop, and lead to more accurate outcomes. To achieve this objective, a process covering all stages of a decision cycle is needed — Observe: Getting connected, contextual, continuous intelligence. Governance is key at this step to know where data is coming from and that it will be used in authorized ways. ML models and strategies must avoid bias and alternative data sources and lending criteria should be considered to expand the business without incurring increased risk. Orient: The next step is putting the data in context. When dealing with models at scale, it’s important to be able to track outcomes through tools such as performance dashboards. Eventually, this will lead to the hyper-personalization of models. Decide: Once models are built, strategy, rule authoring, and approvals are needed. Workflow and collaboration mechanisms help manage the process and accelerate the pace of developing new decisions. Act: Next comes the deployment of models. Using logic to make decisions across multiple applications accelerates deployment, often referred to as centralized management or reuse of decision factors. Feedback: Finally, continuous logging of decisions and effects of decisions. Tracking provides the ability to audit past decisions, explain what was done, and accurately post hoc remediate. Ongoing feedback also enables continuous decision improvement at an accelerated pace. The future of decision management includes decision intelligence In summary, there are five considerations for the future of decision management — A systematic approach to decision making, including a lot more automation and decision intelligence, is clearly on its way. Migration to the cloud is well underway with acceleration thanks to Covid-19. Equilibrium will be reached where some decisions are made at run time at other locations, but most of the development of decision-making, modeling, and strategies will be based on cloud platforms. Data science ML vendors have not focused on decisions. Some may come to realize the reason you do analytics is decisions and broaden the scope of what they do, or they may stay focused and instead partner with other vendors to enable end-to-end decision making. For certain kinds of logic, graph databases and graph analytics can be very powerful. Likely this will become a big part of decision intelligence going forward. Finally, there is huge untapped potential in optimization technology to improve decisions either at development time or even at run time by applying optimization techniques. This could lead to achieving the full vision of artificial intelligence. Related stories Insights in Action Podcast: Identifying the core capabilities your business needs to get MLOps right New Tech Talks Daily Podcast: Machine learning and AI in business — investment trends pre- and post-pandemic In digital transformation, small wins lead to big outcomes

Get the latest from our global experts with these top December headlines, including meeting the demand for digital, increasing consumer expectations, women leading artificial intelligence, and protecting against fraudsters over the holiday shopping season. Investment priorities to meet consumer demand for digital banking In this BAI Banking article, Chris Fletcher, SVP Decision Management & Cloud Services, explores the investment required of financial institutions to transform their use of data and analytics and deliver on credit risk strategies. What’s the proper path for better payments? In context to consumers’ digital expectations post-Covid-19, Progressive Grocer considers the future of payments in food retail and beyond – with contactless payment options already rolling out at a large drug store chain. Wisdom from the women leading the AI industry, with Laura Stoddart of Experian Authority Magazine speaks with Laura Stoddart, Data Scientist, about her career path, her experiences working on ethical AI and using emerging datasets to evaluate risk as well as her thoughts on the future of this industry. #TradeTalks: Increasing consumer demands and expectations Steve Wagner, Global Managing Director of Decision Analytics, joins Nasdaq’s Jill Malandrino to discuss recent research findings on increasing consumer demands and digital expectations, and ongoing considerations for a post-Covid-19 world. A holiday season like no other: What to know to guard your company against fraud Itay Levy, Forbes Councils member and CEO and Co-Founder of Identiq, provides his perspective on the increased preference for online shopping and the need to strike the right balance between customer experience and efforts to mitigate fraud. Stay in the know with our latest insights:

  The need for advanced technologies, such as artificial intelligence, has surged in the wake of Covid-19. The strain of the pandemic on businesses and economies has created tension in operational models requiring a quick and dramatic response to this digital disruption. As transformation efforts continue, there are several considerations for the growing field of AI – including ethical AI, the need for diversity and gender balance, and striving to be consciously unbiased. This final post in our “Game Changers: Women in AI” series takes a deep dive into AI careers. Our experts share important lessons on how to thrive, including having mentors and sponsors, staying relevant with new related skills, understanding problems to be solved, believing in yourself and actively seeking growth opportunities. New to the Game Changers: Women in AI series? Read Part 1 - Game Changers: Women in AI Read Part 2 - Game Changers: Women in AI   Q: What advice would you give to help other women in AI thrive? He: "I would suggest being brave. Don't be afraid of trying new things. Sometimes we fear we cannot do something, but once you try it you find it’s not very difficult, you can do it. You can do it very well. So, I think the first thing is just try it. Don’t be afraid of making mistakes. If you go this route, be confident. Women are very smart and competitive, but they may not recognize how good they are. Also, if you find that you may be interested in this area, find resources and see if this is something you want to dedicate yourself to. There are a lot of options online. Even a lot of the universities now offer their courses online. People also share code online, so there are lots of good resources to help you explore and start learning. Overall, remember to believe you can do it." Kazmi: "For everyone that wants to try AI, or if you’re already working here and want to remain in the industry and do good work, you have to keep yourself relevant — learning and keeping yourself updated with the newest research that's happening. There is no end to learning in this field. At the same time, you need to have business knowledge to truly understand a business problem statement and convert that to a data science problem statement, and then start developing solutions for it. I really think that women can be strong contributors in this regard by leveraging their management and analytical skills to bridge the gap between the two areas." Kung: "I think we all need to be ourselves and respect ourselves. You need to have a goal and work hard for it. I think it is the same for anyone who wants a successful career. You need to set a goal and work hard for it and you will achieve it. Really, it's all about working hard. Also, my experience in AI has included a lot of brilliant women, so I never really felt like this is a job for men vs. women. The truth is we want more people to understand what we are doing – that there are many great things we can do with data. It is not something to fear. It’s not this magical thing. It is statistics. It is computing. It's coding. It's everything good." Peters: "It's so important to reach out and look for both mentors and sponsors, and