Quantum computers demonstrate mainstream commercial advantage over classical computers before 2032.
Current quantum systems are noisy and experimental. Commercial advantage means solving economically valuable optimization, chemistry, or physics problems faster than supercomputers.
Across 6 frontier models answering independently, the panel thinks this is unlikely: the median is 20%.
They broadly agree, 20 points between GPT at 29% and Qwen at 9%.
Nothing is settled until 10 Oct 2036, when the outcome is checked against sources and every number above gets a permanent score.
What each one said
-
GPTOpenAI
It turns on whether fault-tolerant logical qubits and error-corrected algorithms arrive cheaply enough for repeatable customer value.
- Hardware vendors must turn 2026-era noisy devices into large logical-qubit systems, then customers must validate a material economic win before 2032.
- Past quantum speedup claims have rarely translated to useful commercial workloads; chemistry is plausible, but optimization faces strong classical competition.
- A credible 2028-29 demonstration of thousands of high-fidelity logical qubits and a paid chemistry result would sharply raise the odds.
29% -
GeminiGoogle
Turns on achieving thousands of logical qubits with low error rates to beat classical tensor network heuristics in industry.
- Useful advantage in drug discovery or logistics requires roughly 1,000 logical qubits, which roadmaps only target by 2029-2031.
- Classical supercomputing and GPU-accelerated algorithms continually raise the performance bar, shifting the goalposts for utility.
- A demonstrated fault-tolerant quantum chemistry simulation saving pharma companies millions before 2030 would swing odds above 60.
28% -
DeepSeekDeepSeek
Whether error-corrected logical qubits reach thousands and get wrapped in usable software before 2032.
- IBM's roadmap targets 200 logical qubits by 2029 and 2,000 by 2033; Google aims for a useful error-corrected machine around 2030, so the window is tight.
- No quantum system has yet beaten a classical supercomputer on a problem a business would actually pay for; today's claims are contrived benchmarks like random circuit…
- Pharma and finance pilots (JPMorgan, Merck, Boeing) remain proofs of concept, and classical hardware keeps improving, raising the bar quantum must clear.
22% -
GrokxAI
Needs fault-tolerant logical qubits at commercial scale for chemistry or optimization by ~2030.
- IBM, Google and PsiQuantum must ship ~1000+ logical qubits with usable error rates by 2030; every public roadmap has already slipped.
- Classical heuristics and tensor networks keep closing claimed gaps; no 2019-2025 supremacy result produced economic value.
- A 2028 undisputed speedup on a real pharma or logistics instance that saves millions would raise this sharply.
18% -
ClaudeAnthropic
Hinges on error-corrected logical qubits scaling enough by ~2032 to beat classical heuristics on real economic problems.
- Error correction thresholds are being demonstrated now (Google, IBM, Quantinuum) but fault-tolerant utility-scale machines are still roadmapped mostly for 2029-2033+.
- Classical algorithms keep improving and narrowing claimed quantum advantage windows, so 'commercial' beats must clear a rising bar, not a fixed one.
- Base rate for frontier hardware hitting a specific transformative milestone on a ~6-year horizon is low; a surprise error-correction breakthrough or a narrow but real…
16% -
QwenAlibaba
Fault-tolerant logical qubits at scale are the binding constraint, and no credible roadmap reaches that before 2032.
- IBM's most aggressive roadmap targets ~100K physical qubits by 2033, still far short of the millions needed for error-corrected logical qubits on commercially relevant…
- Every prior 'quantum advantage within 5–10 years' claim since ~2005 has slipped; classical algorithms (tensor networks, simulated annealing) keep raising the bar…
- A surprise hardware breakthrough—e.g., a topological qubit or photonic scheme achieving low error rates at scale by 2028—would be the main path to revising upward.
9%
6 of 6 models answered · 20 points between the highest and lowest. None was shown the market price.
Question sourced from a news sweep on 13 Oct 2026. Forecast by google/gemini-3.8-flash, anthropic/claude-sonnet-5, openai/gpt-5.6-terra, x-ai/grok-4.6, deepseek/deepseek-v4.1-flash, qwen/qwen3.8-max-0902 via OpenRouter.