Impact of AI on jobs today

Last week I shared a 2013 article from Mother Jones about the fears of job automation. This week I want to share an article from LinkedIn about how this is now our reality.

This September 2018 post leans on LinkedIn data and the World Economic Forum Future of Jobs report to show some trends of the entrance of AI across industries.

An interesting highlight from the article is a comparison of the occupations with the highest and lowest growth over the past five years.

Image via LinkedIn

Among the highest growing jobs are Human Resources Specialist and Recruiter, which this article suggests are inherently difficult to automate and therefore less likely to see the impacts of AI.

These roles require an understanding of human behaviors and preferences—a skill set which fundamentally can’t be automated.

Igor Perisic, “How artificial intelligence is already impacting today’s jobs,” LinkedIn

I would agree that the top jobs on this list do require an understanding of human behavior that may insulate them in some ways. However, the growth of these jobs also increases the pressure to ensure they are as efficient as possible, and that is the benefit of applying AI in these fields.

Advertisements

How AI can revolutionize HR

My #AIFebruary project is focused not just on learning about artificial intelligence, but also its applications in my field, HR and recruiting. With that in mind, I enjoyed these thoughts from the Forbes Human Resources Council from July. Members were asked what a future with AI might look like in our field. Some top answers:

Enhance efficiency. Stacey Browning, President of Paycor, advocates for humans and technology working together to scale a high-touch and responsive recruiting process.

Automation and a human touch don’t have to be mutually exclusive. Strategically combining them can deliver unrivaled results. In recruiting, automation’s infinitely scalable levels of efficiency mean that, regardless of the volume of candidates, each receives a timely correspondence. For candidates, being kept in the loop with a thoughtful and sincerely worded email is what makes the difference.

Stacey Browning

Reduce bias. Sherry Martin of the Denver Public School system highlights how assessments can be analyzed for bias in language and outcomes and adjusted over time to minimize adverse impact, ideally leading to a wider variety of job candidates.

Simpler sourcing. Sourcing is a popular aim for up-and-coming AI tools, and Heather Doshay at Rainforest QA talks about the impact of improving the ability to match candidates to jobs.

Sourcing is a time-intensive pain point for most talent professionals, and providing well-matched candidates to companies would significantly speed up the top of the recruiting funnel and increase the quality of hires.

Heather Doshay

Replace administrative tasks. This comment from John Feldmann at Insperity Jobs groups together the time-consuming but critical tasks that are part of nearly every recruiting process.

AI will be valuable in automating repetitive recruiting tasks such as sourcing resumes, scheduling interviews and providing feedback. This will allow recruiters and HR managers the opportunity to focus on strategic work that AI will most likely never replace, such as connecting with top talent, providing a more personalized interview experience and establishing training and mentoring programs.

John Feldmann

Stay compliant. Compliance is a critical concern in recruiting and Char Newell thinks that AI could help automate this aspect of the workload for recruiting organizations.

Defining AI

Machine learning and deep learning are two phrases that are related to AI, and I want to be clear on them before proceeding. Here’s the quickest clip I could find on Youtube to get some clarity:

Video from Acadguild tutorial on Data Science

Artificial intelligence is any code, technique or algorithm that helps a machine mimic, develop and demonstrate human behavior.

Machine learning is the techniques and processes by which machines can learn the ways of humans.

Deep learning is drawing meaningful inferences from large data sets, requiring artificial neural networks.

Deep learning is a subset of machine learning, which is a subset of artificial intelligence. These three terms are related but not interchangeable.

Robots will take our jobs

Is artificial going to displace humans? I hear this concern a lot and this article in Mother Jones does a good job articulating the reality of robot colleagues.

Illustration by Roberto Parada

I realize now that a lot of these projections about the rapid acceleration of computer learning rely on Moore’s Law – the historically-true law that computing power (in the original case, transistors) double about every two years). However, Moore’s Law may eventually break down, and outcomes of advancement don’t always match our expectation. For instance, the rise of the computer age led many to assume that paper would soon be phased out… yet we are using more of it than ever.

The most interesting section of this article was the markers we should look for if AI really is taking our jobs:

  • A steady decline in the share of the population that’s employed
  • Fewer job openings than in the past
  • Middle-class incomes flatten in a race to the bottom
  • Corporations stockpile more cash and invest less in new products and factories
  • Labor’s share of national income decline and capital’s share rise

And… hmm. A few markers there but 2019 is looking a bit better than 2013 when this article was published.

Intro to AI via cartoons

waitbutwhy.com always delivers with the thoughtful and funny

Kicking off this month with a primer on AI from waitbutwhy.com. This is from 2015, so there have probably been leaps in AI since this was written, but it is helpful as an anchor.

There are three types of AI:

  • Artificial Narrow Intelligence: We’re here now. AI that can reliably deliver better outcomes than a human brain, such as navigating with a map on your phone and getting customized music recommendations.
  • Artificial General Intelligence: This is AI that can reason and solve problems. We’re not here yet.
  • Artificial Superintelligence: This is intelligence that greatly exceeds human intelligence across all spectrums. This could be closer than we think.