this can be at any age. Mentors are our sounding boards to help with career development. There's some overlap with sponsors, who are opening the doors and speaking about you on your behalf in order to accelerate the track to the next place that you want to be. Mentors and sponsors are good starting from a very young age – and I think that’s a critical aspect of bringing more women along. Find these folks, make those connections, nurture those relationships, and have those mentors and sponsors. I really think that's a key aspect. Also, women do not necessarily need their network to be all women. You need to find the best people positioned to help you in your journey." Stoddart: "Having a mentor is good, especially someone who's more senior in your target field. And, it doesn't necessarily have to be somebody who you're working with or somebody who's your boss. They can be from academia or a different company. It's nice to have the outside perspective. It’s also helpful to network – I’m using virtual events now. I’ve met a lot of women in data science through activities outside of my current role. There are so many opportunities beyond your day to day job. I try to have a few things going at once -- I'll mentor somebody, I'll have a side project or volunteering, and my full-time job as well. For example, for the social enterprise I'm working with, I'm getting experience forecasting. It's nice to give back, but it also makes you a stronger data scientist to work on these different projects." Q: Is there a person or experience you are grateful towards that helped set you on the path to where you are today? He: "First, after graduation, I got a job in transaction analytics, detecting fraud transactions in credit cards. Essentially, it has the same goal as other projects, understanding human behavior from large amounts of data. That's what amazed me and kind of drove me into this direction. After that, I got the job here at Experian and I was exposed to a lot of great innovations and projects." Kazmi: "In the eight and a half years I’ve been in the AI industry, I’ve had the opportunity to work with multiple organizations across different domains. Through this diverse experience, I’ve met and worked with women from different backgrounds both as leaders, as well as colleagues. I’ve seen successful women leaders from all walks of life – from different educational backgrounds, whether from computer science, engineering, mathematics, or economics management, et cetera, or even differing nationalities and ethnicities. It has been impactful to see successful women leaders cutting across industries and localities." Kung: "Professionally, the person that I'm grateful to is my first boss. He was a teacher for me and taught me a lot. Everything that I am today, everything that I do at work, professionally, he was who trained me for it. When I think of the professional Jennifer, I always think of him. I think in my whole career, everyone who was part of my path, they helped me somehow. Maybe in little ways, and maybe in some big ways, they all helped me." Peters: "There are so many people I am grateful to in my career. Overall, where I am today comes down to the opportunities I was given. I had the opportunity earlier on in a prior role to be exposed to big data and frameworks, an exciting precursor to my work with AI. Today, when I think about my work with fraud and identity, AI is such a critical piece of that. And it's becoming increasingly important as we apply these concepts into financial services. I’ve been able to join collaborative and innovative colleagues, fraud experts, in a unified quest to solve the fraud challenge." Stoddart: "I am grateful to the person who brought me into this department. He saw something in me, he understood that I really wanted to learn, and he created a position for me. They were not hiring for a data analyst at the time, so that was really energizing. Also, I don't look for positions that already exist, because if everybody applies for positions that exist, it’s limiting your scope. A lot of the things that I've obtained in my life, it's because I've been a bit brave and asked for it. Even if it's not there on a plate, here I am." Related stories: New Podcast from AI in Business: The evolution of the data business in the age of AI Game Changers: Women in AI (part 1) Game Changers: Women in AI (part 2) Yi He Yi He works as a data scientist in the Experian NA DataLab. She is dedicated to using machine learning and AI to extract information from large amounts of data to identify, understand and help people, and prevent fraud. She aims to bridge online and offline worlds by linking identity data from these unique sources. With a focus on minimizing friction to customers, Yi’s work helps organizations identify synthetic identities to avoid fraudulent applications. Recently, she contributed to a Covid Outlook & Response Evaluator (CORE) Model – a “heat map” of geographic populations across the U.S. most susceptible to severe cases of Covid-19. Deeba Kazmi In her role as a data scientist at the Experian APAC DataLab, Deeba Kazmi is focused on solving business problems with analytics, including the development of consumer and small to medium enterprise credit risk models that leverage alternative data. Deeba is passionately focused on leveraging AI to create solutions that can help address issues faced by developing markets. Most prominently, this work includes her data science leadership contributions to solving a crucial economic and societal problem – financial inclusion. This effort is helping disadvantaged socio-economic consumer groups gain access to vital credit and financial services by leveraging the power of technology to deliver better outcomes. Jennifer Kung Jennifer Kung is an analytics consultant for Serasa Experian Decision Analytics, where she combines her knowledge of financial services with her data analysis expertise. Jennifer aims to harness the power of data through robust, descriptive and predictive analytical solutions to help clients realize the benefits of the massive amounts of data available to them. She recognizes the magnificence in powering discoveries through data analysis and enjoys revealing these capabilities to businesses who can benefit from these robust, yet approachable solutions. Jennifer enjoys knowing that her work helps to simplify and accelerate decisions that consumers rely on at important times in their life. Kathleen Peters Kathleen Peters leads innovation and business strategy for Decision Analytics in North America. As the prior Head of North America Fraud & Identity business, Kathleen is well-recognized as an identity industry innovator, being named a “Top 100 Influencer in Identity” by One World Identity the last two years. As of 2020, Kathleen was named Chief Innovation Officer for Decision Analytics. Kathleen and her team rely on the power of AI to continuously find new ways to solve customer challenges by defining product strategies, new paths to market and investment priorities. Underlying these efforts is a key focus on the ethical use of technology and the need to be consciously unbiased. Laura Stoddart Laura Stoddart is a physicist turned data scientist who works at the Experian DataLab in London. From her first exposure to AI, she recognized how quickly it can have an impact on the world, which has driven her to get and stay involved in the industry – both professionally and personally. Laura’s recent work has focused on ethical AI, having recently contributed to her first paper addressing the removal of bias from models. In addition, she is concentrated on leveraging emerging datasets to evaluate risk. Outside the DataLab, Laura also volunteers her data science skills to good causes such as Bankuet and helps expose others to the world of AI through mentoring.

