The case against Google's claims of "quantum supremacy"
Google’s renewed claims of “quantum supremacy” and a recent quantum error-correction milestone are met with strong skepticism, with critics arguing that the underlying benchmarks are contrived, poorly verifiable, and potentially overhyped. Commenters contrast the media-fueled perception of an imminent quantum revolution – including fears for current cryptography – with the reality that even modest demonstrations of Shor’s algorithm remain out of reach. The exchange broadens into how much to trust unverified quantum results, whether secret state projects could be further ahead, and what meaningful, real-world applications or investment strategies might look like in a post-quantum future.
Context: Google’s “Supremacy” and Willow Claims
- The blog post revisits Google’s 2019 random circuit sampling “quantum supremacy” result in light of the new Willow chip and a dramatic “septillion years” classical runtime claim.
- Some see nothing fundamentally new: same benchmark, slightly larger circuits (53→67 qubits), and argue we could be repeating this cycle with larger devices in future years.
Verification, Benchmarks, and Hype
- Core concern: random circuit sampling isn’t practically important and is hard to verify independently, since the whole point is to exceed classical simulation.
- Earlier “billions of years” estimates were later challenged by better classical algorithms, suggesting the original gap was overstated.
- Several commenters argue Google’s communication style is overly hyped compared to the underlying incremental technical progress.
Shor’s Algorithm and Actual Capability
- Multiple comments note that genuine implementations of Shor’s algorithm haven’t gone beyond trivial numbers (15, 21), and even those used heavily simplified, “cheating” circuits.
- This is contrasted with public perception that widespread cryptographic breakage is imminent.
Quantum Error Correction and Scalability
- The blog author in-thread emphasizes statistical anomalies in Google’s 2019 fidelity modeling (too-perfect fit), not the existence of exponential decay per se.
- Others counter that multiple independent experiments show the expected exponential fidelity falloff consistent with simple noise models, and view the critique as increasingly out of touch.
- There is active disagreement over whether quantum error correction will scale or fundamentally break down.
Security, Intelligence Agencies, and Post‑Quantum Planning
- Debate over whether intelligence agencies might already have secret, more powerful quantum machines; many doubt they could be dramatically ahead of large companies.
- Some stress “store now, decrypt later” risks and blackmail value of old secrets; others argue the storage and collection costs at global scale are enormous.
- On migration: centralized banks are seen as more able to coordinate a post‑quantum transition than decentralized cryptocurrencies, which face governance and overhead challenges.
Applications and Value Proposition
- Mentioned potential uses: chemistry and materials simulation, optimization (e.g., logistics, rail networks), quantum networking (superdense coding, quantum key distribution).
- Several commenters note many current industrial “use cases” look like PR toy problems rather than real business advantages.
- A minority dismiss quantum computing as “vaporware,” while others argue progress is slow but real, comparable to long-running efforts like nuclear fusion.
Skepticism, Optimism, and the Role of Critics
- Thread repeatedly returns to the value of informed skeptics: they may be wrong long‑term but help refine methods and temper overclaims.
- Others warn against both nitpicky contrarianism and uncritical hype, emphasizing that extraordinary claims require robust, reproducible evidence.