AI is the future of finance, but data comes first

By Or Lenchner, CEO at Bright Data

Artificial Intelligence is a necessity for financial services, especially in the current economic circumstances. For example, an Insider Intelligence report estimates that North American banks have the potential to save $70 billion by 2025 by Automating tasks with AI. Banks have been using AI-based tools to assist clients and speed up the onboarding process, while others have been using it to perform better risk assessment or prevent fraudulent activities. Therefore, there’s no doubt that in 2023, financial institutions will continue to implement AI tools to optimise workflows, boost productivity and efficiency and save money where possible.

Publicly available web data as the foundation

However, AI systems are only as good as the information they are fed. They are still capable of creating bias or making ill-informed decisions, which highlights the importance data plays in their success. In fact, 66% of companies claim poor-quality data impairs their ability to deploy and adopt AI effectively, and that poor-quality data is the main obstacle for businesses to create high-quality functioning AI tools, a Refinitiv study found.

To create trustworthy and reliable AI models we must ensure that the data used is extensive, diverse, frequently updated, and tailored to its specific function. DeepMind researchers concluded that to maximise AI models’ performance, they should be trained on larger datasets. However, high volumes of data doesn’t guarantee better quality, the datasets can still be outdated or biased. Especially in the finance sector, it is imperative for AI models to have access to current data. For example, a tool that is used to calculate risk and capital allocation without having the latest, or most accurate market information, can be damaging.

To train emerging AI models, financial services must have access to the world’s largest up-to-date public database in the history of mankind, also known as the internet. Public web data is vital for AI models, as it provides them with diverse sets of frequently updated information and examples. OpenAI’s ChatGPT’s success, for instance, derives from being fed a large public dataset of text scraped from online websites, blogs, articles and forums.

Strict compliance and alternative data

New advancements in web platforms and technology have simplified the process of accessing and gathering public web data, any company, big or small can get their hands on the required qualified data to train their machines without paying huge sums of money or having a full-blown data operation in place.

Alternatively, companies can purchase pre-collected datasets on demand, which hold a tremendous amount of public web data and can be ideal for training AI models. These datasets can be acquired once and be refreshed at periodic intervals as a cost-effective and speedy way for companies to get their hands on massive amounts of frequently updated web data from multiple different sources.

Financial institutions are realising the power of leveraging alternative data by having a public web data provider. Banks now have a 360-degree view of customer interactions, their transaction history, and social media activity to make more informed decisions and ensure they’re meeting their customers’ needs.

With that power comes great responsibility. The US Securities and Exchange Commission (SEC) constantly warns about the use of predictive data analytics, which means that financial institutions must ensure their web data provider is flexible and comprehensive as well as validated and highly compliant. Without proper compliance and ethics guidelines, financial institutions risk incurring massive fines and the same applies to the data that feeds AI tools. Public web data providers who are well versed in compliance, especially in the strict financial domain, can help to eliminate that risk.

Whether AI will automate time-consuming tasks, improve the speed and accuracy of work, or predict potential problems, every business can benefit from it. How well will those tools perform? That’s down to the quality of data they are trained on – the more extensive and reliable the data, the better the performance rate and, consequently, the more valuable the results.