The relationship with artificial intelligence may have started with robots but its integration into the way people interact with the world today looks very different. AI is in our pockets, our homes, our workplaces, and its pay-off is being realized across many industries, including financial services, e-commerce, telecommunications, streaming services, insurance companies, and more. Though some people and businesses still have reservations about its use. In the next article in our “Game Changers: Women in AI” series, we examine the artificial intelligence debate with arguments against and for its use in our everyday lives, and how it can bring real value to our interactions with businesses – whether it’s preventing fraud, increasing financial accessibility, enhancing the digital experience or supporting public initiatives to prevent the spread of Covid-19. Q: What is your view on criticism of AI or arguments against its use? He: "AI is already all around us and sometimes people don’t even realize it. For example, smart devices remember your preferences, try to understand your behaviors, and help you with reminders, goals, or some other alert. For some, this can feel a little bit scary, like they are collecting information and profiling you. But really, AI is helping people by using large amounts of data to train models and find patterns in the information to solve complicated problems." Kazmi: "Since AI is still so new, every time a product or a change in experience through AI is introduced, there are bound to be reluctancy in adoption and initial failures which lead to opposition. But, to establish the final best product possible, we need understanding between AI research teams and business stakeholders. Take the example of Elon Musk. He has come up with SpaceX and Tesla, but there have been so many failures in their development. Still, the entire world was looking up to these ventures, because these products are something that's going to bring huge positive change." Kung: "People need to keep in mind that AI, and all this data science technology, are just tools to help us. It's not that a machine will replace someone. I’ve heard a lot of people saying, "You create things automatically, and machines will replace our job." That’s not how it is. The truth is, we are creating these kinds of things to help us. It improves our lives by saving our time to focus on other useful things that a machine can’t do." Peters: "It’s helpful to consider what got us here. Years back, people would ask, “Are you ready for big data? Do you have big data?” What we found was that as more data was available, even when managed effectively, we needed ways to consume it and to garner insights from it. This underlying piece drove the need for AI and machine learning. Working with these technologies is critical to harnessing the power of data for what we do, to apply these concepts to fuel significant problems, like stopping fraud." Stoddart: "The topic of bias in AI creeps up in the news. If an algorithm is not checked properly, it could mean a portion of the population isn’t reflected. This stems from assumptions inherent in people. If those writing the code are not diverse, you likely miss out on representing whole groups of people in the wider society. This issue of bias emphasizes the importance of team diversity, of driving success by having opinions challenged and ensuring representation across diverse groups." Q: Is there anything you would like to share that could help alleviate fears and show the public that AI is beneficial? He: "It will lessen fears if we can help people realize there needs to be humans involved. To understand the data, to understand human behavior, everything is about the observation and how you interpret it. It also helps to share the benefits people will realize. For example, AI can improve consumer experiences — such as when filling out an application. It can build bridges between different types of data to supplement the details provided. This reduces the friction felt by the applicant by simplifying the inputs required, which is very useful on wearables and mobile devices." Kazmi: "AI can change the world. If you just look around, data science is part of everything nowadays. And, there's often a solution you benefit from but are not even aware that it has AI embedded in it in some way. It’s important to encourage understanding and acceptance and highlight all the good work that people are doing in this industry. We need to acknowledge and encourage endeavors to further these contributions and progress in the AI industry." Kung: "My concern is that people think “Oh, you just put something in the machine and the machine will tell you what to do." It's not like that. People need to realize a human must analyze the results – what it gives you and what you see. It needs to make sense for their business. The machine will not know what you’re analyzing. It will just run the algorithms that you put in it and it gives you a number. It’s up to people to analyze it." Peters: "Whenever you go into a new and somewhat unexplored area, there will always be different aspects to consider. As researchers, innovators, and developers, we need to be aware of inherent risks and keep an eye on the ethical aspects of technology. This focus helps ensure the thoughtful progression of AI, creating the right guardrails to thwart fraudsters and ill-intentioned individuals and equality by being “consciously unbiased” in the models and systems we are building." Stoddart: "I mentioned the need for diversity to prevent bias. I’m proud to be contributing to a project called “fairness.” It’s about tackling bias in models – using AI to help treat everyone fairly. Our work has enabled people to drill down and properly check attributes to ensure that decisions are fair and not discriminating against a certain group. If it’s not fair, it provides the opportunity to fix it. I believe this will be a really important tool going forward." Q: What examples can you share for how AI can bring goodness to the world? He: "At the very beginning of our latest initiative, we were thinking, “how will this development and innovation help