Banking GenAI #12 - Insights for COOs
Welcome to the latest edition of GenAI for Banking COOs. We hope you had a good summer. After a brief summer break, we are back tracking the latest in the world of GenAI for Banking. In this issue, we explore both the benefits and rewards of GenAI as well as the risks including:
**********************
1. Morgan Stanley Launches AI Assistant 'Debrief' for Wealth Advisors
Morgan Stanley continues to double down on its momentum in AI solutions with the launch of AI @ Morgan Stanley Debrief, a GenAI-powered tool designed to enhance the efficiency and effectiveness of its Financial Advisors.
AI @ Morgan Stanley Debrief serves as a virtual assistant during client meetings. With client consent, it acts as an automated notetaker, capturing and summarizing key points discussed during meetings, leading Financial Advisors to save approximately 30 minutes per client meeting, and allowing for deeper, more meaningful client conversations and improved decision-making during meetings. The AI also identifies and highlights action items, ensuring important follow-ups are not overlooked. Post-meeting, it generates a draft email summarizing the discussion, which advisors can review and personalize before sending to clients. Furthermore, the tool automatically saves meeting notes into Salesforce, streamlining record-keeping and client management processes. The rapid adoption rate, with 98% of Financial Advisor teams using the AI Assistant, underscores its value and user-friendliness.
As AI continues to evolve, we can expect to see more innovative applications that not only improve FA productivity but also elevate the quality of financial advice and client service. For leaders in Wealth Management, the message is clear: embracing AI is essential for staying competitive. The future of wealth management lies in the seamless integration of AI technologies with human expertise, creating a synergy that benefits advisors and clients alike.
**********************
2. Goldman Sachs Deploys Its First Generative AI Tool Across the Firm
Isabelle Bousquette from Wall Street Journal reported that Goldman Sachs has successfully deployed its first Gen AI tool, focusing on code generation and seeing enhanced developer efficiency of upto 20%. This code generation is the first to be deployed out of the multiple use cases being developed by the bank, which has centralized all its AI technology on an internal platform that integrates models form OpenAI, Microsoft, Google, and Meta. According to the bank, this approach meant a slower start for use case development but currently accelerates the process by allowing rapid creation and fine-tuning of applications built-in safety and regulatory compliance measures. The bank's centralized AI platform offers a strategic approach to managing AI development. While numerous AI initiatives remain in pilot stages, the successful deployment of this tool signals a shift towards broader adoption within the organization. The centralized approach also enables ‘build once and use many’ strategy such as leveraging portions of code (and associated controls) written for ‘Copilot for Investment Bankers’ to be used for ‘Copilot for asset Managers’.
Goldman Sachs' experience underscores the benefits of a centralized AI strategy and the potential for substantial efficiency gains through Gen AI. This deployment is a key milestone for a GSIB in its Gen AI journey, and the bank plans to continue to increase investments for the development of Gen AI. By investing in AI infrastructure and fostering a culture of experimentation, peer banking COOs can position their organizations too to capitalize on the opportunities presented by generative AI.
**********************
3. bunq partners with NVIDIA to fight financial fraud with AI
Financial fraud has become a significant challenge for financial institutions worldwide. As fraudsters become more sophisticated, traditional rule-based systems struggle to keep up, resulting in high false-positive rates and labor-intensive processes. European neobank bunq has taken a proactive approach by leveraging NVIDIA's GPU-accelerated computing and AI models to help detect fraud and money laundering. bunq's use of NVIDIA RAPIDS has enabled it to train AI fraud-detection models nearly 100 times faster, improving accuracy and reducing false positives. With the use of accelerating computing and the company’s proprietary large language model, more than half of user tickets are now handled automatically, reducing the need for manual intervention and allowing staff to focus on more complex tasks. Additionally, the technology is now used to detect fake IDs during customer onboarding and automate marketing efforts, further streamlining operations
In addition to handling fraud, this European neobank has leveraged AI and GPU-accelerated computing to revolutionize its operations, driving profitability and substantial customer growth. Contributing to this is Finn, a sophisticated personal AI assistant that enhances customer experience through personalized interactions. Finn's capabilities extend beyond basic inquiries, allowing customers to ask nuanced questions about their financial activities. For instance, customers can effortlessly inquire about their monthly grocery expenditure or recall the name of a specific restaurant they visited recently.
