Partner of Eventus Advisory Group, LLC., helping companies meet their CFO, Finance and Accounting needs with fractional teams.
One of my favorite quotes, which has been attributed (and misattributed) to lots of people, is “Artificial intelligence is no match for human stupidity.” That statement captures where we are with AI: There is so much good that AI can do—if human stupidity doesn’t mess it up. AI hype promises unparalleled efficiency, speed and accuracy for the finance and accounting professional, but the reality (both near-term and long-term) is more complex.
AI is touted as the panacea for the challenges faced in many areas of business. Automation of tedious tasks, enhanced data analysis and improved decision making are undeniably attractive use cases for AI. While AI has immense potential, however, let’s look at three points where we need to separate hype from reality to better understand AI’s opportunity in the accounting and finance department.
Hype: ChatGPT 4.0 has an incredible number of parameters.
Reality: “Garbage in, garbage out” still applies.
Much hype about AI is the new models’ data crunching and ability to spot patterns across myriad data points; however, today’s AI models all rely on existing data sets for the “training” required to deliver appropriate outputs for requested tasks. When data sets contain errors or inaccuracies, the AI’s outputs will be flawed. The media regularly reports on AI that provides false information, or even made-up data, to user queries.
Most organizations still deal with imperfect, missing or incorrect data. For AI to be effective, the data has to be clean and high-quality for training the model and then maintained at high quality, as the AI will continue to train its output based on any new data inputs.
Hype: The pace of change will usher in a new era of productivity.
Reality: Today’s investments may quickly become obsolete.
It is true that AI technology is evolving at an incredible pace. New AI models, AI-dedicated chips and new AI companies seem to be launched daily. Deciding where and how to invest in AI for your business is a major obstacle to adoption. Investing in a single platform with all the licensing, training and implementation costs may not make sense for many businesses, as any one AI platform may quickly become outdated.
As the sector evolves, winners and losers will shake out. In the meantime, stay agile and avoid relying on a specific AI solution.
Hype: AI is a paradigm-shifting technology.
Reality: AI is another tool that has great use cases.
AI is a tool, an amazing one to be certain, but it is still just a tool. While the output from an AI model can provide a user with valuable insights, it cannot replace the need to collect clean data and run efficient processes.
AI, as it is today, is a complementary technology tool that might belong in your tech stack, as AI can replace, enhance or automate specific tasks. To make the most of any new technology, however, you need to understand how it fits into your business, so let’s take a closer look at each of these use cases.
Most accounting departments have some basic data entry or reconciliation tasks where manual processes could be replaced with AI; however, making sure the data is correct will still involve human oversight after the AI-run process is complete. Therefore, in AI’s current state, looking for complete replacement is the exception rather than the rule. This leads to where current AI might be very useful: enhancing current processes.
AI’s immediate value is the fact that it can augment already existing human capabilities. If an AI can handle data aggregation and formatting, then accounting professionals have additional time to focus on value-added tasks, such as analyzing the data and making strategic decisions. The AI could also assist in evaluating additional scenarios, giving the user the opportunity to explore more strategic options than was possible without the help of an AI assistant.
The most exciting potential use of AI perhaps lies in its ability to make even the most complex tasks more manageable. For example, an AI can be trained to analyze your company’s financial forecasts and identify the most sensitive assumptions.
Imagine uploading your business plan and a multiyear forecast into an AI program and having it find that a change in the cost of inventory for one specific part of your product has an oversized impact on your profitability. The AI could further tell you when to order the part or show the order of magnitude for different assumptions.
Or more generally, an AI reviews your general ledger and finance data and identifies two line items, which you have control over, have the greatest impact on your financial results. That’s extremely powerful. With such detailed information, you can better prioritize. While all these scenarios are theoretically possible, there is still a long way to go before most accounting departments have access to such a tool.
Wither The Finance Professional?
The introduction of AI in finance has raised questions about its impact on accounting and finance jobs. History shows that new technologies do not always diminish an industry’s relevance. Rather, they improve efficiency and allow professionals to focus on more critical tasks. Finance professionals will continue to play a pivotal business role in any AI-driven landscape. They will serve to not only ensure that AI tools are used correctly but also interpret results, provide context and align AI-generated insights with the broader business objectives.
At present, AI’s role in finance and accounting is more one of promise and potential. The hype surrounding AI must be met with a sober assessment of its capabilities and limitations. Finance professionals will remain indispensable, using AI as a tool to enhance their capabilities and achieve more significant insights. As AI technology continues to advance, the industry must adapt, evolve and stay vigilant to ensure that AI is leveraged effectively and ethically.