Ready to get your mind blown? Listen to an AI generated short podcast on this research here
Welcome to the speed dating future of finance, where artificial intelligence isn’t just a buzzword thrown around at tech conferences, but a powerhouse transforming the financial sector as we know it. You either innovate or become irrelevant to your clients. If you thought AI was only about self-driving cars and robots playing chess, think again. In the world of finance, AI is making waves in everything from trading algorithms to customer service, fraud detection and personalized finance. Forget the days of dusty paper ledgers: AI is here to help financial institutions run faster, smarter, and safer.
Here’s what you’ll find in this research:
- AI in Investment Strategies: How AI is shaking up algorithmic trading, optimizing portfolios, and predicting market moves with impressive accuracy.
- AI in Risk Management: From fraud-busting tech to predicting cyber threats before they even happen, AI is protecting financial institutions and your personal data.
- AI in UX (Customer Experience): Imagine walking into a bank where the service is fast, personal, and kind of… friendly. That’s AI-powered banking solutions, faster loan processing, and predictive insights.
- AI in Operational Efficiency: Cutting costs, streamlining back-office chaos, and using predictive analytics to optimize everything from staffing to resource allocation. Bye-bye, inefficiency!
- Challenges and Barriers: AI has potential, but there are still some bumps in the road, like data security risks, AI bias, and the high cost of implementation (spoiler alert: it’s not cheap).
- Impact on Financial Inclusion: AI is breaking down barriers and opening doors for underserved communities to access essential financial services.
- Future Opportunities: We’re talking about AI-driven green finance, decentralized finance, and supercharged wealth management strategies.
A recent survey carried out by NVIDIA found that “an overwhelming 91% of financial services companies are either assessing AI or already using it in production.” In terms of what AI is already being used for, they found AI is most used for data analytics, generative AI, and predictive analytics, as shown in Figure 1 below. The columns in light green are from 2023, while the dark green represents 2024.
Figure 1. NVIDIA. What AI and machine learning workloads is your company using or assessing (select all that apply)?
In addition, a survey by KPMG also found that executives in financial services companies planned (in 2023) to use AI increasingly for fraud detection and prevention, customer service and personalization, and compliance and risk.
Figure 2: KPMG. How financial services executives plan to use generative AI.
But let’s not get carried away on a wave of techno-optimism. There are still plenty of challenges to tackle.
The bottom line is, AI is not just some flashy trend; it’s the financial sector’s secret weapon to tackling today’s biggest problems and setting itself up for success in the future. So, buckle up. AI is here to stay, and if you’re not on board yet, you’re about to be left behind in the financial revolution. Ready to get into it? Let’s get started!
AI in Investment Strategies
AI tools are already being used for many different investment strategies and will continue to expand. For example, AI-powered algorithmic trading has already become a cornerstone of modern financial markets. In 2021, AI-driven trading systems already accounted for approximately 60-73% of U.S. equity trading volume. These systems use machine learning algorithms to analyze large datasets, identify patterns, and execute trades at speeds that human traders just can’t reach.
Also, firms like BlackRock have integrated AI into their trading strategies to improve their decision-making processes. Part of this has been done by training a proprietary AI model to analyze earnings call transcripts and forecasting the market reaction that is likely to follow. As shown in Figure 3 below, BlackRock found their model was able to predict post-earnings market reactions with high accuracy.
Figure 3: BlackRock. Accuracy of models at forecasting 40-day post-earnings stock returns.
Moreover, the rapid advancements in big data infrastructure and cloud computing have empowered AI systems to process vast amounts of financial data almost in real-time. This capability not only enhances the precision of algorithmic trading but also enables the creation of dynamic strategies that can quickly adapt to volatile market conditions.
The increasing adoption of AI in trading is partly because of advancements in natural language processing and predictive analytics. These improvements allow the analysis of unstructured data sources such as news articles and social media feeds to inform trading decisions, like in the BlackRock example above (with earnings call transcripts).