By embracing these advanced tools, bunq has not only enhanced its fraud detection capabilities but also improved overall operational efficiency and customer experience. bunq reported its first full year of profitability in January and customer numbers now exceed 12 million with deposits in excess of €8bn. As financial institutions worldwide grapple with the dual challenges of combating fraud and staying competitive in a rapidly evolving market, bunq's success story offers a roadmap for leveraging AI models and sophisticated hardware such as NVIDIA GPUs.
**********************
4. Central banks turn to generative AI for enhanced cybersecurity
Arijit Sarkar at Cointelegraph.com brought to our attention a recent survey by the Bank for International Settlements (BIS) that underscores the growing adoption of generative AI (Gen AI) among central banks, with a particular focus on cybersecurity applications. The report reveals that 71% of surveyed central banks are already using generative AI for cybersecurity applications, with another 26% planning to adopt it within the next one to two years. Central banks see the potential for Gen AI to handle routine cybersecurity tasks, thereby freeing up resources for other initiatives. Gen AI has proven effective in detecting cyber threats, accelerating response times, and identifying suspicious trends and anomalies, outperforming traditional tools.
领英推荐
At the same time, concerns about the costs of implementation and risks such as social engineering, zero-day attacks, and unauthorized data disclosure remain which needs to be carefully managed. That said, the adoption rate of GenAI by central banks shows that the value from GenAI in supporting cybersecurity efforts is worth the risks.
**********************
5. U.S. Treasury Secretary Janet Yellen Warns Against “Significant Risks” of the Use of AI in Finance
Julia Shapiro at the Hill wrote about Treasury Secretary Janet Yellen issuing a warning about the significant risks that AI could pose to the financial system, highlighting its complexity, lack of transparency, and inadequate risk management frameworks. The potential vulnerabilities include the widespread use of similar data and models, concentration among a few vendors, and reliance on insufficient or faulty data. In response, both the Treasury and the Financial Stability Oversight Council are prioritizing the assessment of AI's impacts on financial stability. Their plans include launching a public request for information on AI in financial services and hosting a roundtable on AI in the insurance sector. SEC Chair Gary Gensler also underscored the risks of a "monoculture" in AI, where dependence on a few large models could lead to a fragile financial system.
For banking leaders, it is imperative to enhance risk management frameworks to address AI complexities and ensure diversity in AI models and data sources to avoid systemic risks. Evaluating and mitigating risks associated with dependence on a limited number of AI and data service vendors is crucial. Additionally, banking leaders should actively participate in industry discussions and stay informed about regulatory developments to ensure compliance and stability in AI adoption.
**********************
Looks like the summer has been busy in the world of GenAI. We were spoilt for choice on what to include in this edition of newsletter! We're as excited as you are about its potential. Stay tuned for upcoming editions packed with insights, industry intel, and practical tools to navigate the exciting future of GenAI in Banking together.
#GenAI #COO #BankingOps #Innovation #ArtificialIntelligence #BankingTechnology #Banking #Transformation
This issue of the newsletter is brought to you by Accenture’s Rajat Dev | Nazat Dowla | Francesco Erriquez | George Heathcote | Sebastian Cadillo | Alfonso Berguido | Anjana Deepu | Ashley Iu
**********************
For additional details on topics above, please reach out to below:
Editors: Rajat Dev, Nazat Dowla, Francesco Erriquez
1. Morgan Stanley Launches AI Assistant 'Debrief' for Wealth Advisors: Ashley Iu
2. Goldman Sachs Deploys Its First Generative AI Tool Across the Firm: Sebastian Cadillo
3. bunq partners with NVIDIA to fight financial fraud with AI: Ashley Iu
4. Central banks turn to generative AI for enhanced cybersecurity: Sebastian Cadillo
5: U.S. Treasury Secretary Janet Yellen Warns Against “Significant Risks” of the Use of AI in Finance: George Heathcote
?
?