the world?” It was hard to answer until we created different use cases. Currently, we have several meaningful results using AI – linking data to identify a person and deliver the best customer experience and helping detect fraudulent applications using fake or synthetic IDs. We also recently developed a heatmap for predicting Covid-19 severity for more than 3,000 counties in the U.S. We’ve made this tool available to assist public researchers as well as government and policymakers." Kazmi: "I am truly satisfied with the work that I have been doing because it's very exciting to find new ways to have a positive impact. From the day I joined Experian, I've been part of a project called financial inclusion, leading the data science part of it. We are helping people and entities stuck at the lowest level of the financial ladder. This is the beauty of data science, helping consumers and small entities access credit and come out of a vicious cycle, to move up financially, leading to the overall growth of the financially weaker sections of society." Kung: "Within my area of focus, financial services, we can help make life easier and help get things done faster. The important thing is time-saving because we need to get things done quicker. For example, sometimes people try to secure credit and the bank takes too long to give an answer. Or, with a mortgage, there is a lot of paperwork needed. We can use an AI tool to help analyze this paperwork faster, which helps the customer who needs the loan get their home faster." Peters: "Some of the ways that it can bring goodness to the world is where we are just limited by the scale or the speed that we want to move when solving problems based on huge amounts of data, especially in real-time. Where AI can help predict next best actions or best outcomes in a way that usually would require a lot of research or photographic memory. Very relevant today, this applies well to the medical domain, but there are so many areas AI can help us better consume data at our fingertips and predict new innovative areas to explore." Stoddart: "In addition to the fairness project I mentioned, I also use my data science skills volunteering with a social enterprise, helping them obtain the insights they need to determine what food and supplies are most needed at food banks. The insight allows them to prioritize what items to buy in bulk with monetary donations from the public. Usually, food banks are really separated in the UK, so this is a new approach benefitting from advanced technologies." Related stories: Game changers: Women in artificial intelligence (part 1) Impact of technology on changing business operations Forbes: Are we comfortable with machines having the final say? Yi He Yi He works as a data scientist in the Experian NA DataLab. She is dedicated to using machine learning and AI to extract information from large amounts of data to identify, understand and help people, and prevent fraud. She aims to bridge online and offline worlds by linking identity data from these unique sources. With a focus on minimizing friction to customers, Yi’s work helps organizations identify synthetic identities to avoid fraudulent applications. Recently, she contributed to a Covid Outlook & Response Evaluator (CORE) Model – a “heat map” of geographic populations across the U.S. most susceptible to severe cases of Covid-19. Deeba Kazmi In her role as a data scientist at the Experian APAC DataLab, Deeba Kazmi is focused on solving business problems with analytics, including the development of consumer and small to medium enterprise credit risk models that leverage alternative data. Deeba is passionately focused on leveraging AI to create solutions that can help address issues faced by developing markets. Most prominently, this work includes her data science leadership contributions to solving a crucial economic and societal problem – financial inclusion. This effort is helping disadvantaged socio-economic consumer groups gain access to vital credit and financial services by leveraging the power of technology to deliver better outcomes. Jennifer Kung Jennifer Kung is an analytics consultant for Serasa Experian Decision Analytics, where she combines her knowledge of financial services with her data analysis expertise. Jennifer aims to harness the power of data through robust, descriptive and predictive analytical solutions to help clients realize the benefits of the massive amounts of data available to them. She recognizes the magnificence in powering discoveries through data analysis and enjoys revealing these capabilities to businesses who can benefit from these robust, yet approachable solutions. Jennifer enjoys knowing that her work helps to simplify and accelerate decisions that consumers rely on at important times in their life. Kathleen Peters Kathleen Peters leads innovation and business strategy for Decision Analytics in North America. As the prior Head of North America Fraud & Identity business, Kathleen is well-recognized as an identity industry innovator, being named a “Top 100 Influencer in Identity” by One World Identity the last two years. As of 2020, Kathleen was named Chief Innovation Officer for Decision Analytics. Kathleen and her team rely on the power of AI to continuously find new ways to solve customer challenges by defining product strategies, new paths to market and investment priorities. Underlying these efforts is a key focus on the ethical use of technology and the need to be consciously unbiased. Laura Stoddart Laura Stoddart is a physicist turned data scientist who works at the Experian DataLab in London. From her first exposure to AI, she recognized how quickly it can have an impact on the world, which has driven her to get and stay involved in the industry – both professionally and personally. Laura’s recent work has focused on ethical AI, having recently contributed to her first paper addressing the removal of bias from models. In addition, she is concentrated on leveraging emerging datasets to evaluate risk. Outside the DataLab, Laura also volunteers her data science skills to good causes such as Bankuet and helps expose others to the world of AI through mentoring.

