Artificial Intelligence in Accounts Payable
by Jamie Radford
AI in AP
Artificial intelligence (AI) has been a buzz word for some time now, but it’s becoming more and more interesting in the accounts payable realm. It is literally changing the future of AP with some exceptional results such as improved efficiency, smoother processing of accounts, quicker processing times and additional data insights to name a few. And more productive and engaged staff members is just one of the trends we are beginning to see in companies increasingly implementing AI within their AP departments.
AI within AP has some truly amazing applications… Large companies who have creditors globally are able to simply use AI to interpret hundreds or even thousands of contracts, picking out the crucial clauses, analysing them and carrying out risk assessments. What an unbelievable time saver! The vast quantities of data where AI can be implemented to analyse is changing the fact of AP as well as other industries – exciting times for all of us within AP.
Another excellent application of AI within AP is that it can assist in defending against fraudulent activity and attacks. Why? Cyber criminals are becoming progressively sophisticated in their activities and big companies and financial institutions are feeling the impact more and more. Just look at what happened with British Airways not so long ago. AI is able to analyse vast quantities of data from breaches around the world and pick up on even the smallest similarities between previous cases of fraud that would normally not be found by the human eye, to avert them in the future. AI in this context is able to raise a red flag where it may be missed with only human analysis, no matter how in-depth it may be. This more efficient fraud protection can offer a sense of security for managers and company owners, knowing the likelihood of an attack is more rare than may have been a few years ago.
Automation is already offering companies improved protection within their AP departments and maximising efficiency in the combating of fraud, and AI is probably not going to replace these automation techniques completely – but it will certainly complement them, making them even more powerful. As a side note, many people think that automation and artificial intelligence are one in the same, however this is not correct. A succinct definition on each that I found explains it quite well: “Automation is software that follows pre-programmed rules. Artificial intelligence is designed to simulate human thinking.” (Source: medium.com).
SOME TASKS SUITED TO AI
- Clearing invoice payments
- Advanced invoice automation
- Auditing expense reports
- Risk assessment
- Month-end closing
- Data analytics
- Bank reconciliations
- Learning patterns and suggesting general ledger codes, etc.
Besides the obvious benefits that come to mind from reading the above, a great benefit of this digital transformation within AP is the elimination of paper – or at least a department that is almost paperless. Automation and artificial intelligence work together to seamlessly feed results into AP workflows or ERP systems.
One of the benefits which I found particularly interesting was the need for less IT support on a day to day basis. How often are our IT colleagues called for even the smallest questions or perceived errors? With AI in AP, systems can actually be trained and the AP staff member can merely act as a facilitator to help the “intelligence” be learned by the applications. Once the systems have learned the necessary, stretched IT resources will be called upon less and less.
A further benefit is that invoices, once analysed by the AI system can go directly to managers for approval. Graphical dashboards within AI systems will also be able to allow managers to see various vital metrics in real-time. Causes for exceptions and other issues can be discovered quickly and can be resolved almost immediately, ultimately leading to an AP department that is a really slick operation.
In a nutshell, AI is set to enhance responsibilities within AP departments and the financial function as a whole.