This primer on AI goes into a lot of detail on the exponential increase of computing power over time as well.

When I think about AI’s applications for recruiting and HR, I think of many areas that artificial narrow intelligence is just starting to be applied. Recommendations based on past choices can be generated with existing data. And I also see the potential for recreating bias; this comes up over and over again as a concern with AI for recruiting. I’m curious to read more about how that can be addressed.

Welcome to AI February!

Having touched companies’ recruiting processes in some capacity the entirety of my career, I’ve been fascinated with the ways recruiting has evolved in that time. As a snapshot of this change, ten years ago as a new recruiter, I sometimes called an advertising agency to help post retail openings in the local newspaper. Today in recruiting operations, we use Google Analytics and tracking pixels to see performance of online postings, and programmatic advertising can automate some placement decisions.

One of the most exciting advancements in the field is artificial intelligence. Already there are HR tech startups touting their use of AI to enhance and improve the recruiting experience. The tagline is generally something about removing human bias and increasing efficiency – great arguments to move towards technology!

As a data analyst and a former recruiter, I am both excited and skeptical. Can these tools do what they promise? Do they truly apply AI or is this advanced statistics dressed up? And finally – aren’t people a critical component of the hiring practice?

To dive into some of these questions and familiarize myself with the state of AI, I’ve decided to commit my February to learning more about it. Every day throughout February, I will learn something about artificial intelligence and share it here. I’m focused on the following areas:

  • Defining artificial intelligence. What counts as artificial intelligence? What is the difference between AI and machine learning or are they interchangeable?
  • What’s the state of artificial intelligence in recruiting? Who is already doing this and how successful is it?
  • What are the promises and pitfalls of AI?

I realize that these three areas are rich enough that I could probably devote a month (or more!) to each. My intention here is to more broadly explore, though, so I’ll touch on all of these in one short month.

I’m looking forward to this month of learning and hope you follow along! Feel free to share any interesting tidbits, resources or your own interest with me here or via email or Twitter. Here’s to learning!

Business Experiments

This afternoon I’m returning to Columbia Business School to speak to the Economics of Organizational Strategy class about business analytics in the HR space. I’m excited to be back at my alma mater to talk about my work. For this class, the professor assigned The Discipline of Business Experimentation from the Harvard Business Review. From the summary:

The data you already have can’t tell you how customers will react to innovations. To discover if a truly novel concept will succeed, you must subject it to a rigorous experiment. In most companies, tests do not adhere to scientific and statistical principles. As a result, managers often end up interpreting statistical noise as causation—and making bad decisions.

YES. All the buzz about big data and analytics fails to mention that if you are not testing correctly, your data may be driving you to make bad decisions! With businesses ramping up analytics capabilities in a big way, I think this happens much, much more than companies are willing to admit.

The example in the HBR article is about Cracker Barrel testing a switch from incandescent to LED lighting at its restaurants. At restaurants that installed the LED lights, traffic decreased. This would initially suggest that LEDs are bad for business. However, by digging deeper, executives realized that the LED lighting made the entrances look dimmer and customers turned away thinking the restaurant was closed. The LED lighting should have been brighter than the incandescent lighting, but individual store managers were going around the corporate lighting standards to install more incandescent lighting, thus making the store look brighter and more welcoming. Therefore, once these stores adhered to the new LED policy, they had fewer lights and were less luminous.

With all the attention around the tools and methods available with statistical analysis, I’m afraid this deeper digging may get short shrift.

It’s important that business professional not only dedicate more time to digging than analysis, but that we can speak the language of data and statistics. With so much more data available to businesses today, knowing how to use it for decision-making is a competitive advantage.

Magic Bands and A Great Big Beautiful Tomorrow

I’m fresh off what was probably my 20th+ family trip to Walt Disney World, and as always it was such a magical time. After so many visits, it’s fun to experience the parks with a first-timer. and this time it was my sister’s boyfriend, Frank.

The Carousel of Progress sparked a great lunchtime discussion about what technology will look like at the end of this century. The Carousel of Progress is an animatronic stage show that shows an American family at four points in the 20th century, highlighting technology innovations in the home. It kicks off at the turn of the century, showing electric lights and cast iron stoves, and the “future” scene takes places in the 1990s, highlighting virtual reality games, self-flushing toilets and voice-activated appliances. After the ride, we all dreamed up what a 2000s Carousel of Progress would contain. Here in 2016 we’d be the very first scene of the show. 

First day in the Magic Kingdom, Feb. 2016. Em, Frank, me.
 We didn’t have to stretch too far to imagine what that scene would contain – we were wearing it. Disney World has integrated wearable technology into its parks with the Magic Bands. We all had a Magic Band in a chosen color with our name printed on the inside and used it as our hotel key, our park ticket, our credit card… It was everything. Meanwhile, all my ride photos popped up immediately in my Disney mobile app, presumably linked through my Magic Band.

