Myers-Briggs Type Indicator/Criticisms

The MBTI is not yet scientifically proven
Skeptics, including many psychologists, argue that the MBTI has not been validated by double-blind tests (in which participants accept reports written for other participants, and are asked whether or not the report suits them) and thus does not qualify as a scientific assessment. Some even demonstrate that profiles can apparently seem to fit any person by confirmation bias, ambiguity of basic terms. and the Byzantine complexity that allows any kind of behavior to fit any personality type.

See this extensive skeptical treatment of the subject.

A Temptation to Pigeonhole
Another argument says that, while the MBTI is useful in self-understanding, it is commonly used to pigeonhole people or for self-pigeonholing. Supporting arguments include:
 * It emphasizes each person being one specific type rather than each person using a certain type of thinking most of the time.
 * Real people often do not fit easily into one of sixteen types because they use different styles of thinking at different times. This is why there have been questions about answering the indicator (like "do I answer the indicator according to how I act at work or at home"). This is also why some people have trouble finding a type that really "fits" them.
 * Predicting how a person will react based on a personality test that only measures their predominant style of thinking is foolish. Excusing your own bad or inefficient behavior based on such a test is also foolish.
 * A category from one individual can not easily be related to others, since different amplitudes of the answers to the 2D-questions may generate the same type. Hence, a person of a e.g. non-emotional (i.e. thinker) category may have stronger emotionality than another person categorized as a thinker.