Artificial Intelligence (AI): Its Impact to Accountants, and to Electric Cooperatives

By Peggy Boldissar

What is AI? Merrian-Webster (n.d.) defines AI as: 1 : a branch of computer science dealing with the simulation of intelligent behavior in computers

2 : the capability of a machine to imitate intelligent human behavior

MTI College (2018) describes what AI technologies do. AI allows machines (bots) to learn from experience, interpret information, make adjustments and apply what they “know” to perform humanlike tasks.

● Machine learning gives computers the ability to recognize and apply patterns to develop algorithms that they can then fine-tune based on feedback; an example is the way Netflix can offer you suggestions of what to watch based on your viewing history.

● Through deep learning, computers develop the ability to identify relationships and associations.

● When machines “understand” information, they can actually “think” of the implications of that data and analyze it with machine reasoning.

● Computers use natural language processing based on their understanding of human speech.

● They “recognize” people, activities and

objects, and “view” images with computer vision; an example is the way the iPhone X recognizes the face of its user.

Johnston (2018) writes that: AI has matured from technology buried in computer science labs using complex coding techniques to more common algorithms and supporting technologies used as part of the design strategy of new generation products. Many of the developers have known of AI techniques for years but did not have a practical way to apply the algorithms because the compute overhead was too high, the sample of data was too small and the number of techniques that needed to be applied made the code too complex. With centralized computing in SaaS applications and cloud data centers, AI has become much more practical and accurate.

Johnston further describes how AI approaches work. They use:

● Cybernetics and brain stimulation – connection to neurology .

● Traditional symbolic AI – John Haugeland named these approaches to AI “good old fashioned AI” or “GOFAI” exploring the possibility that human intelligence could be reduced to symbol manipulation.

● Cognitive simulation – Economist Herbert

4 Winter 2018 | The Cooperative Accountant

Artificial Intelligence (AI)

Simon and Allen Newell studied human problem-solving skills from psychological experiments resulting in the Soar architecture in the 1980’s.

● Logic-based – John McCarthy in his laboratory at Stanford (SAIL) used formal logic and led to the Prolog language and the science of logic programming.

● Anti-logic or scruffy – Marvin Minsky and Seymour Papert found that solving difficult problems in vision and natural language processing required ad-hoc solutions.

● Knowledge-based – led to the development in the 1970’s of expert systems, introduced by Edward Feigenbaum of Stanford.

● Sub-symbolic – when traditional symbolic AI stalled in the 1980’s was unable to solve problems in perception, robotics, learning and pattern recognition, researchers tried to not encode knowledge.

● Embodied intelligence – Researchers of robotics, such as Rodney Brooks, reintroduced the use of control theory and embodied mind.

● Computational intelligence – neural networks and “connectionism” was revived by David Rumelhart leading to soft computing approaches including fuzzy systems, evolutionary computation and statistical tools.

● Statistical methods – sophisticated mathematical tools to solve specific subproblems that are truly scientific, in the sense that their results are both measurable and verifiable.

● Intelligent agent – a system that perceives its environment and takes actions which maximize its chances of success.

The strengths of AI technologies lie in their ability to process information effectively for large data volumes and complex and changing patterns. And, they are able to do this with consistency, without tiredness, boredom, or biases exhibited by human counterparts. Parsons (2018) notes that AI is not only software that can draw conclusions from large quantities of data and adjust its activities based on those conclusions, but it’s also a system that can learn quickly in real time and be applied to an entire organization (para. 1).

AI does have some limitations. Not every problem or decision is suitable for a machine learning approach. For example, there needs to

be a degree of repeatability about the problem or question so that an automated AI program can generalize its learning and apply it to other cases. For unique or novel questions, the output may be far less useful. The outputs of machine learning models are predictions or suggestions based on mathematical calculations, and not all problems can be resolved in this way. Other considerations may need to be factored into decisions, such as ethical questions, or the problem may require deeper root cause analysis. Some decisions are best made within some context, and AI cannot always deal well with context. Another instance where AI may not be the best application is when not all of the right data is available for program analysis. Success with AI comes when data is sufficient and of the right quality.