Working in analytics, I am curious about the other side of the Band – the big data side. I would love to get a glimpse at what the data scientists at Disney are conjuring up with all the movement and spending data they get through the Magic Bands. Disney has spent over $1 billion on the MyMagic+ program so it’s clear they have big plans for ROI. So far reception has been positive, and the overall creepy factor I felt when my face first magically appeared on my Disney app – screaming on the new Seven Dwarves Mine attraction – was quickly replaced by joy when watching the two-minute video of my family’s reactions at various points on the new ride. From Wired:

No matter how often we say we’re creeped out by technology, we tend to acclimate quickly if it delivers what we want before we want it.

Just like the Carousel of Progress, technological innovation moves fast, and even now Magic Bands are moving to the past. The next scene takes place in Shanghai Disney, where everything will happen seamlessly through a smartphone. I’m adding Shanghai Disney to the bucket list! 

Better Persuasion through Behavioral Economics

A recent Freakonomics podcast called “The Maddest Men of All” perked up my ears. Rory Sutherland, the vice chairman of Ogilvy & Mather in the UK, talked about applying behavioral economics to advertising. Behavioral economics is the study of how individuals react, and while economics assumes rational actors, behavioral economics asks what people really do.

Of course, this has been done for ages in advertising – “Hurry! This offer won’t last long!” The example in this story applied these principles to persuade customers to keep their news subscription through call center interactions.

How to harness behavioral economics for yourself:

1. Channel (gentle) peer pressure

Knowing others are making a similar choice is a very motivating force for us social creatures. “Many people like you are doing this.” Choices presented in this manner still offer options but nudge the listener to the majority action. 

2. Help people avoid loss 

Dan Ariely has written a lot on loss aversion. People are more likely to avoid a loss than reach for a win. Language like, “I wouldn’t want you to miss out” drives the listener to take action. 

3. Stay positive in the face of negativity. 

Even when reading the terms and conditions, call center employees were encouraged to use a friendly tone and phrases like, “I am confident that…”

I like these takeaways because they can be applied broadly and show measurable results. In this study, using one of more of these techniques were three times more likely to be successful in retaining customers. 

How can you creatively apply these to drive action in your business? 

Five Reasons to Get Excited about IBM’s Watson Analytics

This week I saw a demo of Watson Analytics, IBM’s new natural language-based analytics tool. It was presented at the New York Strategic HR Analytics Meetup, though the tool itself is not specialized to any one business function – and that is a big part of its appeal.

My initial thought was, “Uh oh, there goes the boom in data analyst jobs!” I also felt really excited at the idea of analytics operating like a Google search bar. Number crunching for the masses!

The past few years have been an exciting time to be following data and analytics. With market intelligence startups like Food Genius getting regular mentions in the press and tools like Qlik and Tableau entering the common language of non-data heads in the business, it’s cool to see where analytics is going next.

Here are the top five reasons to get excited about Watson Analytics (and one reason to be weary):

  1. You’re no longer at the mercy of your IT team.

You can pull in data from Excel spreadsheets, Oracle, Salesforce, etc. The data has to be uploaded to IBM’s cloud in order to use it in Watson, but data selection and clean-up takes place as a step in the analysis. Current solutions on the market try to approach this by doing transfers and loads on their own, but there’s a lot of planning ahead – and partnering with the data managers in IT – required before you can move data into any system for analysis.

  1. Data clean-up is a cinch.

In the demo, after the rep chose a question, the next page guided clean-up, the process of organizing and formatting the data. IBM estimates that data preparation can take 50-70% of the completion time of a data mining project. To be fair, Excel does this pretty well but has its limits with huge data sets.

  1. The data looks pretty!

Any analyst worth her salt knows that getting the right data story is only half the battle. No one cares about a pivot table – they want to see cool graphics. Compelling visualizations are the vehicle for getting data into viewers’ heads. The visualizations in the Watson demo look good – they have lots of information without looking too busy, they connect well to the data they’re showing and the colors and shapes are bold and appealing.

  1. It’s not relegated to a single business function.

There are great analytics tools for marketing, great analytics tools for operations, great (or at this stage, maybe just good?) analytics tools for human resources. The beauty of Watson Analytics is that it doesn’t specialize. I hope movement towards tools like this leads to asking deeper questions of the data, like how do disparate business functions work together and drive productivity, sales, etc.

  1. Analytics is no longer the domain of analysts.

This tool doesn’t require a PhD to understand a multivariate regression. Analytics and data storytelling are now in the hands of anyone with a question that can be answered with the data on hand.

And my bonus point, one reason to be weary:

  1. Data analysis is only as good as the people communicating it.

A huge job in building good data stories is communicating the finer points of the analysis to those less data-savvy. A difference of 2% can mean nothing or so much depending on the sample size. A regression is only as good as the variables you’re putting into it. These basic statistical concepts may be lost with easy-access analysis. Data is powerful, and it should be wielded wisely!

Ultimately, I am excited to get my mitts on a trial and signed up at IBM’s site. You can too here.

Big data on a little tablet. The future is here!
Big data on a little tablet. The future is here!

So what do you think? Is natural language analytics the next big thing? Will we all be data analysts in the near future?