AI has also revolutionized portfolio management by enabling sophisticated risk assessment and optimization techniques. Investment firms are increasingly deploying AI to analyze historical performance, current market conditions, and financial health indicators to construct optimized portfolios.
For example, BlackRock’s Aladdin platform uses AI (called “Aladdin Copilot”) to assess risk factors and simulate various market scenarios. This helps them to create tailored investment strategies. These AI-driven approaches also allow for quick adjustments to portfolios. The use of machine learning models in this context also includes continuous learning from new data, improving predictive accuracy and investment outcomes over time. Other companies like Betterment and Wealthfront use automated investing tools, with analytics-based and AI-supported trading, rebalancing, and reinvestment.
Understanding market sentiment is crucial for anticipating market movements, and AI has become an invaluable tool in this area. By analyzing large volumes of unstructured data, such as social media posts, news articles, and earnings call transcripts, AI systems can gauge investor sentiment and predict market trends. JPMorgan Chase, for instance, uses AI to analyze the emotional tone of financial reports. This is then used to help traders make more informed decisions. They also work with IBM to integrate AI into models to improve risk management. Other companies like Numerai (who have taken a very radical approach) rely on AI-based market prediction models built by anonymous coders, who compete (through the performance of their models) to win Bitcoin.
The large leap in the use of generative AI is partly because it can increasingly and accurately analyze complex market scenarios and stress-test investment strategies. Firms like Goldman Sachs are exploring the use of quantum computing and generative AI to conduct Monte Carlo simulations, generating a multitude of potential future market states to assess the resilience of portfolios under various conditions. These models can incorporate a wide range of variables, including macroeconomic indicators and geopolitical events, providing a more comprehensive view of potential risks and opportunities.
As AI technologies continue to evolve, their role in investment management is likely to expand, driving innovation and potentially leading to superior investment outcomes. NVIDIA has already reported that AI has increased revenue and decreased costs for most financial services firms, as shown in Figure 4 below:
Figure 4: NVIDIA. AI Is Driving Business Results for Financial Firms.
These benefits are likely to increase as AI tools become more efficient and can work with larger and more complex data sets.
AI in Risk Management
Alongside its application in investment strategies, AI is also being increasingly used to mitigate and manage risks in the financial sector. By analyzing large datasets in real-time, AI systems can identify activities that indicate fraud. For instance, HSBC has heavily invested in AI for financial crime detection. These approaches allow financial institutions to respond more quickly to potential threats, reducing financial losses and protecting customer assets.
Startups like Darktrace use data on companies’ existing activities and patterns to build models that can then send alerts when even subtle but dangerous inconsistencies are discovered. Other companies like Feedzai create individual risk profiles and behavioral biometrics to detect fraudulent transactions and weed out false positives.
In credit scoring and underwriting, AI models also analyze large data sets to predict creditworthiness more accurately. Startups like Zest AI claim to provide 2-4x more accurate risk ranking when offering underwriting services, and that they can accurately assess 98% of American adults. This is carried out with AI models trained on customer data. AI models can create a more accurate picture of an applicant’s financial behavior, potentially leading to fairer lending decisions. AI tools can also process non-traditional data, such as utility payments and social media activity, to evaluate credit risk.
Regulatory compliance is another important aspect of financial operations, and AI-powered solutions are streamlining these processes. According to KPMG, 68% of financial services firms use AI for risk management and compliance. AI systems can monitor transactions, analyze communications, and ensure compliance with complex regulatory frameworks. This reduces the likelihood of breaches and resulting penalties.
The financial sector also faces increasingly sophisticated cyber threats, with the increased use of hacking tools and other cyberattacks from state and non-state actors. AI tools can help to improve cyber security and protection, by analyzing network traffic patterns to detect anomalies that may indicate cyberattacks. For instance, AI can predict potential security breaches by identifying unusual user behaviors or unauthorized access attempts, allowing institutions to take proactive measures.
AI in Customer Experience
Investment strategies and risk management are not the only areas of the financial sector that AI is improving. AI tools are increasingly used to create personalized financial products, by analyzing customer data to understand individual preferences and behaviors. AI algorithms can assess spending patterns and financial goals to recommend tailored investment options or savings plans that suit each customer.