How is AI impacting accountants and the accounting profession? The world of accounting is just the latest in a series of industries being affected by the rapid increase in the use of AI. The Institute of Chartered Accountants in England and Wales (ICAEW) is a member of Chartered Accountants Worldwide (CAW), which brings together 11 chartered accountancy bodies, representing over 1.6m members and students globally. They have described the human decision-making process as follows:

Human Decision-Making: Humans make decisions in two different ways

Institution Much of our thought process is instinctive and unconscious, taking place very quickly and with little effort. This type of thinking is rooted in recognizing patterns based on what has happened before, and is often described as intuitive.

Reasoning We also use logic and reason in order to answer questions and make decisions. This conscious process uses our knowledge and typically takes over when intuition has not produced a satisfactory answer. This process takes time and effort.

ICAEW goes on to further explain: Accountants, as expert decision makers, use both ways of thinking – they apply their knowledge to specific situations to make reasoned decisions,

Winter 2018 | The Cooperative Accountant 5

Artificial Intelligence (AI)

but also make quick intuitive decisions based on extensive experience in their field. Our intuitive thinking is particularly powerful, reflecting quick learning and high levels of flexibility. It provides the foundation for our language, vision, sensing, understanding of the everyday world and ways of interacting with others. However, it is not perfect. It is subject to many biases and inconsistencies, explored in detail by psychologists such as Daniel Kahneman.

● Availability bias – more recent or common examples tend to come to mind, which can skew our decision-making processes.

● Confirmation bias – we tend to see only things that are consistent with our existing views.

● Anchoring – we are strongly influenced by prior suggestions.

It could be that we are at a time in history where how we store and access financial information coupled with maturing technology capabilities will accelerate the digital transformation of our accounting profession. As machines continue to advance technologically, taking on more and more repetitive and timeconsuming tasks, human accounting resources are being freed up to do higher level (and hopefully more lucrative) work for their employers or clients.

Su (2018) writes in a recent Forbes article that “in our upcoming research report on the future of accounting, we expect that by 2020, accounting tasks - but also tax, payroll, audits, banking… - will be fully automated using AIbased technologies, which will disrupt the accounting industry in a way it never was for the last 500 years, bringing both huge opportunities and serious challenges” (para. 5). Su notes that the accuracy of the machine learning algorithms used in most of today’s solutions still needs to significantly improve in efficiency to avoid accounting errors and really fulfill their promise of automation. But, most agree that it is just a matter of time until those improvements are made and available to the masses at affordable costs.

Some of the opportunities that may be associated with this digital transformation include:

● Accountants should be able to handle more work or clients, and deliver added value as they can provide actionable insight rather than

just be a number cruncher.

● Automating accounting processes will also help improve operations and reduce costs.

● The cost of AI technologies is becoming increasingly more affordable to small and midsize companies, not just larger organizations. This will continue to be the case as time progresses.

Some of the areas where AI is already making a difference in improving operations and reducing costs are illustrated by the following examples:

● Accounts payable/receivable processing: There are already AI-powered invoice management systems available that can make invoice processing much more streamlined thanks to digital workflows that are implemented. They can learn the accounting codes that are appropriate for each invoice.

● Supplier onboarding: Machines can vet new suppliers by checking their credit scores or tax information and set them up in the system without human involvement and even query portals to get all the necessary information.

● Procurement: The procurement and purchasing processes for most organizations are filled with paperwork and use different systems and files that are not compatible with one another. As machines through APIs are able to be integrated and the unstructured data is processed, the procurement system will eventually become paperless. Robots are ideally suited to tracking price changes among a number of suppliers.

● Audits: Digitalization of the audit process will help increase its security by allowing a digital trail of when and by whom each file was accessed. Instead of searching file cabinets for the documentation that is required during an audit, auditors will be able to leverage the digital files. A more digital audit improves the efficiency and accuracy of audits and makes an audit of 100% of a company’s financial transactions possible instead of just a sample.

● Monthly/quarterly close process: The faster one can get the numbers, the more time the organization has to think strategically about what to do with the numbers. Machines can post data from a number of sources, consolidate and reconcile it. Not only will

6 Winter 2018 | The Cooperative Accountant

Artificial Intelligence (AI)

the monthly/quarterly close process be speedier, it will also be more accurate thanks to the support of machines in the process. Additionally, technology for generating and processing natural language can turn financial tables into written documents, give structure to large volumes of unstructured financial data, and pore through long documents and contracts to extract needed information.

