The Role of Phenomena in Reason Bias
Scott Foote
Cybersecurity Executive, Board Advisor, CISO, Chief Privacy Officer/DPO, Chief Risk Officer, CAIO | CISSP, CCSA, CCSP, CISM, CDPSE, CIPM(IAPP), IDPP, AIGP, CRISC, GRCP, CISA
Thought for the Day… Anomalies (i.e., phenomena) in any set of observations hold great power, and often result in very positive or very negative interpretations.
On the positive side, anomalies can help to detect/expose malicious activity in human behavior (for example) that might otherwise have gone unnoticed.
In Cyber Defense, we use these phenomena productively every day as we hunt for malicious actors hiding on our networks, exploiting our organizations.
On the negative side, anomalies can create significantly biased summary interpretations (e.g., a skewed central tendency) of an entire set of observations.
An example of this… the media use phenomena destructively every day, as they ignore the “mundane” in favor of sensationalizing only extreme events to push their ratings… willfully allowing their audience to misconstrue the central tendencies of society.
The more I dive into Machine Learning (supervised and unsupervised), the more concerned I am that “artificial intelligence” will develop the same biased inferential and inductive reasoning that so many humans have come to embrace.
Copyright © 2022 Phenomenati – All Rights Reserved.
Cyber Security Strategist
2yWhilst on a project at the company we worked, I came to this conclusion at during an assessment discovering bias effectively programmed in. My conclusion was dismissed at the time but I never waivered from the conclusion that people can purposefully or "accidently" inject bias into these.