May 23, 2021

How is Finance Winning the Automation Race?

Automation
RPA
ML
Finance
Faisal Iftikhar, VP Global Hea...
4 min
Faisal Iftikhar, VP Global Head of Automation at Ciklum, share his thoughts on finance lending itself seamlessly to automation

How is Finance Winning the Automation Race?

Faisal Iftikhar, VP Global Head of Automation, Ciklum

 

As automation becomes more prevalent across sectors and industries, the largest uptake area has been across finance processes. There are several reasons why finance lends itself so seamlessly to automation, and other sectors and functions can learn a lot if they pay attention.

 

Automation is changing the way finance functions operate, with technology now enabling finance professionals to push the boundaries of efficiency and effectiveness. The future of finance is here, and leading organisations are adopting these new ways of working to give them a competitive edge.

 

Why Automation?

Simple question, yet still a mystery for some. 

 

Understanding the capability of automation (RPA, AI, ML, Analytics, etc.) can help when businesses are strategising. In a growing number of organisations, automation is playing an important role in driving business objectives around process efficiency and effectiveness.

 

Over the past 5 years, there has been a huge drive for automation and digitisation across all industries and sectors. Studies show that almost all (95%) large organisations will embrace some form of automation by 2022.

 

Recent surveys showed that 77% of senior executives are re-evaluating their automation agenda and the speed of adoption due to the current climate. 

 

Automation and multi-faceted benefits:

Image removed.Image removed.


 

Why is Automation so pertinent for finance functions?

Automation targets processes that are highly transactional, mundane and repetitive.  Finance functions have always been challenged by the business to provide more value through business partnering, and this is exactly the role that finance wants to play in an organisation.  

 

However, more often than not, finance functions spend too much of their time performing routine based transactional processing and therefore do not have the capacity to perform the value-added tasks for the business.

 

Automation can address this issue by taking away the transactional processes, enabling finance personnel to focus on the value-added tasks.  

 

There are a number of processes across finance that are ripe for automation. Some can be automated end-to-end, driving substantial process efficiency, whereas others can be partially automated, augmenting robotic/virtual workers with finance staff.

 

For example, invoice processing can be automated end-to-end using a combination of OCR and RPA to process the invoices, extract the data, and post the information into ledgers. An example of partial automation is for performing financial analysis - the report generation and creation can be automated and then handed over to a finance professional to perform the value-added insights on the reports.

 

Image removed.


 

Automation is here to stay

Finance is usually one of the first functions organisations think about when it comes to automation, and this has been the case for most successfully scaled automation programmes.  

 

Image removed.



 

These are the top three reasons why automation is so suitable within finance functions:

 

  1. Excellent return on investment

 

Not only are there financial benefits associated with automation, but reducing errors, improving compliance and enabling resources to focus on more value-added tasks results in a significant business case.

 

For finance shared service centres, automating the transactional processes allow more work to be pushed into the centres to drive more process and operating model optimisation.

 

  1. Processes

 

Due to the nature of processes, there are various automation opportunities across the purchase-to-pay, record-to-report and order-to-cash functions. Therefore, a prioritised backlog of automation opportunities can be created to build a substantial benefits case.

 

With so much success across finance functions, there are multiple use cases that can be leveraged to help with the identification of opportunities. Finance functions are usually quite similar regardless of industry, which provides opportunities to learn from others more seamlessly. The seasonality of month-end, quarter-end, and year-end processes also makes a strong case for automation.

 

Finance processes also allow intelligent automation to be a fast follower of RPA. Robotic Process Automation can be utilised for transactional processes, OCR can be leveraged for document ingestion and virtual agents/chatbots for helpdesk functions.

 

  1. Finance personnel

 

Finance function personnel typically have a continuous improvement mindset, and therefore, implementing and adopting automation works well in these functions. Historically, VBA macros have been developed in-house within the functions to drive process optimisation, and this culture helps when introducing tools like RPA.