● Expense management: Reviewing and approving expenses to ensure they are compliant with an organization’s policies is time-consuming for an accounting team. Machines can read receipts, audit expenses and alert humans when a possible infraction has occurred.

● AI chatbots: Chatbots are used to efficiently solve common questions or queries. For example, AI chatbots from customers can include the latest account balances, when certain bills are due, the status on accounts and more.

Charpentier (2018) discussed the latest advancements using AI in Accounts Payable (AP).

Before AI, accounting teams manually created and processed invoices, purchase orders, or delivery orders on paper documents. Those documents were then manually entered in computer systems, coded, and finally transmitted to the managers for approval and payment. Today, thanks to AI, there are no more manual processes! The AP workflow process is automated by software that analyzes, recognizes, directs, and exports data into a company’s ERP/financial system. Before automating the AP workflow, suppliers had little to no insight into payment timing details; now, they have full access to this information in real time.

The advancements in AP automation have been powered by the following technology deployments:

● Algorithms have become more and more reliable, flexible, and adaptable, permitting solutions to automatically manage documents with variable structure, such as invoices. As a result, data is automatically recognized in an exhaustive and reliable way, with no prior configuration.

● Cloud solutions are available to millions of users, which results in constant technological enhancements. This contrasts to older onpremise solutions where usage was limited.

● The self-learning — machine learning — capabilities of cloud-based software solutions are constantly improving. These solutions essentially “learn” from their mistakes and do not make them again once humans correct them. AI-driven AP programs are increasingly able to:

● Identify and interact with suppliers;

● Automatically intake, code, process and route invoices, using optical recognition technology; and,

● Denote payment deadlines, approval workflows, and the approvers. Everatt (2018) notes that Gartner research predicts that by 2022, 80 percent of smartphones shipped will have on-device AI capabilities. This change is set to impact accountants and very quickly the industry will see the implementation of optical character recognition (OCR) technology. We’ll also see more intelligence from smartphones when performing functions. For instance, he notes, if an employee is driving, the smartphone might prompt the user to see if they would like to make a claim for their mileage, or auto populate an expense claim by scanning a receipt.

For auditors, who perform audits of financial records, they can access AI tools with natural language processing capabilities to interpret thousands of bits of information. The technology can extract key terms and compile and analyze that information to perform risk assessments or other functions that are vital to creating and executing audit plans. Additionally, these tools can perform a variety of analyses, designed by humans, and then provide lists of exceptions for the auditor to evaluate. Machine learning comes into play as the auditor confirms the exception or invalidates that exception and the machine learns to “look” at the auditor’s conclusions and try to identify additional data points about the positives or negatives to apply to additional exceptions it identifies. In this way it learns to better identify exceptions. The auditor role, for both internal and external auditors, will switch from performance of the procedures to design of the procedures, interpretation of the results, and

Winter 2018 | The Cooperative Accountant 7

Artificial Intelligence (AI)

monitoring the effectiveness of the interpretation.

While AI technologies are fairly wide-spread in use, there has been limited use of some AI technologies within the accounting industry. These include things such as: 1. using machine learning to code accounting entries and improve on the accuracy of rules-based approaches, enabling greater automation of processes;

2. improving fraud detection through more sophisticated, machine learning models of ‘normal’ activities and better prediction of fraudulent activities;

3. using machine learning-based predictive models to forecast revenues; and

4. improving access to, and analysis of, unstructured data, such as contracts and emails, through deep learning models.

Expectations are for technology improvements to be deployed in these areas in the future.

AI’s impacts to electric cooperatives AI and the energy industry are not mutually exclusive. In fact, they are becoming increasingly interconnected as computing power, data collection, and storage capabilities scale exponentially on an annual basis. The advances in both industries offer opportunities to meld new technologies to improve the performance in both arenas. AI technologies are still in their early stages of development and implementation, and the future expectations are that they will be poised to revolutionize the way energy is produced, transmitted, and consumed. Wolfe (2018) notes that an added benefit of AI is its ability to limit the energy industry’s environmental impacts at a time when “demand is steadily growing, our energy production portfolio is diversifying, and we are witnessing the ramifications of fossil fuel consumption on biodiversity, air quality, and quality of life” (para. 3).

