Q&A with Étienne Labrie-Dion, a data scientist from Canadian software development leader, GSoft
AI is permeating business at a much faster rate than expected. Forbes and the 2018 Deloitte Human Capital Trends Report describe adoption rates in this area as dramatic. Almost half of Deloitte’s respondents in their study report that “their organization is deeply involved in automation projects with 24% using AI and robotics to perform routine tasks.”
If you’re not living and breathing AI in your work every day, it can be daunting to think that a quarter of businesses have already replaced some human labour with robots.
So, how can you leverage AI within your organization today? How will it impact your job? As an HR professional and hiring manager, what do you need to know?
These questions were top of mind for us, so we recently caught up with an expert who could help us better understand this trend – one that is clearly transforming our work environment:
Meet Étienne Labrie-Dion, a data scientist who works for GSoft, one of Canada’s leading software development companies that describes itself as “An experimental lab on the future of work.” This organization develops workplace-efficiency and employee-engagement software, and was recognized in the 2017 PROFIT 500 as one of Montreal’s Fastest-Growing Companies.
(And if you want to peek into their very cool office space – complete with a mini-ramp and in-house barista, check out this feature on NotableLife.com.)
Here is our Q&A with Étienne:
Q: When it comes to understanding AI’s applications in the business world, what are we getting right, and where are we misunderstanding things?
A: Many people get caught in the hype of it all. For narrow applications such as translations, image recognition, consumer behaviour prediction, today’s AI applications are quite good. In fact, they are a lot better than they were 10 years ago. A lot of business executives don’t realize just how useful some narrow applications can be. But if you’re not expecting the right things out of it, you’ll be disappointed.
When it comes to pulling business insights using AI for example, this technology is still in its infancy.
Net-net: AI is good at finding solutions to a given problem. But people tend to expect too much performance in one sense, and not in another. It is critical to understand scope and right now the opportunity lies mostly in predicting value or user behaviour.
Q: Can you give us an example of the benefits in predicting value or user behaviour?
A: In some applications, we can look at how many users we anticipate coming to us on a given day. Then, we can look at how to better serve those customers. Our main data analytics goal could be to figure out when it is best to speak to our users so we don’t bombard them with messages at the wrong time. If we share info too early, perhaps it won’t be of interest to them. But if we can connect with them at the right time, to give them the information they are seeking, they will be interested, and we can deliver value.
Q: So, in contrast, what are some of the barriers we face today, using AI?
A: Let’s look at it from the other side first: There are two essential things that you need to make AI work well today. The first is: the quality and quantity of data. Both have to be good. The second is: the right expertise in knowing how to use that data.
Once you have these two elements, it’s important to plan the project well and manage expectations. Because technology is speeding things up so drastically nowadays, the business world sometimes expects the development of AI to also work at that pace. But these are not the kinds of things that could, for example, be completed in one week.
Measurement is also an essential part of this conversation. Sometimes, the impact of AI can be felt immediately, while at other times, the benefits will surface further down the road. Planning accordingly is very important.
Q: What surprises you the most about AI?
A: Definitely how fast it is growing. There are so many new technologies and I’m in awe of how fast companies are starting to apply them. It’s hard to keep up. I’m surprised at how much you can do with basic AI and machine learning today. There is so much data and processing power. It’s amazing to see how you can take huge data sets and apply algorithms to make predictions.
Q: Will AI replace humans at work?
A: No. AI helps humans perform better. It doesn’t replace humans. AI will help make people better at their jobs. Here’s an example: spreadsheets didn’t replace accountants, right? While spreadsheets can add up the numbers, the implications of those numbers, and the smart business recommendations based on those numbers must still be determined by humans.
The main goal for AI is to do some automated tasks so humans have more time to do other things. It’s about streamlining the process. AI is replacing some of the redundant work that humans do today.
Q: We saw an article in the Globe and Mail recently that reported on how Monster is utilizing AI to streamline the job search process for jobseekers. It used AI in building a résumé assessment tool to help candidates see how strong their applications were for the job postings to which they were responding.
In this context, what advice would you give to HR and hiring managers who may want to explore utilizing AI within their organizations?
A: Of course, it’s hard to keep up with everything. I would say the first step is to look into your company to see who could develop that expertise in AI. Consider people you may have in technical data roles, and examine what areas there are for improvement. What kind of tasks can be automated within your organization, and what kind of data do you have that can help support such a move? Such questions can be answered in partnership with your IT personnel who deal with your organization’s data analysis and technology. Together, you can figure out what needs to be done.
I would also recommend staying informed on what other businesses in your industry are doing with regards to AI, and where others are experiencing successes. You may be thinking, for example: “We want to use AI for x.” Knowing where you are in relation to your industry can help you make better informed decisions, preparing for any implications, and setting up your expectations for success.
Étienne Labrie-Dion, was a research scientist in neuroscience for 10 years. He then traded his lab reports and microscopes for data sets and algorithms. Today, in his own words, Étienne focuses on “improving GSoft’s products through data analysis and user behaviour prediction.”
We sincerely thank Étienne and GSoft for giving us their time and expertise for this article.