Will Google Quantum AI publish a peer-reviewed paper claiming to have crossed Milestone 3 before 31 December 2027?
deadline 2027-12-31
confidence low
tier 4 evidence
8%PolySignal estimate
Resolution criterion
Yes if Google publishes in a peer-reviewed journal a result that Google itself presents as reaching Milestone 3 of its error-correction roadmap (a logical qubit with a lifetime substantially better than the best physical qubit).
Verified starting state
Google validated Milestones 1 (2019) and 2 (2023–2024, below-threshold correction on Willow, 105 qubits). Milestones 3 and 4 remain active research targets, with no peer-reviewed publication. In March 2026 Google opened a second modality in neutral atoms. Google is targeting 2029 for the useful machine and took close to five years between M1 and M2.
What pushes it up
- Google’s recent expansion into neutral‑atom qubits suggests potential acceleration of error‑correction progress [tier 4]
What pushes it down
- No peer‑reviewed milestone 3 result yet; Google’s own timeline targets 2029 for a useful machine, implying a 5‑year gap similar to M1→M2 [tier 1]
- The lack of any announced or published milestone 3 evidence in the last year pushes the probability down [tier 4]
What would move this number most
Whether Google’s neutral‑atom program will deliver a logical qubit with lifetime exceeding the best physical qubit before the end of 2027.
How this number is built
| Deadline | 2027-12-31 |
| Base rate | 5% |
| Best evidence | tier 4 · 3 article(s) used |
| Independent runs | 4 · 12 (spread 8 pts) |
| Confidence | low |
| Evidence cap | not triggered |
| Last revised | 2026-09-27 09:15 |
| Model | openai/gpt-oss-20b (Groq) |
Sources consulted
Probability history
2026-09-27 09:15 8% 4 article(s) on file · openai/gpt-oss-20b (Groq)
- What pushes it up. Google’s recent expansion into neutral‑atom qubits suggests potential acceleration of error‑correction progress [tier 4]
- What pushes it down. No peer‑reviewed milestone 3 result yet; Google’s own timeline targets 2029 for a useful machine, implying a 5‑year gap similar to M1→M2 [tier 1]
- What pushes it down. The lack of any announced or published milestone 3 evidence in the last year pushes the probability down [tier 4]
- What would move this number most. Whether Google’s neutral‑atom program will deliver a logical qubit with lifetime exceeding the best physical qubit before the end of 2027.
2026-09-26 09:22 15% 4 article(s) on file · qwen/qwen3.8-27b (Groq)
- What pushes it up. Google has successfully validated M1 and M2, demonstrating a track record of hitting quantum milestones, though the time gap between M1 and M2 was ~5 years [tier 1 context].
- What pushes it up. The opening of a second modality (neutral atoms) in March 2026 suggests accelerated R&D capacity [tier 1 context].
- What pushes it down. The hand-verified state explicitly states M3 remains an active research target with no peer-reviewed publication as of the current date [tier 1 context].
- What pushes it down. Google's official target for a 'useful machine' is 2029, suggesting M3 (a key step toward utility) is unlikely to be fully published and peer-reviewed by end of 2027 [tier 1 context].
- What pushes it down. The provided recent press articles are mostly tier 4 aggregators or unrelated tier 2 news, providing no new tier 1 or 2 evidence of imminent M3 breakthroughs [tier 2/4].
- What would move this number most. Whether the 'substantially better' lifetime threshold for M3 has been achieved in internal labs but is being held back for a larger 2028/2029 announcement, or if the technical barrier is still unsolved.
2026-09-20 09:09 31% 3 article(s) on file · gemini-3.5-flash
Reasons not recorded for this revision (detailed history has only been kept since recently).