As artificial intelligence (AI) and its various subsets — such as machine learning (ML) — become more complex and developed, their presence within the financial realm has increased drastically.
According to a 2022 report by Nvidia, over 75% of companies operating in the financial sector apply machine or deep learning to optimize their internal operations.
Moreover, the study notes that 91% of financial firms are now driving critical business outcomes with investments in AI, with many of the surveyed companies stating that the burgeoning technology has helped them yield more accurate prediction models.
More than 30% of respondents claimed that using AI and ML has increased their annual revenue by more than 10%, while over 25% of the surveyees stated that AI has helped them reduce their annual working costs by more than 10%.
Data processing redefined
Despite its relative nascency, AI is poised to bring significant changes to the financial sector, with its potential being similar to that of computer-driven trading models introduced by Wall Street traders in the 1980s.
Jeroen Van Lange, founder and analyst for YouTube channel The Blockchain Today, told Cointelegraph:
“AI is being used to develop machine-learning trading models, detect transactional irregularities, and even analyze complex blockchain data with an exceptionally high level of accuracy.”
“Moreover, ML-based tools are being used to analyze risk from borrowers to assess their creditworthiness using a broad range of data sources like their social media activity and online behavior,” he added.
Van Lange highlighted that since most cryptocurrency exchanges provide real-time data linked to their order books, ML algorithms can study these comprehensive data sets to predict short-term price movements.
Similarly, in the case of derivative exchange data, these models can sort out and process information like open interest, funding rates and taker buy/sell ratios much more rapidly than humans, thus allowing traders to make better investment choices.
“This is something we have not yet seen before, that programs are thinking for themselves and improving their decision-making capabilities on the fly,” Van Lange said.
A new standard for data security
The introduction of AI and ML has allowed blockchain systems to enhance their security capabilities.
AI-enabled platforms can provide users with real-time threat feeds while allowing them to gain actionable insights into various scams, rug pulls and threats.
Earlier this year, Forta’s monitoring systems detected the attack on the Euler protocol minutes before the hack, which saw $197 million stolen.
While Forta was able to provide some advanced notice to Euler, the protocol’s team was unable to respond in time.
Similarly, Forta’s Attack Detector module was also successful in flagging the $3.3 million hack of decentralized finance platform SushiSwap back in April, as well as the flash loan attack on Yearn.finance, leading to a loss in excess of $11 million around the same time.
As a result of its threat detection capabilities, Forta has accrued the financial backing of several prominent industry players, including Coinbase Ventures, a16z, Blue Yard and Blockchain Capital, among others.
Solving the issue of liquidity fragmentation
Even as the crypto market matures and grows, it still faces several issues around illiquidity, especially when compared with traditional finance.
Ahmed Ismail, CEO and co-founder of FluidAI, an AI-based crypto aggregation platform, told Cointelegraph that digital asset liquidity is currently siloed with a few major players, making the market extremely inefficient. He added:
“Even the most stable cryptocurrencies, such as BTC and ETH, are fragile. The crypto market needs high-quality liquidity aggregators so that when volatile conditions are witnessed, market participants can access funds quickly and at the best possible price to maintain some sort of equilibrium.”
When asked how AI can help address these problems, he noted that aggregators — including FluidAI — use the technology to predict digital asset order book prices in real time, thus providing deeper liquidity for relevant trading pairs. “FluidAI uses an AI-bolstered Smart Order Router and Matching Engine to connect to major centralized and decentralized exchanges and enhance liquidity reserves.”
Moreover, Ismail said that his platform uses customized algorithms like volume-weighted average price, time-weighted average price, arrival price and volume participation to minimize adverse market impacts and prevent information leakages during the execution of large orders.
In today’s globalized economy, sentiment analysis continues to play a larger role across various industries, including crypto.
With AI, businesses can now understand customer sentiments in real time, allowing them to curate and personalize their marketing efforts.
A recent study from researchers at University Canada West notes that AI-powered sentiment analysis tools can comprehend the tone of a statement instead of merely recognizing certain words within a piece of annotated text as being positive or negative.
Companies can also use these tools as part of broader business strategies to help them outperform competitors, attract and retain consumers, perform live research to assess client interest in certain themes and understand market conditions.
Lastly, these tools are scalable and suitable for companies handling vast amounts of feedback data. By analyzing this feedback information, it is possible to address areas for improvement, respond to issues promptly, and make informed decisions to enhance customer satisfaction.
What lies ahead for the future of finance?
Despite the nascency of AI and blockchain technology, Ismail believes these innovations have the potential to complement each other and reshape the way we perceive global finance:
“Distributed ledger technology offers immutable data storage capabilities with enhanced transparency and traceability. AI, on the other hand, can process enormous amounts of blockchain data to provide intelligent insights and accurate prediction models. By combining the two, market participants can make informed decisions to maintain healthy market metrics.”
Ismail further claimed that every major financial institution should look closely at utilizing technologies such as natural language processing, deep learning, reinforcement learning, generative models and edge computing to stay ahead of their competition.
A fairly similar point of view is shared by Cerullo, who believes that using AI within today’s existing financial structures can help investors rake in better returns, at least for certain select trades. That said, he did concede that AI is not some magic wand that can automatically enhance productivity.
“It can, however, serve as a valuable assistant.”
Collect this article as an NFT to preserve this moment in history and show your support for independent journalism in the crypto space.