The artificial intelligence (AI) market is expected to grow 159% by 2025 to $190.61 Billion, according to Markets and Markets, and there’s considerable value for businesses and consumers. In our July global survey of businesses and consumers, we found that 60% of businesses planned to invest in advanced analytics and AI to better support their customers' financial needs during Covid-19. As more businesses adopt AI, processing their vast amounts of data with advanced analytics for automated decisions, human oversight is and will remain key to ensure transparency and explainability. This “human element” in AI was the inspiration for our latest “game changers” series. We recently sat down with five industry experts to get their view on how AI is making the world a better place, and how its use in financial services can be realized. Yi He, Deeba Kazmi, Jennifer Kung, Kathleen Peters, and Laura Stoddart are visionaries and leaders in data science and innovation making a real difference in how advanced technologies are helping consumers and businesses engage more meaningfully. Q: What excites you most about the AI Industry? He:  "As AI is more involved in our lives, it provides benefits we couldn’t imagine before – such as using your face to unlock your phone security. With the development of AI and machine learning, we can find patterns in data or in behaviors of people to solve complicated problems. That’s really it; helping people make life easier." Kazmi: "The main thing is that AI is not only transforming the way we live and communicate, it's changing the way almost every industry around the world is going to operate. To positively contribute to this growth, it’s not just that you need to learn and then deliver, but to keep innovating and coming up with new solutions that others learn from." Kung: "The technology improvement excites me. Things are getting easier, giving us more time to focus on what really matters. We usually don’t have time to focus on some of these areas because we are used to doing things manually. Now with AI, we have a machine to do a job that is manual, so we can focus on analysis and improvement." Peters: "What’s most exciting for me are ways AI technology can augment human decisions and innovation, in new directions that we historically run out of horsepower for. And, it can be applied to virtually every industry — the ways that it can better help us leverage big data, robotics, the Internet of Things — there are so many directions we can go with AI." Stoddart: "One of the most exciting things about AI is that people benefit from it every day — using social media, or maps to get to the shops, sometimes without even realizing it. And, if you can create an algorithm that can help somebody get credit who previously couldn't, you can have a real impact on the world that actually changes people's lives for the better." Q: What concerns you most about the AI industry? He: "I think the key things are data security and privacy protection. People are more and more sensitive about their information being used and released, which is understandable, and why opportunities exist to opt-out of information being used or sold to third parties. The key is to offer comfort by building in how to secure the data and protect privacy." Kazmi: "There are pros and cons of everything, especially with a stream of faster evolutions in prominent areas affecting our day-to-day lives. Since it’s still so innovative, when AI is introduced, there’s bound to be reluctance. But, to progress, we need acceptability, encouragement and patience; an understanding between AI research and stakeholders that these developments are going to bring huge positive change." Kung: "My main concern is that we need to keep in mind that AI is just a tool to help us. The machine will not replace humans and it cannot tell you what to do. An algorithm can give you a number based on its design. You need to analyze that result and ensure decisions make sense for your business." Peters: "The more we know and learn about AI, the better we can anticipate potential risk areas. These include the ethical aspects of technology, and striving to be consciously unbiased. As we progress, explainability and other model governance practices will help us stay within the right guardrails and mold the necessary regulations." Stoddart: "Lack of diversity concerns me – both in the boardroom and on the programming side. Decisions that we make in our programming are based on assumptions as human beings and our lived experience. If the people writing the code are not diverse, you’re missing out on whole groups of people in the wider society." Q: Can you share with us the “backstory” of how you decided to pursue this career path? He: "My educational background includes cognitive science, neuroscience, and psychology, and it involved a lot of data analysis and modeling. I wanted to understand how humans behave. In my first job, I did essentially the same work — understanding human behavior from large amounts of data — but to detect fraud. That amazed me and driving my focus today." Kazmi: "My education included subjects around analytics, and had a lot of flavor of data science, predictive modeling, mathematics and statistics. AI was very new at the time. I studied these topics and began to understand how data science is developing, and what's the future of it. I really got excited and interested into it. And once I started my career, there was no looking back." Kung: "As a child, I thought I wanted to be an engineer. Statistics was my second choice. But, I am really glad I had the opportunity to follow this path, because statistics and data analysis are amazing. When I started my course, I was so amazed at how data analysis can help you discover a world. You can do anything with data. I realized that this was my true passion." Peters: "I became interested in AI from the business aspects – working in a big data environment, we really needed machine learning and AI to handle data at scale. When joining Experian in the identity and fraud area, our mission was clear – harnessing the power of one of the largest data assets in the world to make a difference; finding new ways to stop fraud." Stoddart: "I studied physics at university and attained a master's in particle physics. But, during my final year, I started to learn about AI and machine learning. It was inspiring, especially how quickly they can have an impact on the world compared to academic research, which can be over many years. Realizing how quickly it was progressing, I thought it would be really exciting to get involved." Q: Can you tell our audience about the most interesting projects you’re working on now? He: "Recently, I’ve been working on use cases and projects surrounding identity. We have been working to link identity data from various sources – online and offline. Here at Experian, we have information from many sources, across different business areas. This project is providing a platform to link all this data together, which in the past was not very easy to accomplish. With this platform to provide linkages, it provides a 360-degree view of a person and helps provide conclusions such as whether two identities are the same person. To do this, we utilize machine learning techniques and AI. It’s very exciting." Kazmi: "I would like to mention something I'm very proud of, which has been a turning point in the way I look at data science solutions. I have the privilege of playing a prominent role in solving for a crucial economic and societal problem of the world, financial inclusion. This issue has historically blocked growth for financially weak and less established sections of society. I am leading data science as part of the initiative, exploring different sources of information beyond credit history, to increase access to financial products. This is the beauty of data science and how it helps us." Kung: "At Experian, I work in a consulting area, so I advise our customers and show them the power of data. Often, it’s not easy for a client to recognize this power. That’s our job – showing them how data can help their business or their