Explore Ray Kurzweil's AGI and superintelligence timeline: 20 sourced statements from 1990 to Oct 24, 2024. Compare what was said, when it was said, and any predicted arrival date.
This is a curated record, not a complete history or a scorecard. AGI and ASI labels group related language; the original wording and caveats remain below. How we document claims.
Ray Kurzweil's statements, earliest to latest
Prediction · AGI
Expected 2020–2070
Established, broadly accepted success on a full Turing test would arrive sometime between 2020 and 2070.
Source summary · not a verbatim quotation.
Kurzweil distinguished narrow successes and disputed demonstrations from broadly accepted unrestricted intelligent conversation. He explicitly cautioned that his estimate might be inaccurate.
By 2029, computers would routinely pass apparently valid Turing tests, while debate about the full breadth of their intelligence continued.
Source summary · not a verbatim quotation.
The 2029 chapter forecasts autonomous learning and knowledge creation with little human involvement. It also preserves controversy about whether machines match every aspect of human intelligence; its 2019 forecast was only for disputed test reports.
By 2029, the software needed for intelligence would be largely mastered.
Source summary · not a verbatim quotation.
The article forecast machines expressing a full range of human intellectual and physical skills within several decades. Its separate 2019 date referred to computing power, not completed intelligent software.
“we will achieve machines in a few decades that have the complexity and richness of human thinking”
Kurzweil forecast machines with human-like behavioral complexity. He discussed consciousness as a philosophical question rather than claiming a current system was conscious.
Bet that a machine would pass the agreed Turing test by the end of 2029.
Source summary · not a verbatim quotation.
His wager with Mitch Kapor specified lengthy text interviews and judging criteria. Passing this particular test was the proposed proxy for human-level intelligence.
He described this as machine intelligence becoming behaviorally indistinguishable from biological intelligence. This is a renewed forecast in a separate interview.
By 2029, machines should combine human cognitive flexibility with superior memory and speed.
Source summary · not a verbatim quotation.
He reiterated that both the required hardware and software would exist in 2029. His threshold encompassed the breadth of human intelligence, not only a narrow task.
He forecast that machines would combine human-like reasoning with far greater memory and reading capacity. Contemporary systems were examples of progress, not declared AGI.