AIMultiple ResearchAIMultiple Research

Top 13 Ways to Mitigate the Ethical Risks of RPA in 2024

With the widespread adoption of RPA, it’s important to discuss RPA ethics. Who should be held responsible for its malfunctioning? What goal should it serve? Or how advanced should it get?

These are all questions that warrant definitive answers. And that is what we aim to do in this article. Specifically, we will talk about what RPA ethics is, the values it should uphold, and the 13 different risks that RPA technology poses to society, the economy, and businesses alike.

This article primarily builds upon, and summarizes, arguments made by Dirk Beerbaum, a scholar, in a paper published in 2021, on the ethics of AI and RPA.  

What is RPA ethics?

RPA technology is dependent on human rationality for development and implementation. Therefore, RPA ethics refers to the dilemmas that programmers, users, and adopters of RPA face when dealing with the societal, operational, and economical consequences of the technology.

What are the operational risks of RPA? 

Operational failures are incidents, hiccups, barriers, or mistakes that prevent a business process from running smoothly. For instance, a production plant fire, caused by overheated wires, is an operational failure that interrupts the regular flow of the manufacturing process.

In terms of RPA technology, a sudden malfunctioning can have a negative downstream effect on workflows that depend on it. You can read our article on RPA failures to learn more about the negative effects and best practices towards such scenarios.

Therefore, to mitigate operational risks of RPA, the following conditions should be upheld:

1. Safety

The scripts for RPA and the underlying IA technologies should ideally be safely programmed and securely kept for the duration of the system’s operational lifetime. For instance, data leakage poses a safety issue.

2. Digital footprint

If RPA technology results in operational, administrative, or societal harm, it should be possible to digitally track the process back and determine what caused it and how.

3. Easy-to-understand explanation

When something goes wrong, the programmer must explain it in a way that a human authority who isn’t familiar with the lingo may understand.

4. Human control

There should be human-led-controls applied to the technology, and to the underlying codes and algorithms, wherever there’s a possibility of a systematic breakdown. This makes human intervention possible.   

Risks of value misalignment 

In order to bridge the gap between ethical values and RPA’s alignment with it, the following principles should be upheld:

5. Principal-Agent problem

The responsibility for RPA’s moral implications, its use, and misuse should lie with the designers, programmers, and the stakeholders of the technology.

6. Social and moral values 

RPA and AI should be built from the ground up with respect to social and moral ideals of human dignity, rights, freedoms, cultural diversity, non-discrimination, and unbiasedness. 

AI bias is a common issue and developers should take it into consideration while implementing RPA.

7. Non-subversion

The RPA and AI technology should be built to uphold and strengthen the social and civic processes rather than trying to subvert them. 

One use case of RPA in healthcare, for example, is claims management. RPA should ideally be used as an impartial and rule-based system that evaluates each patient’s claim based on their documents, and the hospital’s policies.

In this way, the outcome is programmed to be as equitable as possible. It would be contrary to such a framework if the technology favored some patients/claims over others.

8. Common Good 

RPA and AI systems should stay true to the promise of democratizing automation for the masses. Therefore, any developments in those fields should be in the service of the common good rather than the benefit of specific individuals or organizations. 

9. Accessible to all

The AI technology has to be accessible to all users without discrimination or apprehension. Open-source RPA, for instance, is an instantiation of such premise, wherein the underlying code for an RPA tool is freely shared on the Internet in order to be leveraged and used by citizens and professional developers alike to create one of their own.  

Risks of negligence 

These risks refer to negligence over not applying proper controls over the development path of the RPA and an unrealistic expectation over the development stage of the technology. 

10. Recursive self-improvement 

RPA and AI systems that leverage ML and other similar technologies for evolution and self-improvement should constantly be watched and monitored to ensure their development are in line with safety, legal, ethical, and other control measures. 

11. Caution

Because, theoretically, there are no “limits” to how advanced and scalable the AI technology could get, assumptions regarding the subject matter should be made cautiously. 

On the end-user side, for instance, the business managers who wish to adopt RPA should be realistic about how expansive automation can get vis-a-vis their day-to-day activities. Overly optimistic automation promises to stakeholders should be avoided.

Market risks

Market risks are posed by agglomeration of, mainly, operational and supply chain risks, because of AI-enabled failures. These risks negatively affect a business’ market position in terms of pricing, competitiveness, outreach, and popularity.

12. Supply chain risk 

Users and the adopters of RPA should be aware of the risks they pose to supply chain continuity. For processes that depend on timely execution and scheduling, a malfunction could lead to distribution delays, manufacturing and/or assembling delays, and market shortages.

Taken to the extreme, these supply chain disruptions would have a downstream effect on sudden price changes and consumer panic. 

The dependency of supply chain management on RPA and other AI-enabled technology should always be mitigated through “exit plans” that would ensure continuity until the technological mishap is resolved. For instance, one exit plan could be the adoption of process mining. 

Process mining can enable logistics firms to detect deviations, delays, and errors occurring in their processes and discover the root causes behind these issues. As a result, the firms can identify the processes that they need to implement RPA or they can solve the problems by simply changing the order of tasks.

13. Operational risk

Operational risk is any plausible circumstance that might disrupt or endanger the operational continuity of a business.

For instance, if the RPA technology mistakenly sends company’s confidential information to a wrong email address, the operational continuity of a business, and reputation, might be at risk.  

By applying human-led controls through different stages of the automation processes, as well as having a transparent process map of the RPA technology’s functionality, businesses can mitigate the operational risks that autonomous technologies might pose.

For more on RPA

To learn more about RPA’s technology and use cases, read:

To gain a more comprehensive overview of RPA, download our whitepaper on the topic:

Get RPA Whitepaper

And if you believe your business would benefit from an RPA solution, we have a data-driven list of RPA software vendors prepared.

Go through them, and we will help you choose the correct tool for your business:

Find the Right Vendors
Access Cem's 2 decades of B2B tech experience as a tech consultant, enterprise leader, startup entrepreneur & industry analyst. Leverage insights informing top Fortune 500 every month.
Cem Dilmegani
Principal Analyst
Follow on

Cem Dilmegani
Principal Analyst

Cem has been the principal analyst at AIMultiple since 2017. AIMultiple informs hundreds of thousands of businesses (as per similarWeb) including 60% of Fortune 500 every month.

Cem's work has been cited by leading global publications including Business Insider, Forbes, Washington Post, global firms like Deloitte, HPE, NGOs like World Economic Forum and supranational organizations like European Commission. You can see more reputable companies and media that referenced AIMultiple.

Throughout his career, Cem served as a tech consultant, tech buyer and tech entrepreneur. He advised businesses on their enterprise software, automation, cloud, AI / ML and other technology related decisions at McKinsey & Company and Altman Solon for more than a decade. He also published a McKinsey report on digitalization.

He led technology strategy and procurement of a telco while reporting to the CEO. He has also led commercial growth of deep tech company Hypatos that reached a 7 digit annual recurring revenue and a 9 digit valuation from 0 within 2 years. Cem's work in Hypatos was covered by leading technology publications like TechCrunch and Business Insider.

Cem regularly speaks at international technology conferences. He graduated from Bogazici University as a computer engineer and holds an MBA from Columbia Business School.

To stay up-to-date on B2B tech & accelerate your enterprise:

Follow on

Next to Read

Comments

Your email address will not be published. All fields are required.

0 Comments