Can data analytics narrow the gender gap?

16

October

2019

5/5 (1)

Companies are tapping into their data by using advanced analytics techniques in finance, marketing, sales in order to gain competitive advantage against their competitors. With the rise of artificial intelligence – machines making decisions instead of humans – it is hard not to wonder whether those decisions are ethical. Those techniques can discriminate (women, for instance) or highlight human biases and aid ethical decision-making.

On the shadow side of AI-driven decision making, we see Amazon for which automation has been a key in the industry dominance. However, in 2015, Amazon found out that its secret AI recruiting tool was picking men over women for tech positions. It happened because the algorithm was learning using historical data of a company where men outnumber women, thus making it biased (Dastin, 2018). Another example, which was found by Lambrecht and Tucker (2019), is that gender neutral advertisements (i.e. STEM careers ad) on Facebook were shown more to the men because of cost-optimizing algorithm which discriminated women who were a prized demographic and more expensive to display ads to.

On the other hand, things are not as gloomy as they may seem. There are also examples of data analytics tools which are used towards eradicating the gender gap. As people are becoming aware that at a current rate it would take 202 years to achieve gender parity (Sanz Sáiz, 2018), they are rolling up their sleeves to quicken the pace using AI algorithms and Data Science to solve this problem. An example of these efforts is Syndio a Human Resource analytics platform. It strives to solve gender pay issue by providing insights to companies on where they stand in the gender pay parity journey. Syndio helps companies in three ways – encourages them to commit to ongoing pay analyses, use valid methodology to analyze the data and ensure transparency. Their clients are able to see dashboards in which they can identify the changes they need to make in order to compensate women fairly. It is important to note Syndio’s algorithms have been created together with National Women’s Law Center and has been vetted by federal and state agencies (Kasunich, 2019) and have the potential to become a standard procedure.

As we are still grasping the potential of AI algorithms, it is eventual that some solutions may backfire. However, I believe that we can use these mistakes as a cautionary tale and become more rigorous in the way we create them. On the bright side, as the saying goes “what gets measured gets done”, I think that transparency which data analytics can bring to the table, would help us to make decisions based on reality rather assumptions and therefore make decisions more ethical.

References
Dastin, J. (2018). Amazon scraps secret AI recruiting tool that showed bias against women. Retrieved 16 October 2019, from https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G
Kasunich, C. (2019). Syndio Establishes Pay Equity Standards for Global Corporations Based on Validated Methodology. Retrieved 16 October 2019, from https://www.prnewswire.com/news-releases/syndio-establishes-pay-equity-standards-for-global-corporations-based-on-validated-methodology-300839523.html
Lambrecht, A., & Tucker, C. (2019). Algorithmic Bias? An Empirical Study of Apparent Gender-Based Discrimination in the Display of STEM Career Ads. Management Science.
Sanz Sáiz, B. (2018). Five ways data analytics can help close the gender gap. Retrieved 16 October 2019, from https://www.ey.com/en_gl/digital/five-ways-data-analytics-can-help-close-the-gender-gap

Please rate this

No time to exercise? Get AI-powered exercise dose in 9 minutes

12

October

2019

5/5 (1)

It is not a surprise that Artificial Intelligence is being applied in numerous sectors – we see virtual chefs, superhuman doctors, chatbots as sales assistants. The field of fitness is not an exception. As lack of time is being quoted the most as a reason for not exercising regularly (Green, 2019), researchers are interested in looking into how to optimize our fitness habits in order to get maximum returns for our efforts (Green, 2019).

One of the products of interest in this area is CARdiovascular Optimization Logic (CAR.O.L) which is an Artificial Intelligence powered exercise bike which wants to change the way we do exercise by providing the optimal and personalized exercise solution to each of us. It optimizes each workout and meticulously calibrates the intensity based on how fast you lose muscle power (CAR.O.L, 2019). To be more precise, CAR.O.L Fit Ai exercise session includes 2 minutes of warm-up, 20 sec sprint, 3 minutes recovery alongside another 20 second sprint and 3 minute cool-down (CAR.O.L, 2019). This regime focuses on depleting glycogen resources as quickly as possible, as the depletion triggers your organism to improve fitness, fat metabolism and to remodel the muscles (CAR.O.L, 2019).

CAR.O.L Fit AI was founded by mechanical engineer Ulrich Dempfle and microbiologist in training Ratna Singh who has spent her career advertising and management consulting and was the first entrepreneur-in-residence at McKinsey & Company’s in Silicon Valley (Asprey, 2019). They clinically tested the exercise bike through a randomized peer-reviewed controlled trial sponsored by American Council on Exercise (Green, 2019). Surprisingly, people who were doing 40 seconds of personalized intense exercise three times a week outperformed people who were jogging 30 minutes five times a week. CAR.O.L group received double the gains for a fraction of their time – they improved their cardio fitness by 78% blood pressure by 196% and blood sugar by 48% more than the jogging group (Green, 2019). These findings show Artificial Intelligence abilities to provide optimal exercise solution which maximizes the health benefits of exercise in the shortest time.

Personally, I enjoy exercising because I can do it with friends or while spending time in the nature, thus it is not as hard for me to convince myself to exercise. However, for people who do not enjoy exercising but would like to receive the health benefits and optimize their exercise regime, such AI application in the fitness realm can be an attractive solution.

References

Asprey, D. (2019). [Video]. Retrieved from https://www.youtube.com/watch?v=8ZJS8y0NjCk
CAR.O.L | Stationary Bike for High Intensity Interval Training (HIIT). (2019). Retrieved 12 October 2019, from https://carolfitai.com
Green, D. (2019). What Are the Acute and Chronic Responses to Reduced-exertion High-intensity Training?. Retrieved 12 October 2019, from https://www.acefitness.org/education-and-resources/professional/certified/may-2019/7267/ace-sponsored-research-what-are-the-acute-and-chronic-responses-to-reduced-exertion-high-intensity

Please rate this