 

Finance personnel have taken up roles within the automation centre of excellence due to their skillset. This has often worked well when building out capability and offering this as a service to the business. It’s no surprise that so many Automation CoEs started in Finance before going enterprise-wide.

 

Right now, banks, insurance companies and utilities are the industry leaders in embracing the benefits of automation. Within organisations, the finance function has a great opportunity for automation. Any business that asks for advice on where they should start their automation journey, I say look in your finance function. It’s the most natural one, increasing your ROI and providing a learning curve for other functions within your company. 



 

Share article

Jun 17, 2021

Facebook Develops AI to Crackdown on Deepfakes

Facebook
MSU
AI
Deepfakes
3 min
Social media giant, Facebook, has developed artificial intelligence that can supposedly identify and reverse-engineer deepfake images

In light of the large tidal wave of increasingly believable deepfake images and videos that have been hitting the feeds of every major social media and news outlet in recent years, global organisations have started to consider the risk factor behind them. While the majority of deepfakes are created purely for amusement, their increasing sophistication is leading to a very simple question: What happens when a deepfake is produced not for amusement, but for malicious intent on a grander scale? 

 

Yesterday, Facebook revealed that it was also concerned by that very question and that it had decided to take a stand against deepfakes. In partnership with Michigan State University, the social media giant presented “a research method of detecting and attributing deepfakes that relies on reverse engineering from a single AI-generated image to the generative model used to produce it.” 

 

The promise is that Facebook’s method will facilitate deepfake detection and tracing in real-world settings, where the deepfake image itself is often the only information detectors have to work with. 

Why Reverse Engineering? 

Right now, researchers identify deepfakes through two primary methods: detection, which distinguishes between real and deepfake images, and image attribution, which identifies whether the image was generated using one of the AI’s training models. But generative photo techniques have advanced in scale and sophistication over the past few years, and the old strategies are no longer sufficient. 

 

First, there are only so many images presented in AI training. If the deepfake was generated by an unknown, alternative model, even artificial intelligence won’t be able to spot it—at least, until now. Reverse engineering, common practice in machine learning (ML), can uncover unique patterns left by the generating model, regardless of whether it was included in the AI’s training set. This helps discover coordinated deepfake attacks or other instances in which multiple deepfakes come from the same source. 

 

How It Works 

Before we could use deep learning to generate images, criminals and other ill-intentioned actors had a limited amount of options. Cameras only had so many tools at their disposal, and most researchers could easily identify certain makes and models. But deep learning has ushered in an age of endless options, and as a result, it’s grown increasingly difficult to identify deepfakes.

 

To counteract this, Facebook ran deepfakes through a fingerprint estimation network (FEN) to estimate some of their details. Fingerprints are essentially patterns left on an image due to manufacturing imperfections, and they help identify where the image came from. By evaluating the fingerprint magnitude, repetition frequency, and symmetrical frequency, Facebook then applied those constraints to predict the model’s hyperparameters. 

 

What are hyperparameters? If you imagine a generative model as a car, hyperparameters are similar to the engine components: certain properties that distinguish your fancy automobile from others on the market. ‘Our reverse engineering technique is somewhat like recognising [the engine] components of a car based on how it sounds’, Facebook explained, ‘even if this is a new car we’ve never heard of before’. 

 

What Did They Find? 

‘On standard benchmarks, we get state-of-the-art results’, said Facebook research lead Tal Hassner. Facebook added that the fingerprint estimation network (FEN) method can be used for not only model parsing, but detection and image attribution. While this research is the first of its kind, making it difficult to assess the results, the future looks promising. 


Facebook’s AI will introduce model parsing for real-world applications, increasing our understanding of deepfake detection. As cybersecurity attacks proliferate, and generative AI falls into the hands of those who would do us harm, this method could help the ‘good guys’ stay one step ahead. As Hassner explained: ‘This is a cat-and-mouse game, and it continues to be a cat-and-mouse game’.

Share article