Wall Street, Manhattan, New York
Andrey Denisyuk | Moment | Getty pictures
In the feverish breed for the takeover of artificial intelligence, the financial world is at a critical time. The attraction of the all-purpose AI, classified by Tech Giants, is undeniable. However, this approach is a dangerous mirage for financing, an area of ​​complicated regulations and specialized jargon.
It is time for a reality test: finance needs a own AI, not a uniform solution.
The idea that a generalized major language model (LLM) can navigate the complexity of asset management, asset management or insurance seamlessly is generally incorrect. These are domains with their own jargon, private data, specialized workflows and intermediaries, similar to the healthcare system or the law.
A model trained on a wide internet data is to struggle with the precision that is necessary for financial calculations and compliance with official compliance. It will also not indicate the multi -stage process to navigate decision -making trees unless a frame is provided.
Models that were finely coordinated with private, public and user -used data in the real world and were further improved by synthetic or simulated data using basic (and sometimes small) language models, for certain application cases using knowledge graphs and detailed workflow schemas, to enable thinking, the quality of your AI application in finances will soon determine.
Extracting the language from a document is one thing; Thinking and interaction with a specialist in a financial context with its unique methods and schemes is different. This leads to a natural inference: even the hyperscale horizontal players – the Microsofts and Amazons – and the application developers – the Salesforces and Palantirs in the world – need specialized collaborators in finance. The Generalist -Ki platforms lack powerful, but the necessary domain expertise is missing.
Specialized AI
The depth that is required in areas such as asset management and asset management is simply too granular. These managers inevitably have to work with industry specialists who have intimate knowledge of workflows, regulations and user experiences in finance.
The era of Bulldozing -llms by domains is over. The future lies in the verticalization, where, in cooperation with experts who understand the subtleties of the financial world, a special AI creates. This vertical complex financial services is also large enough to justify these partnerships. At the same time, traditional financial services companies have to give up the hybris to use these general platforms to build up in the house. The initial impulse to build and own the technology based on domain expertise is understandable – sometimes because the providers are not mature or stable enough in an emerging industry. However, this is an expensive and often unsuccessful undertaking.
The AI ​​landscape develops at burning speed. What is innovative today is out of date tomorrow. This requires repeated new reviews, a culture of thinking for a clean leaf and an organizational design that rewards the speed. Financial institutions risk being included in an eternal cycle of development and maintenance and steering the resources from their core business. If an application for the industry is common, there is the possibility that a fintech that focuses on this application is faster on a better product, scales, learns and maintains than an internal team.
A relevant parallel is the early development of CRM systems: The attempt to build up their own internal solution in the early 2000s when specialized partners have appeared has now clearly proven to be short-sighted. In some cases in which the company is large – e.g. For example, a JPMorgan or Morgan Stanley – and the resources have to build the establishment of internal teams that are tackling applications for you. It can also make sense whether the platform is used to generate and improve its intellectual property of Kern. Suppose you can move quickly.
As a result, the intelligent step for the Generalist Technology players and the incumbent financial services companies is to accept partnerships. Companies should concentrate on what makes them unique – their special sauce – and it lets up -to -see fintechs cope with the additional heavy lifting.
In summary, the financial world must recognize that its AI needs are different. Special solutions are needed. More strategic partnerships are needed between Tech giants and financial experts. Traditional companies are needed to resist an isolationist go-it-all approach. The operations are high. Generalist technology company and specialized financial operators: Caution.
Dr. Vinay Nair is the founder and CEO of Tifin, a FinTech asset platform that uses AI and Investment Intelligence to serve the asset and asset management industry. Previously, Nair was the founder 55ip, which was taken over by JPmorgan Chase.