In addition, conversational AI such as chatbots and virtual assistants, is also being used more and more for customer service in banking and finance. These AI-driven tools handle inquiries from account balances to loan applications and can be available 24/7 and at a much lower cost than having a large staff of customer service employees. Tools like KAI, made by a startup called Kasisto, can search banks’ and financial institutions’ information networks to provide fast and reliable answers to their customers. This reduces call volumes, saves time, and minimizes administration costs.
The Commonwealth Bank of Australia has recently taken steps to use an AI agent for its business customers, in collaboration with Amazon’s AI Factory. The AI agent is intended to enhance transaction efficiency and customer experience and position CBA at the forefront of AI adoption in banking. The AI tool is still only a trial though, and more experiments will need to take place to figure out how these tools work for real customers.
For customer service behind the scenes, AI tools can also improve loan processing by automating data analysis and decision-making. Approaches like embedded lending (EL), in which loans or credit are offered through non-financial platforms such as retail, e-commerce, or travel services, allow customers greater and more streamlined access to financial products. AI is increasingly used to facilitate such approaches. Ernst & Young also report, in Figure 5 below, that customers are becoming increasingly aware of such approaches and are more willing to use them than even just a couple of years ago:
Figure 5. Ernst & Young. Increased customer awareness and usage.
By evaluating creditworthiness using diverse data sources, AI can make it faster to approve (or decline) loans and can minimize human biases. Online platforms like Blend can be integrated into financial institutions to provide cloud-based financial services and lending tools that can be accessed easily through a platform, instead of needing to make appointments at a physical location.
AI approvals and tools can lead to fairer lending practices and improved access to credit for a broader range of applicants. However, there have been cases in which AI tools have led to more discrimination, such as research from the University of Bath that shows discrimination against women worsens when AI tools are used to approve or deny loans. Banks will need to work these issues out so that AI can be used responsibly for both their business goals and social equity.
Finally, the same tools used for preventing fraud can also be used to improve customer service and offerings. For instance, data analytics using AI can help banks to predict customer behavior by examining transaction histories and interaction patterns (which are then used to detect anomalies). These predictions can also enable proactive service offerings, such as personalized financial advice or product recommendations, improving the customer service for banking customers.
AI in Operational Efficiency
More behind-the-scenes improvements come from AI in back-office operations in the financial sector, through approaches like robotic task automation. KPMG reports that 71% of organizations use AI in their financial processes, such as financial planning, accounting, and tax reporting. As shown in Figure 6 below, you can see that many organizations already have wide adoption of such technologies, with almost all organizations either currently using these tools or planning to:
Figure 6: Fintech News. Why AI Has Become Essential in Banking.
In addition to these measurable gains, many financial institutions are now deploying integrated AI platforms that merge cloud computing with real-time analytics. This integration not only automates routine processes but also provides dynamic dashboards for monitoring key performance indicators, enabling instant adjustments in response to market fluctuations and operational challenges.
By automating simple and routine tasks such as data entry and transaction processing, AI tools can minimize errors and free human workers up to do other tasks. Some financial institutions are deploying automated AI tools to handle compliance reporting and customer onboarding processes (such as “know your customer” steps), saving time and increasing accuracy. Tools like Automation Anywhere provide automation tools for customer onboarding, loan processing, underwriting, fraud detection, portfolio management, document verification, and more. They claim their AI tool can complete nine years of work in 14 days.
The integration of AI in financial services has already led to cost reductions. A survey by Bain & Company found that generative AI applications have already resulted in an average productivity gain of 20%. Manual tasks like document processing and customer service are being made more efficient with AI, reducing the need for manual intervention (i.e. human workers) and lowering operational expenses (i.e. employment costs).
