1966
ELIZA at MIT
A 1966 chatbot fooled people with simple rules
Joseph Weizenbaum’s ELIZA, built at MIT, played a therapist by spotting keywords and turning your own words into questions. It understood nothing, yet some users confided in it. Seeing understanding where there is none is still called the ELIZA effect, which is worth remembering with today’s fluent AI.
Source: J. Weizenbaum, Communications of the ACM (1966)