Is the Algo Right?
Sometimes we are good predictors of what to expect from a computer’s output, other times well we can be off by tremendous margins. I just ‘googled’, Love sucks! Let’s see what my social feeds will conjure up! Diving in head first into the wild and wacky world of algorithms — those digital magicians that sometimes get it right, and other times, well, let’s just say they end up doing a little dance in the wrong direction!
Back in yore, Robert Morris developed the self-replicating program to test the boundaries of the Massachusetts Institute of Technology (MIT) network, — a sort of penetration testing. Yes he had valid points to test out his theory as any researcher would, but unfortunately, the program didn’t run as expected, This became the first malware that utilized your resources while replicating itself and finally denying you service — (DoS attack).
Last week, while scrolling through my entertainment app’s “Because you watched Queen Cleopatra” section — or whatever they call it these days, — I stumbled upon a gem. The app suggested I give “Alexander: The Making of a God” a whirl, boasting a 98% match rating for my viewing pleasure. And you know what? It hit the spot! Like when I find your G-spot. The series’ portrayal was spot-on, and I was hooked from start to finish. I feel like I have used a lot of ‘spots’ back there…mmh..anywho — Cheers to the algorithm for hitting the mark and delivering exactly what my brilliant mind craved.
Another interesting example is that of the Twitter bot called Tay. At first, it seemed to be a rosey affair with chocolates and wine on the side, It drastically took a wrong turn — like Wrong turn The movie. It was a massacre on the streets of Twitter by Tay! The idea behind this was that the more you chat with Tay said Microsoft, the smarter it gets, learning to engage people through “casual and playful conversation.”The starts of the conversations were thoughtful and nice but as the conversations grew deeper and more sinister the tool became a mixture of Hitler and Benito Mussolini with a hint of Jeffrey Dahmer.
That just sends chills down my spine!
And speaking of funny blunders, The Nikon Camera and its blink detection system is our next stop. The idea here was not to snap photos when it saw closed eyes. That makes sense right? When the damned thing arrived in the Asian market the Camera wasn’t ‘cameraing’ this was because our friends have Monolid eyes. The algorithm thought their eyes were closed. This was a mistake of bias that we can attribute to the type of dataset these algorithms are trained on.
The rapid adoption of algorithms is transforming various industries, with key sectors like Financial Services, Transport and Logistics, and High Tech Communications leading the charge… Like when a rhino charges, Powerful stuff…, The above examples may not directly impact our daily lives, but it’s crucial to recognize the growing influence of this technology and its potential implications. We can already witness its biased impacts in Western countries that use this technology in their healthcare and Judicial environments.
As we integrate algorithms into our everyday routines, we must proactively address issues of bias and fairness. Organizations like the Algorithm Justice League are working tirelessly to establish standards and promote equity in algorithmic decision-making. By acknowledging and understanding the complexities of algorithmic bias, we can pave the way for a more inclusive digital landscape where diverse perspectives are valued and user experiences are not hindered by discriminatory algorithms.
Teddy Lumidi aptly reminds us that embracing these challenges allows us to strive for a future where technology serves everyone equitably, fostering innovation and progress while upholding fundamental principles of fairness and equality. Let’s join forces to ensure that algorithms work for the betterment of society as a whole.
