Managing assets, the GenAI way

After almost two decades of robust growth, the asset management industry stands at a crossroads as costs rise, fees decline, and investor expectations evolve.

No longer confined to automating routine tasks, Generative AI (GenAI) is emerging as a strategic enabler in portfolio optimization, risk management and client service.

GenAI can help asset managers cut costs by enhancing operational efficiency, diversify revenue streams by driving innovation, and offer personalised service by fostering a client-centric approach.

By leveraging its ability to process vast datasets, uncover hidden patterns, and generate actionable insights, GenAI is redefining how asset managers approach decision-making and client engagement.

As the industry evolves, asset managers who harness the power of GenAI will be better equipped to navigate complexity, seize opportunities and deliver exceptional value to their clients.

GenAI versus discriminative AI

Artificial intelligence (AI) is a field of computing that has the ability to learn, solve problems and perform tasks that typically require human intelligence.

It can be broadly categorised into two types: discriminative and generative.

For years, AI has been part of the asset management industry primarily in the form of discriminative AI, utilised for predictive analytics and forecasting based on historical data and market trends.

However, discriminative AI works only on supervised machine language (requires humans for manual labelling) and learns by understanding rules and identifying boundaries to classify or predict outcomes.

In contrast, GenAI absorbs and processes vast amounts of data to create new content, learning rapidly and answering complex questions.

GenAI is set to disrupt and revolutionise the asset management industry, serving as a powerful tool for analysing financial data, predicting market trends and simulating hypothetical scenarios.

That said, it is not intended to replace human expertise but rather empower it. 

In the asset management industry, the capabilities of GenAI are vast and diverse. GenAI offers capabilities that extend far beyond traditional automation and analytics. Its ability to learn, adapt, and generate human-like outputs has transformed how asset managers approach complex tasks, enabling greater efficiency, accuracy, and scalability. Below are nuanced perspectives on the key capabilities of GenAI

• Search and analysis: Provides enhanced search capabilities, intelligent analysis of media and textual content from multiple sources, and democratisation of access to information
• Information interpretation: Reformulates unstructured data into structured formats by leveraging techniques such as Optical Character Recognition (OCR) and Natural Language Processing (NLP)
• Content creation: Generates customised content in multiple mediums; analyses existing data to learn patterns in order to create new content aligned with specific user objectives or preferences
• Summarisation and classification: Condenses lengthy documents into concise summaries, highlighting key insights, and classifies data with enhanced accuracy and efficiency
• Personalisation: Analyses user preferences, behaviour and historical data to create personalised recommendations, content and provide human-like answers to inquires or prompts
• Voice recognition: Transcribes speech into text and vice versa, enhancing accessibility and facilitating communication 

The adoption of GenAI is not merely a technological trend - it is a strategic imperative driven by the need to navigate increasingly complex markets, extract actionable insights from vast datasets, and deliver superior investor outcomes. As asset managers face mounting pressures to enhance efficiency, manage risks, and uncover-alpha generating opportunities, GenAI emerges as a transformative solution. Below are they drivers fueling the adoption of GenAI:

• Processing and analysis of data at scale: Extracts valuable insights and patterns from large volumes of data - news flow, market trends, company filings, market commentaries etc. at quick and timely manner
• Navigating complexity of financial markets: Simulates various scenarios by not just analysing historical data but create synthetic data if needed, and optimises portfolios based on risk-return profile
• Real-time decision making: Enables swift response to market signals and execution of trades with precision by processing and analysing data in real-time.
• Analysis of diverse data sources: Identifies investment opportunities and potential risks by analysing unstructured data such as social media sentiment, satellite imagery and geolocation data. Provides distinctive insights into investment opportunities and potential risks
• Risk management and regulatory compliance: Models risk scenarios, stress-tests portfolios, predicts market downturns, and provides accurate and auditable records of investment decisions
• Alpha-generating opportunities: Uncovers hidden patterns and correlations in data. According to Asian Bureau of Finance and Economic Research (ABFER) research paper¹, Hedge funds using Gen AI generated 1.8%-3.5% higher annualized abnormal returns vs non-adopters
• Cost reduction and management of increased workloads: Automates routine tasks, such as data analysis and research, and accommodates growing volumes of data and portfolios
• Personalisation of investment recommendations: Takes into account individual investor profiles and preferences and offer hyper-personalized investment advice and portfolio recommendations

