In the last month of 2018, synced gathered all the artificial intelligence failures that have happened so far this year. Artificial intelligence has made breakthrough and some scientists are still working to create a masterpiece like Master Algorithm by the great Pedro Domingos who can crack all the problems proposed by the human mind. But nothing in this world can be made flawless hence everything accompanies some errors and blunders in them.
Therefore, we must be reviewing some AI failures of 2018. This article has no such aim of degradation or mocking, we are just here to highlight some loopholes of the research so that we can avoid them in the future.
Here are the 10 most significant failures made by Artificial Intelligence in 2018.
1. Uber’s self-driving car killed a pedestrian
In the history of auto-vehicle related pedestrian death, the self-driving Uber SUV hit a female pedestrian in Arizona on March 2018. The Uber’s auto-mode was unable to install the emergency breaking system due to which it was unable to recognize the pedestrian. Due to this mismanagement, the autonomous mode was suspended in the US by both the Uber and Toyota. However, the suspension was revised after the 8 months to the accident.
2. DeepFakes reveals AI’s unseemly side
Last December, many porn videos were uploaded on Reddit. These videos were featuring the most rated international celebrities. DeepFakes started swapping the celebrities faces with the porn stars. However, face swapping was considered as something bad at that time. but DeepFakes showed that if you have a pretty face you can make your own fake videos.
The social pages were flooded with such fake videos, but the outcomes were serious and alarming that made it a serious failure.