decisions. We developed a credit decisioning model for one client using machine learning. This showed them how powerful it can be to use the data we make available to them. They were so amazed with the results. It was a really great experience." Peters: "The newest aspect of my role is leading innovation and strategy for decision analytics in North America. I am constantly on the watch for opportunities to incubate and try to apply Experian’s data and analytics and AI capabilities to solve new problems. We are looking at the role of identity and how we might apply capabilities in new ways. There is an expansion of needs, especially as the world evolves, and how we’re identified is evolving. So the application of Experian’s differentiated capabilities to new areas and markets is an area of focus of mine that I'm really excited about right now." Stoddart: "One of the most interesting projects I've worked on since joining the lab is around fairness of machine learning algorithms, decision-making. It’s about tackling the bias that can come when you use machine learning in a real world scenario. This happens when an algorithm is not being checked properly and it's discriminating against a certain group. To be part of building this vision about treating everybody fairly is great. Especially to be part of a company that values this effort and recognizes that it's going to be increasingly important going forward." Related stories: What is the right approach to AI and analytics for your business? Four fundamental considerations Maximizing impact from AI investment: 4 pillars of holistic AI Forbes: Are we comfortable with machines having the final say? Yi He Yi He works as a data scientist in the Experian NA DataLab. She is dedicated to using machine learning and AI to extract information from large amounts of data to identify, understand and help people, and prevent fraud. She aims to bridge online and offline worlds by linking identity data from these unique sources. With a focus on minimizing friction to customers, Yi’s work helps organizations identify synthetic identities to avoid fraudulent applications. Recently, she contributed to a Covid Outlook & Response Evaluator (CORE) Model – a “heat map” of geographic populations across the U.S. most susceptible to severe cases of Covid-19. Deeba Kazmi In her role as a data scientist at the Experian APAC DataLab, Deeba Kazmi is focused on solving business problems with analytics, including the development of consumer and small to medium enterprise credit risk models that leverage alternative data. Deeba is passionately focused on leveraging AI to create solutions that can help address issues faced by developing markets. Most prominently, this work includes her data science leadership contributions to solving a crucial economic and societal problem – financial inclusion. This effort is helping disadvantaged socio-economic consumer groups gain access to vital credit and financial services by leveraging the power of technology to deliver better outcomes. Jennifer Kung Jennifer Kung is an analytics consultant for Serasa Experian Decision Analytics, where she combines her knowledge of financial services with her data analysis expertise. Jennifer aims to harness the power of data through robust, descriptive and predictive analytical solutions to help clients realize the benefits of the massive amounts of data available to them. She recognizes the magnificence in powering discoveries through data analysis and enjoys revealing these capabilities to businesses who can benefit from these robust, yet approachable solutions. Jennifer enjoys knowing that her work helps to simplify and accelerate decisions that consumers rely on at important times in their life. Kathleen Peters Kathleen Peters leads innovation and business strategy for Decision Analytics in North America. As the prior Head of North America Fraud & Identity business, Kathleen is well-recognized as an identity industry innovator, being named a “Top 100 Influencer in Identity” by One World Identity the last two years. As of 2020, Kathleen was named Chief Innovation Officer for Decision Analytics. Kathleen and her team rely on the power of AI to continuously find new ways to solve customer challenges by defining product strategies, new paths to market and investment priorities. Underlying these efforts is a key focus on the ethical use of technology and the need to be consciously unbiased. Laura Stoddart Laura Stoddart is a physicist turned data scientist who works at the Experian DataLab in London. From her first exposure to AI, she recognized how quickly it can have an impact on the world, which has driven her to get and stay involved in the industry – both professionally and personally. Laura’s recent work has focused on ethical AI, having recently contributed to her first paper addressing the removal of bias from models. In addition, she is concentrated on leveraging emerging datasets to evaluate risk. Outside the DataLab, Laura also volunteers her data science skills to good causes such as Bankuet and helps expose others to the world of AI through mentoring.

Public and private organizations worldwide are embarking on ambitious digital identity initiatives, from the tiny country of Estonia to efforts that encompass much of Africa and India. At the core, the broad goal is often the same: Use blockchain or equivalent technology to provide individuals with a unique digital identifier. That digital identity then enables seamless, secure access to services—governmental, financial, or otherwise. However, as you delve into the details of each program, there remain more differences than similarities. Organizations may have different drivers for pursuing digital identities and varied approaches. And in these early days of digital identity development, there’s not yet a single plan for aligning initiatives across the public and private sector or even within the financial services industry. So how do organizations evaluate where to invest and when to act, when efforts are progressing and changing in real-time? The impetus now is to understand the fundamentals of digital identity programs and then evaluate what your organization stands to gain—or potentially lose. Get that sorted, and you’ll be ready to make smart digital identity decisions at the right time for your company and customers. The fundamentals of blockchain Much of the digital identity conversation centers around the notion of blockchain-based digital identity programs and their benefits to consumers or citizens. Broadly, these programs enable individuals to have a digital identity profile, which is tied to a basket of attributes and stored on a blockchain. Those attributes are verified when the identity is established. Consumers then use their digital ID, for example, to access their financial applications. And organizations can verify the person via their digital identity token. Such programs provide privacy for consumers; they also promise to accelerate and secure all sorts of processes from applying for loans to paying taxes. That’s because, with a digital identity, consumers don’t need to re-submit documents or provide personal information to various businesses and entities. Instead, they can allow institutions to access their digital identity for proof of who they are. The potential for such programs is already exciting, and we’ve likely just scratched the surface of what’s possible. Still, most of the discussions leave out a critical component. That is: how will programs establish a digital identity in the first place? As financial institutions assess the digital identity landscape, digging into how programs ensure that the right information makes it into the system is paramount. As the saying goes, it’s garbage in, garbage out. Regardless of how innovative the technology is, a consumer’s digital identity is only as ​trustworthy as the information that created it. The digital identity trade-offs The security of digital identities is very compelling—especially as cybercriminals become increasingly sophisticated. Businesses can easily authenticate customers, and consumers have more control