Join Our Growing Quantum Community

Breaking down Google’s latest quantum supremacy claim.

Google’s quantum team is back with another "we’ve outclassed classical computers" announcement—this time with a 70-qubit experiment claiming to perform a task beyond reach of existing supercomputers. But what does this actually mean, and how does it compare to their 2019 Sycamore milestone?

The Core Claim

1. Problem: Random circuit sampling (same as 2019, but scaled up).
2. Hardware: New 70-qubit "Sycamore-3" processor, allegedly with lower error rates.
3. Classical Benchmark: Google estimates it would take Frontier (the world’s fastest supercomputer) ~47 years to simulate their 70-qubit circuit.

Key Questions

1. Is this really supremacy, or just a harder benchmark?
2. Unlike 2019, no classical team has contested the claim yet—partly because simulating 70 noisy qubits is tedious, not necessarily impossible.
3. Critics argue the task is still synthetic (no practical use), but Google counters that it’s about proving scalability.
4. How improved is the hardware?
5. The paper hints at better gate fidelities, but full specs aren’t public. If true, it’s a quiet win for error mitigation.
6. Still no error correction, though—this is NISQ-era supremacy.
7. What’s next?
8. Google’s roadmap suggests 100+ qubits by 2025, but the real goal is logical qubits. For now, they’re dominating the "quantum Olympics" (winning at artificial benchmarks).

Broader Implications

1. For IBM/Quantinuum: Pressure’s on to demonstrate similar scale.
2. For theory folks: The classical simulation counterattack will be interesting—expect clever tensor network optimizations.
3. For everyone else: Still zero practical applications, but the engineering feat is undeniable.