From the article, in bold:

The really exciting point comes when we can re-run all this analysis, basing it on actual job performance, rather than interview results

How will they gather data about the job performance of the people who did not pass the interview? Without that, the study is only halfway done.

That's true of any hiring system though: how do you gather data on the job performance of the people who don't pass a resume screen? A coffee date? A phone screen? Missing counterfactuals everywhere.

To run a full analysis, you need to hire people randomly - both people who fail and pass your hiring process - and assess their subsequent performance. This never happens in real life.

However, you can still run an informative statistical analysis based on the variability in interview scores and performance scores. For example, the people who scored 5/5 on the interview should perform better than the ones who scored 4/5.

They said they would be correlating that at a later time. This is the beginning of what will hopefully be a much larger study.

Even so, an automated quiz that could predict with high confidence that someone would fail an interview would be a big win, saving both interviewers and interviewees a lot of time.