The mission of the utility industry, including electric cooperatives, is to produce and transmit power under a mandate of low-cost, reliable energy as a public good. The utility grids in the US were constructed decades ago, with many current systems averaging 30 to 40 years old. The current challenge in the industry is the rise of distributed generation, where private sector users are generating and using their own energy from renewable sources (i.e. wind or solar). In

balancing the past grid investments with the deployments of present energy technologies, AI may help to bridge the gap in providing information regarding supply and demand requirements on power systems. The older grids were not built with the new emerging energy technologies in mind at the time. The energy industry is currently in a transformation period working to adapt and incorporate the new energy technologies into existing energy portfolios. With the influx of smart grid technologies, as mandated by the US Department of Energy, the energy industry has been working over the last decade towards a more automated power delivery network that monitors and controls consumption by customers. The capture of automated meter data in smart meters will be a prime example of where AI technologies can be merged with large amounts of consumption data collected by these smart meters in order to better manage load requirements. If the smart grid is able to use fossil fuels in the most efficient way possible through increased incorporation of renewable resources as those technologies advance in sophistication and capability, the entire system may be able to reduce its carbon footprint. On the supply side, AI could allow the transition to an energy portfolio with increased renewable resource production and help to minimize disruptions that come variability in sunlight and wind intensity.

One of the major concerns with the smart grid is the increased use and reliance on the internet and computer processing power. The computer industry has become a large contributor of greenhouse gas emissions in recent years as companies shifted to machinerun operations, and the use of the internet has increased dramatically. With the increase in the capture of large amounts of data from customers’ smart meters, additional computer machines and computing power will be needed. As a result, the impact of energy consumption on the environment from further greenhouse gas emissions will likely continue to increase. Players in the AI energy grid industry will need to address this problem. Industry leaders have been aware of this challenge and have been taking steps to address this. Interestingly, the three leading greenhouse gas emitters in this industry are computer makers, data centers, and telecoms. One example of how this issue is being addressed is that computer makers are investing in new

8 Winter 2018 | The Cooperative Accountant

Artificial Intelligence (AI)

hard drives, screens, and fuel cells; data centers are monitoring temperatures, pooling resources and researching cloud computing; and telecoms are looking into network optimization packages, solar-powered base stations, and fiber optics (Wolfe, 2018, para. 15).

AI is continuing to make inroads in the energy industry. Some of the use cases for AI include:

● Consumer energy efficiency - AI can track household behaviors and feed smart thermostats. This could result in devices that learn the habits of the household occupants and adjust temperatures automatically to accommodate them.

● Storage - Smart storage applications can make pulling energy from sources more intuitive. This could result in devices that track behavioral data like peak demand and lows to figure out how fast battery storage should charge, or when to dispatch energy, and how much energy it should store.

● Automating Grids - Grids can respond to fluctuations in demand, peaks and lows, and meet them when it’s most necessary. Additionally, utilities could control many processes and equipment remotely. This means, though, that the network they operate on could be vulnerable to attacks.

Summary It will be the effective pairing of people and machines that provide for the most valuable deployment of an organization’s resources, allowing each to contribute in the areas where

they are best suited. Machines can handle a high volume of information rapidly, and identify and quantify patterns in data which could speed up decisionmaking or problemsolving. It will fall on people to be the ultimate decisionmakers and problemsolvers, as no machine has yet been able to replace certain value added activities performed by people. Present-day AI technologies have yet to mirror a human’s ability to handle decision making based upon context.

