The stock market has long been dominated by algorithmic trading, but these clever tools are almost always going to trade based on how they’ve been coded to do so. What about completely independent AI? AI research and advisory consultants Emerj believe that machine learning (ML) and AI will give traders an advantage not seen before by allowing them to access unstructured data, and make trades based on their own learned datasets. Questions remain, however, over how accurate AI traders can be, and the ethics - and legality - over how fair the technology really is in the modern market.
There’s a lot of automation in stock trading already. Stock market research platforms bring together data and third-party information from a range of sources and then present it to traders in such a way that they can make an informed decision on trades. Where AI diverges from this process is in its total automation of decisions. An AI can be given a set of instructions and then will make its own decisions. Technology in this field is seeing huge growth - CNBC reports that Microsoft AI platform C3.ai, already deployed by banks to automate the loan application process, has the potential to be set up by traders and then left to run, making trades as they see fit.
AI trading is already underway, according to the Tech Times. Several platforms have already fully automated the stock trading process, and to great effect, though notably on low risk trades - few traders are willing to put their money behind high-risk, high-value trades when the technology is in such early days. As AI is put into the stock market it will learn, however. This is the benefit of ML and AI based approaches to work - the system is always learning, and this is part of the guarantee of smart learning systems that makes them so attractive.
Ethical and legal questions
AI poses a lot of concerns in regards to ethical and legal oversight. One journal, published by Springer Open Access, asserts that AI must continue to be developed along the same regulatory objectives that all stock market developments are typically designed - that is, in the interests of market fairness. However, there are questions over whether an AI can truly be fair. It allows traders to trade at a far faster rate than a human would be able to. It takes in a greater amount of information and is constantly learning in a way that is somewhat inaccessible to traders - the sheer level of data processing and knowledge can be hard to challenge. Getting AI right, and ensuring that those with the best technology can’t simply take over the market, is a huge challenge for developers.
In a world where there’s a level playing field, however, AI is a great tool for traders. It can help both retail and institutional investors to find more value in the market, and reduce the human burden of trading. Developers need to find that ethical balance, however, for it to be fair.