over their information, which is an issue of growing importance. A recently released Experian study shows that consumers are most concerned about protecting their financial data over other types of information. As privacy and security assurances become part of the financial service value proposition, digital identity programs will likely be a differentiator for companies. That said, doubling down on digital identity can initially seem at odds with another dual technology priority: Taking advantage of data to provide hyper-personalized financial products and services. By tokenizing identity information, organizations may need to forgo some of the data that enables that personal, customized approach. In the long run, I believe companies will find creative ways to balance privacy with personalization needs. For instance, customers may rely on digital identities to navigate their financial networks and then opt to provide additional information about themselves in return for better, more personalized service. Financial institutions will need to weigh some similar factors when leveraging digital identity programs to improve customer experience. Digital identity programs promise to remove the friction caused by customer recognition and authentication. Again, the organization may give up some data collection to enable that seamless experience. But in the long run, companies will likely find that the related improvements and revenue opportunities gained more than makeup for any sacrificed information. ​At the same time, against a backdrop of an increasing number of stolen identity records, the idea that a digital identity program can help reduce the excessive proliferation of sensitive personal data is a significant benefit. The road ahead Financial institutions should prepare for the pending digital identity journey—even if they haven’t yet embarked. There are still multiple issues that the industry, consumers, and regulators will have to settle. For instance, there’s the question of adoption and how long it will take for businesses and consumers to use digital identity programs regularly. As we’ve discussed before, consumer trust ​and availability will remain a considerable component in driving that adoption. What’s more, we’ll likely see regulations follow digital identity efforts as specific initiatives gain steam and popularity. The rules may accelerate adoption or, conversely, increase the investment expense on behalf of financial service firms. For these reasons, financial institutions need to be involved early and voice their concerns often to ensure that regulations serve consumers without adversely affecting the business. In the meantime, businesses should remain aware that digital identity is a fragmented market, which may ultimately settle into an “ecosystem of ecosystems” across programs. It will be critical for enterprises to plan accordingly if they want to become early adopters. Or, at the very least, companies with a more moderate strategy should wait until a leading program emerges before making a significant investment. Digital identities represent a dramatic shift in how consumers navigate their online world and how companies continue to meet their online expectations and needs. Keep these developments on your radar, and you’ll be prepared to make smart digital identity decisions and investments. Related stories: Infographic: Global Identity & Fraud Trends, February 2020 The impact of Covid-19 on Consumers and Businesses, July 2020 The impact of Covid-19 on Consumers and Businesses, Oct 2020

In this episode of Insights in Action, we talk with Derek Garriock, Design & Innovation Director at Experian and David Bernard, SVP of Global Marketing & Strategy at Experian Decision Analytics, about the future of banking and trends and opportunities arising in the post-Covid-19 crisis era. The future of banking is being shaped, in part, by people's response to Covid-19 There is adaptation to the current crisis, but even as we start to progressively get out of lockdown in a number of countries, banks have realized there are a number of deeper things around their use of analytics, the fine-tuning of their scorecards, lending strategies and risk strategies that have to be redone. Also, there’s the general, longer-term trend towards moving some of their banking structure to the cloud, making sure that their decision strategies are fit for purpose, that they are flexible enough, building attributes into the system. So, there are a number of programs that are continuing and sometimes accelerating. David Bernard, SVP, Global Strategy & Marketing, Experian Decision Analytics Questions answered include: Are we already on the path to a different way of banking? Speed, convenience, and choice have gained a different meaning, accelerating digitalization efforts and demands virtually overnight. What are the current areas of focus for the industry based on experiences with financial institutions globally? Has this Covid-19 crisis further challenged the status quo in the industry and what is the anticipated impact between traditional financial services and fintech challengers? What are the pillars of a successful modern banking infrastructure, and what promising technologies can help meet new market dynamics? Related content: The role of the virtual assistant: What businesses can do to ensure consumer demand is met while taking care of customer experience Maximizing impact from AI investment: 4 pillars of holistic AI Be mindful of these 3 Strategies when engaging customers digitally

As consumer organizations settle into the so-called new normal, behaviors have dramatically changed and expectations have been redefined. Speed, convenience, and choice have gained a different meaning, accelerating digitalization efforts and demands virtually overnight. Recently, we spoke with our internal experts – Derek Garriock, Design & Innovation Director at Experian and David Bernard, SVP of Global Marketing & Strategy at Experian Decision Analytics – about the future of banking and trends and opportunities arising in the post-crisis era. Here’s highlights of that discussion: A different way of understanding and doing banking – a viewpoint by Derek Garriock Industries are redefined by changes in consumer behavior, and certainly, the crisis that’s been unfolding across the globe has had a big impact in terms of how we live our day-to-day lives. These changes are reflected in the demands made of banks, as we try to manage our money in a different way. The challenge that the banks and lenders have seen across the globe is obviously different levels of reaction from consumers and businesses — depending on the jurisdiction that they’re in and the immediate need that’s created. This challenge is more about how you are able to adapt given that going forward this behavioral change will be no doubt be one of the lasting impacts of the crisis. At a very basic level for banks, we still have some of the pre-existing challenges around how they deliver change in a digital world to really serve customers and give them the best possible experience and journeys to serve their needs. Obviously, there’s a lot of regulation banks are required to observe and follow as an organization doing the type of business that they do — but the current needs shine a light on big areas of focus that probably haven’t changed in the last decade — around how do you digitize your business to reduce cost, to better serve your customers, and to be in a place where you drive deeper share of wallet with customers to grow your business. What we’ve seen through the crisis is really a spotlight shone on this area and in the context of how to move quicker, faster, better, and to deliver against some of those core objectives. Current areas of focus for the global banking industry – a viewpoint by David Bernard Thinking about the immediate reaction to the crisis, we have a number of banks that are still focused on coping with lockdowns and business continuity across the globe — managing going into lockdown and out of lockdown across different countries. For example, we had banks in the UK that have call centers