Select your language: EN JP

Don’t worry about whether or not Google achieved quantum supremacy

Lewie Roberts, Senior Research Associate
October 23, 2019

Recently, a number of news sites rushed to report that Google claimed it could display quantum supremacy, an exciting-sounding term that  – like most technical jargon – can be given slightly different meanings depending on who you're speaking with. The basics of quantum supremacy revolve around how algorithms practically scale (how long they take to solve a problem with increasing inputs) on classical computers vs quantum ones; importantly, a problem chosen to test for quantum supremacy generally does not need to be useful, and can be selected simply to highlight the strengths of a quantum computer.

Source: Google

In nontechnical terms, the original report indicates that a Google researcher claimed to prove that an undisclosed problem (perhaps a metric Google has described before) could be solved by using its quantum computer, while an existing supercomputer could never do so in any practical time; the paper described a ratio of 20 seconds to 10,000 years. Although big wins over classical architectures are important for the advancement of research in applied quantum computing, the technical metrics that researchers will use for validating the technology are unlikely to be immediately useful for any practical applications. In addition, academics will likely debate the validity of Google's results over the coming months (indeed, IBM has already provided an argument). Regardless of the outcome, know that this news doesn't indicate major disruption coming anytime soon.

By contrast, many startups in the quantum computing space (see Lux's Tech Page for a full market map) are attempting to use the technology to achieve incremental gains in computation, as opposed to massive ones. Example problems include those that companies are claiming to solve measurably faster (see HQS Quantum Simulations) or more accurately (OTI), or provide tangential results to existing tools (ProteinQure). While the eventual goal is to develop a wide library of massive improvements, like those that Google is researching, we recommend focusing on exploring work in the near-term applications and how they compare to today's existing tools.





- Blog: Can AI Make Innovation Any Easier?

- Analyst Insight: IBM points out chemical simulation applications for near-term quantum computing (Members Only)

- Tech Page: Quantum Computing (Members Only)

- Blog: $6.1 Billion in Funding and China Still Not Leading in AI Innovation 

Download Case Study  Schedule Demo