Accountants will continue to notice this transition from task-driven activities to one of empowerment in which systems driven by AI are in charge of the data entry, data verifications, referrals, and fraud detection. With this continuing technological transition, accountants’ time will be increasingly freed up to produce “real added value with time for analysis, strategy, creative thinking, and decision-making” (Charpentier, 2018, para. 11). Professional accountants do much more than keep track of receipts and provide basic reports, they act as consultants who advise on tax planning, discuss operations, and review employer and client goals and more. Additionally, the rapid pace of change in client industries and the expansion of complicated regulations means that human controller services will be necessary to ensure that compliance requirements are met and financial controls are sound

AI is largely being used to digest and analyze large volumes of data at speeds well beyond what any person or team of people could do. Experts generally believe that some accounting jobs could be lost as machines continue to take over more and more automated tasks, but most believe that the net impact to jobs available for accountants will be negligible. Some even believe that accountants may be in more demand. The biggest impact will be the shifting of work duties from less sophisticated to more

Winter 2018 | The Cooperative Accountant 9

Artificial Intelligence (AI)

sophisticated or complex work. Everatt (2018) contends that “Accountants are now able to use the time saved through AI-based technologies to add insight to the information provided and increase value, allowing them to move towards more legislative and compliance-orientated tasks” (para. 11). Although AI technologies are not new, and the pace of change is fast, widespread adoption in business and accounting is still in its early stages. While accountants have been using technology for many years to improve what they do and deliver more value to their employers and clients, emerging AI technologies present an opportunity to reimagine and radically improve the quality of business decisions.

The recommendation to company management is to form a strategy for AI technology deployment in their organization’s strategic plans because AI isn’t coming, it’s already here. Parsons points out that “the difference between leading a financial services team toward new, technological opportunities – or obsolescence – is in the strategic focus of leadership” (para. 2). Some of the areas of focus recommended for management might include imagining how AI can help better serve the organization’s mission, exploiting powerful technologies, thinking radically, and remaining adaptable to change as it becomes warranted.

The recommendation for accountants and auditors is to gain database and IT skills by taking on special projects or work assignments, attending seminars, and completing classes or self-directed learning to learn about latest technologies. It is important for them to stay upto-date on the latest accounting trends, emerging technologies and industry news. Accountants need to learn not to rely on AI tools blindly, but instead be able to critically analyze and spot important information produced from automated tools. The AI evolution will also be reflected in the skills required of accountants. Some roles may require deep knowledge of machine learning techniques, but other roles may just need a more superficial knowledge of machine learning to be able to have informed conversations with experts and other parts of their organization. Either way, critical thinking and communication skills are likely to become increasingly more important. The role of the accountant could change into a managerial position where technology will increase the accuracy and efficiency of work freeing up

accountants to take on a partnership role with other areas of the business.

Accounting regulators and standard setters also need to build their understanding of the application of AI technologies and be comfortable with any associated risks. The recommendation is that standard setters and regulators take a proactive approach to understanding AI and how it is being deployed by accountants. For example, standard setters in audit will want to examine where auditors are using these techniques to gain evidence, and understand how reliable the techniques are. If organizations and audit firms increasingly rely on

AI models in their operations, more thinking will be required about how these regulators and standard setters gain comfort in their correct operation. Regulators can also actively encourage and even push adoption where it is aligned to their requirements.

References CBInsights. (March 8, 2018). 5 Ways The Energy Industry Is Using Artificial Intelligence. Retrieved November 7, 2018 from https://www.cbinsights. com/research/artificial-intelligence-energyindustry/

Charpentier, L. (July 25, 2018). Voices From artificial intelligence to accounting intelligence. Retrieved November 7, 2018 from https://www. accountingtoday.com/opinion/from-artificialintelligence-to-accounting-intelligence

Everatt, N. (July 20, 2018). Will AI help or hinder the future of accountancy? Retrieved November 7, 2018 from https://www. accountancyage.com/2018/07/20/will-ai-help-orhinder-the-future-of-accountancy/

Institute of Chartered Accountants in England and Wales (ICAEW), IT Faculty. (2017). Artificial intelligence and the future of accountancy. Retrieved November 7, 2018 from https://www. icaew.com/-/media/corporate/files/technical/ information-technology/technology/artificialintelligence-report.ashx?la=en

Johnston, R. (May 22, 2018). What AI Means for the Accounting Profession: Part II. Retrieved November 7, 2018 from https://www. cpapracticeadvisor.com/article/12407031/what-aimeans-for-the-accounting-profession-part-ii

Marr, B. (June 1, 2018). The Digital Transformation Of Accounting And Finance - Artificial Intelligence, Robots And Chatbots.

