Confirmation bias is the well-documented tendency to notice, seek, and remember evidence that supports what we already believe, while discounting evidence that doesn’t (Nickerson, 1998). It is not a character flaw — it is a default setting of ordinary human reasoning. Exploitation happens when someone feeds that default on purpose: curating a stream of information that only ever confirms your existing view, so your conclusions feel independently reached when they were in fact spoon-fed.
The tactic is powerful precisely because it is invisible from the inside. Each confirming item feels like more proof, and growing confidence feels like being right rather than being managed. Recommendation algorithms industrialize the process, learning what holds your attention and serving more of it; partisan media and radicalization funnels do it deliberately, pairing affirmation with a villain to resent. The result is a belief that grows stronger while its actual evidence base does not.
The defensive tell is asymmetry: honest inquiry exposes you to disconfirming evidence and gets more cautious under scrutiny, while manipulation shields you from it and gets more certain. When a source has never once made you uncomfortable, it is not confirming your judgment — it is farming it.