Ray Kurzweil and Mitch Kapor have a $20,000 bet over whether a computer will pass the Turing Test by 2029.
Kapor bets no, arguing human knowledge is tacit and experiential, not book-learnable.
Kurzweil bets yes, citing the representational power of language and progress in AI.
What happened
This article outlines a 27-year bet (2002–2029) between Ray Kurzweil and Mitch Kapor over whether any computer or "machine intelligence" will pass the Turing Test. The stake is $20,000, going to the Electronic Frontier Foundation if Kapor wins, or the Kurzweil Foundation if Kurzweil wins.
Why it matters
The bet centers on the essence of the Turing Test—whether a computer can impersonate a human in text-based conversation well enough to fool a judge. Kapor argues that human knowledge is largely tacit and experiential, not found in books, so a machine cannot acquire the depth needed to pass. Kurzweil counters that written language represents human-level thinking, and that machines are already passing narrow forms of the test, like Deep Blue's chess play.
What to watch
The outcome hinges on whether Kurzweil's approach—reverse-engineering the brain with nano-scale technology to simulate its algorithms—can achieve human-level intelligence by 2029. Kurzweil predicts it will, while Kapor doubts it, expecting that the brain-as-computer metaphor will be superseded.
Ask the AI about this article →
This bet, placed in 2002, frames a central AI debate: whether human intelligence is fundamentally replicable in machines. Kurzweil's optimism relies on the idea that the brain's information processing can be reverse-engineered, but Kapor counters that human cognition is rooted in embodiment, emotion, and tacit knowledge—factors not captured by book learning or algorithms. The decades since have seen AI advance in narrow tasks, yet the Turing Test remains a benchmark, highlighting the gap between data-driven responses and genuine understanding.
The bet's structure—with proceeds to rival foundations—underscores the philosophical stakes. Kurzweil's reference to Deep Blue's chess victory suggests progress, but Kapor's skepticism about the limits of such narrow successes remains a caution, as expert systems in the 1980s failed to generalize beyond rigid domains. Whether machines will achieve true human-like conversation by 2029 is unresolved, but the argument highlights deeper questions about consciousness, creativity, and the nature of intelligence itself.
For example, today's edition would include:
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. Q&As are published on this page for other readers too.
Infosys, a founding partner of CrowdStrike's Project QuiltWorks, is bringing enterprise context to CrowdStrike…
Cisco has announced a target to have zero engineers writing code by the end of October
OpenAI's latest model, Astra, can do more thinking off the scratchpad, according to Transformer
Cecilia Ziniti, former general counsel at Replit, left the company in November 2023 and founded GC AI, an AI s…
Governments and companies outside the US and China are building local AI models to avoid relying on foreign sy…
Users report Instagram's "AI Content" label is wrongly appearing on images they edited with tools like Canva's…