McKinsey also notes that AI is increasingly used across the banking and finance landscape, from “front office” tools such as chatbots and facial scanning, to “back office” tools such as fraud detection and transaction analysis. In Figure 7 below, you can see some additional examples of these types of tools:
Figure 7. McKinsey. Banks are expanding their use of AI technologies to improve customer experiences and back-office processes.
Like in other industries, AI can also be used to improve logistics efficiency. In the finance and banking sector, this is particularly related to cash distribution and payment processing. For example, AI algorithms can predict cash demand at ATMs, ensuring optimal cash levels and reducing replenishment costs. In global payment systems, AI also streamlines transaction routing, improving efficiency in payment flows.
Challenges and Barriers=Opportunities
Although there are many benefits for the sector, there are also several challenges and barriers. One of these is that AI systems rely on large data sets including sensitive financial data, making cybersecurity and privacy a top concern. Cybercrime is already expected to cost the global economy $10.5 trillion annually by 2025, and the financial sector is not immune. AI-powered fraud detection helps to mitigate some risks, but data breaches, such as the Capital One hack in 2019 affecting 100 million customers, show that banks are still incredibly vulnerable to breaches or loss. With more “attack surface” available through the use of AI tools and add-on software that could include back doors or be vulnerable to access, the financial sector needs to ensure that security measures are robust.
Like other sectors, regulatory issues still abound in relation to AI. The lack of standardized regulations creates uncertainty, forcing financial institutions to anticipate and adapt to changing compliance requirements. Luckily, regulatory bodies worldwide are already working to keep pace with AI-driven financial services. The U.S. Securities and Exchange Commission (SEC) and the Federal Reserve are evaluating AI-related risks, while the EU’s AI Act could establish some global rules around AI governance and risk assessment.
KPMG notes that privacy concerns increased for financial services executives from March 2023 (in purple) to June 2023 (in blue), as shown in Figure 8 below, with greater awareness of the privacy and security issues involved with AI. The need for regulatory action is also already clear to executives, who noted that it is necessary for security, data privacy, and information transparency concerning the use of AI.
Figure 8: KPMG. Generative AI privacy concerns.
Another issue, as mentioned briefly above, is that AI-driven financial decisions can inadvertently reinforce systemic biases if models are trained on historical data that reflects existing inequalities. A 2022 study found that Black and Hispanic borrowers were charged higher interest rates than white borrowers, even when AI was used in lending decisions. Ensuring fairness in credit scoring and underwriting will require model training to improve, active oversight to determine biases, and continual model development with new and enhanced (such as synthetic) data sets.
Additional data quality issues and system integration problems can prevent AI tools from being as useful as they otherwise would be. With low-quality data, models output low-quality results. In addition, when systems are not integrated well, data accessibility and functionality can be significantly reduced. In Figure 9 below, you can see the results of an MMA Global report and the slow progress made by many firms on AI deployment and data quality:
Figure 9. The Financial Brand. The data issues coming to the forefront of AI deployment.
The opportunity here is for firms that invest in Explainable AI (XAI). AI models offering a clear explanation for their decisions will build client trust, reduce compliance risks and attract socially conscious investors. While this may not appear to be appealing today given the movement away from DEI, large companies like BlackRock, JPMorgan and Wells Fargo are integrating ethical AI frameworks to enhance transparency. Fintech startups focusing on Fair AI in Lending (Zest AI, Upstart) are already outperforming traditional banks in credit approvals for underserved markets. Investors should watch for companies embracing AI ethics & transparency as a potential competitive edge.
Finally, AI adoption in finance comes with significant costs, including data infrastructure, computing power, and specialized talent to operate relevant AI tools. While many people are increasingly upskilling and learning how to use AI for their benefit (and the benefit of their organization), this still takes time and costs. According to a McKinsey report, major banks already invest billions annually in AI and digital transformation. Smaller firms may have difficulties competing with high-end models and may be unable to spend similar in-house resources to develop their own models (such as the BlackRock in-house model mentioned above). Cloud-based AI solutions and open-source models may help reduce barriers, but the cost remains a challenge for widespread adoption.
