artificial intelligence and business analysis

I was first introduced to artificial intelligence in 1983 as I watched the movie War Games. I was riveted as I watched a young man, David (played by Matthew Broderick), hack into a military computer to play games. The computer, Joshua, prefers playing chess but agrees to play Global Thermonuclear War. Winning the game involves escalating the threat of war and launching nuclear missiles. David believes he is playing a game. However, Joshua doesn’t understand the difference between real life and the game. The computer was using the actual systems of the US in playing the game and intended to start a nuclear war!

This was terrifying to watch. A machine was taking control of our national defense while the humans watched! Finally, the developer who created Joshua intervenes and talks Joshua into simulating all the scenarios in the game. Joshua responds to him, “A strange game. The only winning move is not to play. How about a nice game of chess?” The world was saved, but I was left afraid for the future!

That was 1983. Systems today are much more evolved in their learning. We have subtly been conditioned to accept AI as part of daily life; it is changing how we react to and interact with technology. If you have shopped with Amazon, asked Siri a question, or told Alexa to play your favorite song, then you have interacted with AI. In some cases, we are now dependent on it!

Artificial intelligence relies on machine-based learning driven by data inputs to simulate human-like tasks. Instead of writing code, progressive algorithms are trained to adapt to our likes, dislikes, and behaviors (for example: Amazon product suggestions and self-driving cars). This isn’t intelligence in the sense that computers are smart and now understand the meaning of life; it is intelligence in the sense that they are improving their decision-making ability to predict actions based on large data sets.

Where does this leave business analysis?

One of the primary functions of business analysis is to perform the work necessary to make a good decision. So, what would happen if we found ourselves in a “War Games” scenario? In other words, what would be left if the AI machines took over and started performing the analysis and subsequent decision making that business analysts perform today?

I have good news… while AI won’t go away, the need for analysts will not be replaced! A BA skill set is primed to work alongside AI. Analysts will be necessary to help improve the algorithms that will improve the quality of AI decisions.

I have three predictions on how analysis will add value as AI becomes more ingrained in our future:

1. Data Analysis

Artificial intelligence is based on algorithms, which are based on data. Data analysis is essential to AI. Questions must be asked and answered in order to provide accurate data for the algorithm to learn from. Not only must the right data be provided, it needs to be correct data – data without flaws or bias.

Garbage data in creates garbage decisions out. Organizations reliant on big data need to be aware of the risk and ensure their data is correct. As we work on solutions going forward, ensuring that we have sufficient data governance and safeguards in place to prevent bad data from entering our environments will be increasingly more important for accurate AI. Excellent elicitation and communication skills will continue to be necessary to provide the correct data to fuel the algorithms that guide good decisions.

Unfortunately, reliance on data creates bias. In her Ted Talk, The Era of Blind Faith in Data Must End, Diana O’Neil describes algorithms as opinions embedded in code. In many cases, the opinions behind the algorithms are flawed, and often people utilizing AI don’t understand the code behind the tool. The opinions are the status quo, the way we have always done things… not necessarily how things should be done.

If we don’t want AI to make decisions based on the way we have always done things, analysis of the algorithms’ learning and conclusions must be constantly assessed. Imagine how unintelligent a machine would be if it is engineering and reengineering itself based on bad or biased data. Feedback is essential to successful AI.

Analysts should always understand data, and in the future, they should be even more focused on understanding the right data to capture, store, and use. It’s the same as business rules being buried in code that are impossible to untangle. This is a problem with the big data engines. Hence, I see a lot of reverse engineering to remove bad data on the horizon. Identifying what the bad data is will also be a critical skill. Develop your data analysis skills in our Detailing Business Data Requirements class. 

2. Human Interactions

AI can’t replace the complexity of an interpersonal relationship. Artificial intelligence is smart but it will not eliminate human intervention in its assessments and decisions. Qualitative analysis is still crucial: asking the “Why” questions. Human beings are intensely complex, and predicting how they will behave in a given situation is difficult. The human hunch or gut instinct can’t be programmed. The decisions we make are influenced by our mood, the moods of others, hunger, and even temperature. As a machine, these human elements aren’t considered when determining AI’s logic.

While AI will enable more effective business, we will continue to work with people. A machine will not become your new CEO; they will not be the ones in charge. Additionally, uncovering the real problems that need to be solved (beyond what an algorithm indicates) will require human interaction. Customer-focused product development and human-centered design will continue to drive business decisions. An analyst skilled in facilitation, elicitation, and understanding people and how to work with them won’t be replaced.

TIP: Don’t forget to hone your communication skills for virtual communication. As the global world becomes more connected, virtual interaction will become more common. The ability to effectively elicit information with a virtual team will be a necessity. Learn from our virtual facilitation experience: 8 Virtual Facilitation Lessons Learned

3. Change Agents

.The other element essential in this journey with AI is transitioning organizations successfully through change. As some functions and processes are automated with AI, there will be a wake of people left behind. What will they be focusing on? What new opportunities await them?

As change agents, analysts must be prepared for human-centered change in the wake of future technology. Analysts have the empathy skills needed to see the change through another person’s view and are skilled in working with people to understanding how change impacts them. We must be able to learn quickly and then turn around and teach what we have learned as part of this change.

Part of our role as an analyst will be seeing how humans augment business with new products, processes, and systems. We are positioned to support organizations to simplify the complexity in order to help others understand the impacts and accept change. Organizations that do not focus attention and resources on people will be unable to compete in the future.

Yes, They Can Co-Exist!

There is a place and a purpose for AI and analysts to work alongside and support each other. The human element will continue to influence systems in order to gain a better machine-learning experience with better data and improve the quality of AI decisions.

Be excited for the future that is ahead; it’s bright, and with great analysis, it will be brighter. I know someone I am looking forward to having in my home is Rosie from the Jetsons. I could use her around; that would leave me more time for great analysis!

– Heather


More on Co-Existing with AI

Detailing Business Data Requirements Course/a>

8 Virtual Facilitation Lessons Learned

Agile BA Checklist


Pin It on Pinterest

Share This