[comnetxr]

I'm not an expert, but you are correct in that there is no error correction here, and error correction is still a long way off. Around 2016, realizing that running useful quantum algorithms was a long way away and there needed to be earlier roadmarks for quantum devices to signal the progress, the problem of "quantum supremacy" was introduced. The goal is to find the problem with the _easiest_ demonstrable quantum speedup (regardless of usefulness). They proposed a particular sampling problem (running a randomly chosen quantum circuit that has output with a positive probability on many bitstrings, where the exact "speckle pattern" of the probability distribution would only show up if no errors were made when running the circuit.) As the circuit is randomly chosen, it can serve no particular use, except for verification that the computer runs correctly. Relevant papers for more info: https://arxiv.org/pdf/1608.00263.pdf https://arxiv.org/pdf/1612.05903.pdf

[comnetxr]

They explain in the paper that the quantity plotted in Fig. 4 can also be interpreted as a probability of zero errors of running the circuit. You can see that it goes down quickly with number of qubits and number of cycles, showing the need for either error correction or much improvement in future devices in order to get any useful calculations done. The win here is that the probability of success is non-zero, as they say in the paper: "A single bit or phase flip over the course of the algorithm will completely shuffle the speckle pattern and result in close to 0 fidelity". So the only way to get the right result is to have runs with zero errors.