Impact on Financial Inclusion
Despite these challenges, AI has the potential to significantly improve financial services, particularly for the 1.4 billion adults worldwide who do not currently use banking services. While the term inclusion brings up certain negative connotations, 1.4 billion is 1.4 billion.
Leap-frog development in African countries and other developing countries has already allowed numerous people to jump straight to mobile banking, microloans, and digital wallets, without ever having experienced old banking systems such as teller windows or waiting in line for cash. AI-driven mobile banking and digital wallets, such as M-Pesa in Africa, allow users to access financial services without traditional banking infrastructure.
In addition, credit assessment models using AI tools can facilitate microloans by analyzing alternative data sources such as mobile payments, utility bills, and social media activity. Companies like Tala use AI to develop personalized financial services in the developing world, from instant credit to loan options, and bill management options. The global microfinance industry, valued at $124 billion, is increasingly using AI to reduce default rates and improve loan accessibility.
The difference between traditional banking and AI-driven banking is huge: traditional lending models often rely on credit scores that exclude many financially responsible individuals. Underwriting using AI tools and customizable models can consider broader financial behaviors, helping to improve financial inclusion. For example, Zest AI (mentioned above) also uses machine learning to make fairer lending decisions, increasing approval rates for underrepresented groups while maintaining low default rates.
Future Opportunities
There is also a wide range of future opportunities for AI to grow in the financial services and banking sector. One of these is the role of AI in ESG investing and sustainability reporting. With global ESG assets projected to reach $53 trillion by 2025, AI is playing a key role in evaluating ESG risks. (Despite the recent moves by the US, this remains a robust growth opportunity especially as the frequency and severity of storms and fires occur.) AI-powered analytics assess company sustainability metrics, detect greenwashing, and optimize ESG portfolios. For example, BlackRock’s Aladdin platform uses AI to analyze climate risks and ESG factors in investment strategies. Companies like Sylvera also use data on carbon emissions, capture, and avoidance, to give ratings for projects from an ESG and sustainability perspective.
Looking ahead, emerging technologies such as quantum computing and edge AI promise to further revolutionize financial operations. Quantum algorithms could dramatically enhance the speed and precision of complex risk simulations and portfolio optimizations, while edge AI solutions, processing data locally rather than in centralized data centers, offer improved security and reduced latency for real-time trading and fraud detection.
In addition, AI is being increasingly used with decentralized finance (DeFi) systems to enhance risk management, optimize trading strategies, and detect fraud. DeFi is a blockchain-based financial system that allows customers to trade, lend, borrow, and earn interest without banks or intermediaries. It relies on smart contracts, such as by using Ethereum. Smart contracts automate lending and yield farming decisions, while blockchain-based AI solutions like SingularityNET create decentralized financial intelligence. AI is also being used to detect illicit activities in crypto markets, strengthening compliance efforts. Others like Sahara AI are creating tools that allow AI to become better integrated with blockchain tools.
Financial institutions are also increasingly using AI to develop predictive models for economic trends, risk assessment, and long-term financial planning. Macroeconomic forecasting tools can also analyze complex datasets to provide strategic insights. These models help institutions prepare for market shifts and economic downturns.
Banks also haven’t just reserved chatbots and service bots for customers: robo-advisors are also being used for high-net-worth individuals. Personalized AI assistants analyze your spending habits, investment goals, and tax strategies, and can create financial plans for you to follow. Whether or not these are any good, time will tell.
Startups like Albert have already begun producing apps that automate your budgeting, savings, and investments. Other tools using AI for portfolio rebalancing can also already be used to change asset allocations based on real-time market conditions and can be supplemented with more real-time information-gathering tools.
Other tools like Dataminr’s AI platform can already analyze information on news events (such as the CrowdStrike outage and Baltimore bridge collapse), providing information even before it is reported in mainstream media. When these tools are integrated with portfolio and market information, the financial sector companies who take advantage of them have speed on their side.
