Every generation throughout history believes it has lived in innovative times, and yet, every generation brings its own innovation and change. The reality is that defining what innovation looks like can be quite hard. Steve Jobs described it as “putting a ding in the universe”; Thomas Edison as “finding a better way to do things”; and science fiction writer, Arthur C Clarke as “going beyond the limits of the possible”.
For the asset management industry, innovation has been driven by the proliferation of data; advances in technology, including the widespread adoption of artificial intelligence (AI); and commitments to ambitious sustainability goals – all of which have caused significant disruption to the business, people and investment models of organisations.
Economist and Santa Fe Institute external professor, W. Brian Arthur, maps this digital and data revolution over the last 50 years – from integrated circuits, processors and memory chips in the 1970s/80s; to the connection of digital processes and computers via the internet; to the development of magnetic, gyroscopic, radar and other sensors. The latter is critical as these sensors brought us oceans of data and it is estimated that the asset management industry has nearly tripled its spending on data since 2017. The challenge for our industry today is how to make sense of it all, while providing benefits for its stakeholders, with the use of artificial intelligence playing a leading role.
Benefits can’t be understated
The benefits of artificial intelligence to our industry cannot be understated and we see investors trying to harness its power through the use of natural language processing, image recognition and machine learning.
From processing unstructured ESG data from alternative sources with the aim of assessing company risk; to using AI in private markets to source deals and conduct due diligence on businesses; to improved customisation of products and client experiences. We also see its benefits in improved trade-execution algorithms; searches for new sources of alpha through alternative data and the generation of synthetic data points and scenarios; and reduced costs for data management.
Indeed, around 63 per cent of banks and investment firms surveyed confirmed that they are currently deploying or already using AI, with a further 28% intending to deploy it over the next 1-3 years (Gartner Data and Analytics Transformation Survey, 2022).
However, it is the development of generative AI – where machine learning models are trained to generate new content and data by training on existing data sets – that has caused divisions: the optimists who see the significant opportunities to drive work efficiency, allowing our workforce to do different, higher-value tasks (the National Bureau of Economic Research (NBER) recent working paper, Generative AI at Work, which points to a 14 per cent productivity improvement, with the greatest impact on novice and low-skilled workers); the pessimists who emphasise AI’s potential to propagate misinformation and create widespread disruption to jobs or even existential risk to human life; and those in-between who see lots of opportunities for AI’s use but strongly highlight the need to mitigate and manage its risks.
Social technology generally lags behind the development of physical technology and, as such, we need to be aware of the risks and put in place guardrails, while embracing its benefits.
We also need to not underplay the roles of human intelligence (HI) as a complement to rapid advances in AI use cases. Indeed, the combination of AI + HI will be especially powerful if we are to learn the intrinsic limitations of this technology and adjust our part in this combination.
Empathy, judgement and the ability to inspire
The reality is that AI cannot yet fully replicate human behaviour in all its dimensions. Traits such as creativity, empathy judgement and the ability to inspire others are very much the reserves of humans. We are also reminded that the skills of the future (The investment professional of the future, CFA Institute, 2019) are not just technical, but also include soft skills such as relationship and building social capital; leadership skills such as crisis management and instilling an ethical culture; and T-shaped skills including situational fluency and adaptability. And we also need judgement and inference skills to consider data in its full context where simple causality is not present in a complex system and where trade-offs need to be made between highly objective/valid hard data and softer more subjective data that may be more material.
Data science and analytics have become a vital part of the investment business. But the ultimate test of quality in data and technology will be related to the quality of decision-useful information and the connected insights, judgements, processes and algorithms applied to it.
AI can indeed be a game changer for our industry – it is a systemic opportunity – but only if we are able to mitigate the risks that have and will come from multiple sources. It is the powerful combination of AI + HI that will truly deliver long-term value – enabling us to make better decisions quickly and more consistently, with the human touch.
Marisa Hall is head of the Thinking Ahead Institute at WTW, an innovation network of asset owners and asset managers committed to mobilising capital for a sustainable future.