in India. When the India lockdown happened, very suddenly they lost their ability to respond to clients over the phone — so we see some immediate impacts of the crisis with banks coping with a situation where different parts of the globe are challenged from a business continuity perspective. Banks also had to adapt to a number of government programs such as government-sponsored loans for small businesses and individuals. They had to adapt details like their scorecards for lending, or their scorecards for debt collections — evaluating their approach to debt collections since suddenly you have a lot of people that lost their jobs. Asking for last month’s bank statements doesn’t quite give you the right view of their personal situation. There is adaptation to the current crisis, but even as we start to progressively get out of lockdown in a number of countries, banks have realized there are a number of deeper things around their use of analytics, the fine-tuning of their scorecards, lending strategies and risk strategies that have to be redone. Also, there’s the general, longer-term trend towards moving some of their banking structure to the cloud, making sure that their decision strategies are fit for purpose, that they are flexible enough, building attributes into the system. So, there are a number of programs that are continuing and sometimes accelerating. There is also the example of digital interfaces where it looks like you can do something in an app on the website, but behind the scenes, a number of banks have analog processors — non-digital processors — where there are people reading data internal in the system or doing some manual task behind the scenes and the whole crisis is shedding light on those examples and forcing more complete digitization across the board. Listen to the full podcast: https://bit.ly/IIA_FutureFS Related articles: Digital transformation through cloud-first decisioning by Chris Fletcher, SVP Decision Management & Cloud Services & David Britton, VP Of Industry Solutions Maximizing impact from AI investment: 4 pillars of holistic AI by Shri Santhanam, Global Head Of Advanced Analytics & AI & Birger Thorburn, Chief Technology Officer, Global Decision Analytics How rapidly changing environments are accelerating the need for AI and Machine Learning in business by Birger Thorburn, Chief Technology Officer, Global Decision Analytics

Due to Covid-19 , the focus on analytics and artificial intelligence (AI) has significantly increased. However, while companies have made significant investments in AI, many are struggling to show a tangible impact in return. One executive commented, “We have data science teams and a data lab where advance techniques like neural networks, GANs, etc. are successfully being used. However, less than 10% of our actual operational decisions and products are powered by AI and machine learning (ML). I would like us to be driving greater measurable impact and Covid-19 is exposing some of our execution gaps.” And, he’s not alone. Despite the investment, the true impact is elusive, and many businesses are not getting the desired effect from their efforts. Achieving the results needed to justify continuous investment will take a holistic approach. So, what can companies do to achieve this impact? The four pillars of holistic AI: performance, scaling, adoption and trust Achieving impact from AI requires taking a more holistic approach across four pillars — beyond just the delight of the data scientist producing a better performing model. 1. AI performance — outperforming the status quo and quantifying the impact This pillar is where most data scientists and companies tend to focus first, for example using modern AI techniques to create an underwriting model that performs better than traditional models. The so-called ‘data science moment of truth,’ where the data scientist declares that he has built a model which outperforms the status quo by 10%. However, it’s important to note model performance alone is not sufficient. We should look beyond the model to understand business performance. What quantifiable business impact does the 10% improvement deliver? How many more credit approvals? How much lower will the charge-off rate be? This reasoning provides the important business context around what the incremental performance means. 2. AI scaling — having the right technical infrastructure to operate models at scale This area is often ignored. The risk with data science teams is they can see their job as being completed with creating a better performing model. However, that’s just the beginning. The next important step is to operationally deploy the model and setup the operational infrastructure around it to make decisions at scale. If it is an underwriting model, is it deployed in the right decisioning systems? Does it have the right business rules around it? Will it be sufficiently responsive for real-time decision making, or will users have to wait? Will there be alerts and monitoring to ensure that the model doesn’t degrade? Are there clearly defined, transparent and explainable business strategies, and technology infrastructure and governance to ensure all stakeholders are aware? Is the regulatory governance around this model in place? Does the complexity in the model allow it to scale? Too often we see data scientists and data labs create great models that can’t scale and are impractical in an operating environment. One banking executive shared how her team had developed 5 machine learning models with better performance, but were in ‘cold storage’ verse in use, because they didn’t have the ability to scale and operationally deploy them effectively. 3. AI adoption — ensuring you have the right decisioning framework to help translate business decisions to business impact With better performing predictive models and the right technology, we now need to present the information in a way that is ‘human-consumable’ and ‘human-friendly.’ At one bank, we found they built a customer churn ML model for their front lines, but no one was using it. Why? They didn’t have the contextual information needed to talk to the customer — and the sales force didn’t have faith in it — so didn’t adopt it. Subsequently, they built a model with a simpler methodology and more information available at their fingertips — where decisions could be made. This was immediately adopted. This pillar is where the importance of decisioning tools is highlighted. The workflow and contextual information to allow a decision to be orchestrated and made is critical in driving AI adoption. 4. AI trust – having governance, guardrails and the appropriate explainability mechanisms in place to ensure models are compliant, fair and unbiased This final pillar is probably the most important for the future of AI — getting humans to trust it. In recent times we have seen numerous examples like the Apple Card, where the underlying principles and models have been called into question. For scalable AI impact, we need an entire ecosystem of people who can trust AI. To achieve this effect, you need to consistently apply the right principles over time. You also need the right decisions to be explained — like adverse action calls. Explainability capabilities help manage communication and understanding of advanced analytics, contributing to established AI trust. And, when fairness and bias issues come up, you need to provide good answers as to why decisions were made. AI is poised to fundamentally change the way we do business, and studies show that $3 to 5 trillion in global value annually, up to $15 trillion by 2030, is likely to be created. We believe the four pillars highlighted above will be key to accelerating the journey to driving positive results and capturing this value. At Experian, we are making investments to drive impact for our clients by delivering against these four pillars. Related articles: What is the right approach to AI and analytics for your business? Four fundamental considerationsHow rapidly changing environments are accelerating the Need for AI What’s new in online payment fraud Part 2: How AI and evolving regulation are driving change