The financial sector’s AI transformation is accelerating, with spending on AI in the sector expected to increase at a rate of 18% per year according to Forrester’s 2022 Global AI Software Forecast. While challenges remain, particularly in regulation, data security, and ethical AI, financial institutions that adopt AI are highly likely to reduce risks and costs, while increasing economic growth.
Discussion
Financial institutions that invest in AI early are likely to benefit from faster decision-making, improved risk management, and personalized customer experiences, all of which increase revenue growth and operational efficiency. A 2023 PwC report states that AI could contribute $1.2 trillion to the global banking industry by 2030, with early adopters benefiting the most.
Statista predicts that AI in the finance sector will continue to grow, continually expanding at a steady rate, as shown in Figure 10 below. Statista notes that “the banking sector, in particular, has emerged as a front-runner in AI investments, allocating 20.6 billion U.S. dollars in 2023 alone.”
Additionally, as AI systems are deployed at scale, ongoing investment in robust cybersecurity measures and ethical oversight will be paramount to protect sensitive data and mitigate the risk of biased decision-making. Establishing cross-functional oversight committees that include experts in technology, compliance, and ethics is emerging as best practice for ensuring that AI remains a force for good in the industry.
Also, the integration of AI across the financial sector also calls for a paradigm shift in workforce skills and management strategies. Financial institutions are now investing heavily not only in cutting-edge technology but also in upskilling and reskilling their workforce to work effectively alongside AI. This dual focus on technology and talent is essential to fully capture the strategic advantages that AI offers.
Figure 10. Statista. Market size of generative artificial intelligence (AI) in the financial services industry from 2022 to 2023, with a forecast until 2033 (in billion U.S. dollars).
The increasing use of AI in finance brings risks, particularly around algorithmic bias, data privacy, and regulatory compliance. Transparency in AI decision-making is crucial to maintaining public trust and avoiding negative consequences. Initiatives such as the EU’s AI Act and the U.S. National AI Initiative reflect an increasing concern around responsible AI governance, particularly in sensitive areas such as finance and wealth. Banking and finance firms need to use AI tools responsibly, to ensure fairness and accountability.
AI’s transformative potential in finance also extends beyond individual firms: it requires collaboration between financial institutions, technology providers, regulators, and policymakers. Joint efforts in cybersecurity, fraud prevention, and AI ethics can strengthen the financial ecosystem and lead to rapid growth. Firms like Goldman Sachs predict that AI will have a “potentially large impact on productivity and growth, which our equity strategists estimate could translate into significant upside for US equities over the medium-to-longer term.”
Partnerships between fintech startups and traditional banks, as seen with Visa’s acquisition of AI fraud detection company Featurespace, also increasingly highlight how cooperation and collaboration can be used to increase innovation. Governments also play a role in funding AI research, setting industry standards, and addressing regulatory gaps to ensure AI’s benefits are widely accessible.
Finally
AI is no longer just a futuristic concept: it is already an integral part of the financial system, reshaping daily services as far-reaching as investment strategies and risk management to customer experiences and operational efficiency. With AI-driven finance projected to expand rapidly, firms that resist adoption risk losing relevance. The next decade will be defined by how well financial institutions leverage AI to stay ahead in an increasingly data-driven economy.
While AI in operational efficiency brings significant benefits in terms of cost savings, error reduction, and speed, its true value lies in the strategic insights it offers. As the financial industry continues to evolve, those institutions that successfully balance technological innovation with robust ethical and regulatory frameworks will lead the next phase of growth and transformation.
While barriers such as cost, compliance, and trust remain, those who embrace AI with a well-structured approach will benefit significantly, while contributing to progress in the industry itself. As AI continues to evolve, its role in finance will also expand beyond automation and predictive analytics. Emerging areas such as quantum computing, decentralized finance, and predictive analytics will shape the next phase of financial innovation. While AI cannot eliminate uncertainty in markets, it significantly improves decision-making, reduces inefficiencies, and allows decision-making to take place in real-time.
The institutions that not only adopt AI but also invest in human capital and maintain rigorous oversight will be best positioned to capitalize on the digital revolution. Innovation or irrelevance.