10 Winter 2018 | The Cooperative Accountant

Artificial Intelligence (AI)

Retrieved November 7, 2018 from https://www. forbes.com/sites/bernardmarr/2018/06/01/ the-digital-transformation-of-accountingand-finance-artificial-intelligence-robots-andchatbots/#2ba1024f4ad8

Merrian-Webster Dictionary. (n.d.). Definition of artificial intelligence. Retrieved November 7, 2018 from https://www.merriam-webster.com/ dictionary/artificial%20intelligence

MTI College. (August 15, 2018). How Artificial Intelligence (AI) Impacts Accounting. Retrieved November 7, 2018 from https://mticollege.edu/ artificial-intelligence-accounting/

Najjar, D. (July 15, 2018). Is Artificial Intelligence (AI) the Future of Accounting? Retrieved November 7, 2018 from https://www. thebalancesmb.com/is-artificial-intelligence-thefuture-of-accounting-4083182

Parsons, S. (July 31, 2018). How AI and Automation Technology Can Help Accountants. Retrieved November 7, 2018 from https://www. business.com/articles/ai-and-accounting/

Su, J.B. (January 22, 2018). Why Artificial Intelligence Is The Future Of Accounting: Study. Retrieved November 7, 2018 from https://www. forbes.com/sites/jeanbaptiste/2018/01/22/whyartificial-intelligence-is-the-future-of-accountingstudy/#571c5191337b

Wolfe, F. (August 28, 2018). How Artificial Intelligence Will Revolutionize the Energy Industry. Retrieved November 7, 2018 from http://sitn.hms.harvard.edu/flash/2017/artificialintelligence-will-revolutionize-energy-industry/

Articles of Interest Harris, J. (August 9, 2018). Is Artificial Intelligence Set to Replace Accountants in the Future? Retrieved November 7, 2018 from https://finance. toolbox.com/articles/is-artificial-intelligence-setto-replace-accountants-in-the-future

KDP Certified Public Accountants, LLP. (June 1, 2018). The Effects of AI on Accounting. Retrieved November 7, 2018 from https://www. kdpllp.com/the-effects-of-ai-on-accounting/

Kubicki, F. (August 14, 2018). AI in Accounting: How Artificial Intelligence & Machine Learning Technology Has Changed the Industry. Retrieved November 7, 2018 from https://www. intellichief.com/ai-in-accounting/

Mohamandi, T. (September 28, 2018). What accountants can learn from AI. Retrieved November 7, 2018 from https://accountingweekly.

com/what-can-accountants-learn-from-ai/

Mooney, G. (May 17, 2018). 10 Ways Utility Companies Can Use Artificial Intelligence And Machine Learning. Retrieved November 7, 2018 from https://www.digitalistmag.com/ digital-economy/2018/05/17/10-ways-utilitycompanies-can-use-artificial-intelligence-machinelearning-06167501

Ovaska-Few. (October 10, 2017). How artificial intelligence is changing accounting. Retrieved November 7, 2018 from https://www. journalofaccountancy.com/newsletters/2017/oct/ artificial-intelligence-changing-accounting.html

Shimamoto, D.C. (April 8, 2018). Why Accountants Must Embrace Machine Learning. Retrieved November 7, 2018 from https://www. ifac.org/global-knowledge-gateway/technology/ discussion/why-accountants-must-embracemachine-learning

Warawa, J. (November 1, 2017). Here’s Why Accountants (Yes, YOU!) Should Be Driving AI Innovation. Retrieved November 7, 2018 from https://www.cpapracticeadvisor.com/ news/12378218/heres-why-accountants-yes-youshould-be-driving-ai-innovation

UTILITY COOPERATIVE FORUMGeneral Editor Peggy Boldissar LCEC (Lee County Electric Cooperative, Inc.) Manager, Financial Accounting PO Box 3455 North Fort Myers, FL 33918-3455 Phone (239) 656-2117 Fax (239) 656-2256 peggy.boldissar@lcec.net

Winter 2018 | The Cooperative Accountant 11