The hustle, hype, frenzy, fear, loathing, and even panic surrounding AI today is daunting. However, the practical uses of AI, such as helping predict inflation and curbing inflation problems, are important if companies use the right AI tools.
For example, the emergence of OpenAI’s GPT-3 and ChatGPT has overturned conventional wisdom about the nature and role of AI. Even AI scientists did not predict how large-scale language model (LLM) applications would outperform previous AI applications on various metrics. As a result, people are turning to AI for help with almost everything.
In the context of the financial sector and dealing with inflation, having the right AI tools is one of the key considerations. Another thing is to use them the right way.
How AI can and cannot help with financial forecasting
No matter how great ChatGPT or any other LLM may seem, it’s important to remember a few key facts. First, they are not good at everything. You may be able to pass the basic conversational Turing test or get a passing grade on the bar exam, but your rationale is poor, your structure is poor, and your writing is poorly written. It may produce inappropriate results.
It also tends to: Hallucination, or just a false answer. When questioned, most LLMs stick to their answers and may even cite fabricated evidence in the form of fabricated statistics or citations. Without significant intervention in training and oversight to make AI products stick to actual facts and their implications, this tendency to mix lies and truths is a major obstacle to the use of his LLM by companies.
Thankfully, AI consists of many areas of development besides LLM.
The financial services industry has been at the forefront of AI applications for many years. Financial services and fintech companies have primarily focused on machine learning (ML) for analytics and forecasting. An ML developer feeds them a plethora of relevant data points and feeds back answers to questions to get algorithms to learn important parameters in the data stream. The goal is to build a model that not only gives you the right answers, but also adapts to changing data streams—a model that allows you to navigate your financial data and find insights.
One area of ML that is particularly important to the financial services industry is deep learning. Deep learning applications use neural network models that combine multiple layers of analysis. Financial services companies can use the vast amount of stock market movements to train deep learning models and develop algorithms that, for example, help predict how the market will move next. The insurance industry is already training models based on vast amounts of demographic, policyholder and weather data to predict where crop, fire and flood premiums will need to rise in response to climate change. I’m here.
Offsetting the impact of inflation with AI
AI cannot completely avoid inflation. Business leaders experience the effects of inflation when suppliers raise prices. However, ML applications can help offset the effects of inflation by improving corporate efficiency.
Consider a retailer. With access to information about orders, warehouse inventory, and shelf inventory on the one hand, and sales, customer, and demographic information on the other, ML systems can learn to optimize. You can determine how much ordering, shipping, and warehousing inventory is required to meet your company’s sales targets in a given time while saving costs. The same concept of optimization with AI can be applied to manufacturers.
Retailers and fast food companies can use ML to find not only material and transportation efficiencies, but also location efficiencies. Specifically, you can train an ML model to decide where to move new locations, or whether to grow, shrink, close, or move existing locations. Service providers can also use such models to optimize staff numbers, skill sets and placement.
Machine learning can help with the effects of inflation, but quantum computing and quantum AI may play a bigger role in solving such problems in the future. For some types of problems, quantum AI can help companies find solutions faster. Quantum AI allows machines not only to find better solutions to some optimization challenges, but also to find the best possible solutions. Given that generative AI is rapidly taking the world by storm, this future may be closer than you think.