The adoption of GenAI is driven by its ability to address critical challenges in asset management while unlocking new opportunities for growth and innovation. As firms continue to embrace GenAI, they are not just optimising operations - they are redefining the future of asset management.

GenAI is making asset management more data-driven, agile and client-centric. Its versatility allows firms to address complex challenges while unlocking new opportunities for value creation. Its ability to combine advanced analytics, automation, and personalisation makes it a cornerstone of innovation in asset management and beyond.

• Predictive analytics and limited-partner (LP) reporting: Provides personalised investor communications, predictive analytics and real-time market insights, helps preparing customized pitch or RFP response documents
• Portfolio management and optimisation: Enables creation of portfolios that are finely attuned to prevailing market conditions, and resilience of portfolios in complex and uncertain environments
• Scenario analysis: Simulates various market scenarios, assessing portfolio resilience and potential risks, to enable proactive decisions and strategy adjustment
• Asset allocation optimisation: Analyses real-time market data and makes dynamic adjustments to portfolio allocations to maximise portfolio returns while mitigating risk 
• Bespoke investment strategies: Develops tailored investment strategies aligned with clients’ financial objectives, risk tolerance and ethical preferences, as well as evolving market conditions
• Operational efficiency: Automates routine tasks, data analysis and reporting, enhancing operational efficiency by minimising errors and freeing up resources for high-value tasks
• Trading: Substantiation of trading wheel rules by incorporating market information in real-time. Identifies subtle patterns and correlations in how counterparties adjust their pricing or volumes, allowing for more informed and profitable trade decisions.
• LP reporting: Automates creation of tailored financial reports and client communication, ensuring consistency, precision and personalisation in conveying investment strategies and performance
• Market research: Analyses financial data and market news to identify emerging trends, assess market sentiment and predict market movements
• ESG-aligned investing: Analyses data to identify companies with strong ESG practices and associated risks, balancing financial returns, regulatory compliance and sustainability in portfolios

Outdated infrastructure, poor-quality data and AI models’ high computational demands are hindering asset managers’ ability to integrate GenAI effectively and produce reliable results at scale. Moreover, clients are hesitant to trust AI with their financial futures. In fact, GenAI's personalisation capabilities can build trust by delivering tailored experiences. 

The Aug 2024 CFA Institute Survey² noted, a need for standards and upskilling to ensure responsible and ethical use of AI and GenAI. 85% of the survey participants noted a need for industrywide standards and ethical guidelines for AI/GenAI, with 82% stating the lack of such standards hinders faster adoption.

To overcome scepticism, it is essential to: 

• Clearly communicate the role of AI as a complementary tool 
• Ensure responsible use of GenAI in terms of data quality and transparency 
• Address training bias, identified as the primary barrier to responsible AI adoption 
• Implement robust governance frameworks and regular audits to ensure trustworthy AI outputs

Asset managers have framework in place to oversee the quality of gen AI outputs, though the extent of that oversight varies widely

GenAI has introduced a transformative approach to optimising portfolios, managing risk and delivering client service in the asset management industry. By integrating GenAI in their operational frameworks, asset managers are not merely adopting a new technology - they are embracing a strategic leap that positions them at the forefront of innovation.  

As GenAI continues to advance, it is crucial for asset managers to strike a balance between harnessing its benefits and ensuring transparency, accountability, and regulatory compliance. 

By adopting a strategic and responsible approach to GenAI, asset managers can gain a competitive edge as well as shape the future of the industry.

¹ Generative AI and Asset Management
²Aug 2024 CFA Institute Survey