That was Super-K, the same detector she's talking about in the video. The incident occurred in 2001 and set back the start of the data taking for them quite a bit, they were able to begin data taking by re-distributing the unaffected photo-detectors to cover the gap where the imploded detectors were. Eventually (2005/2006) they replaced all the destroyed photo-detectors and took data from then on out with a full-suite of sensors. Following this incident all Japanese experiments with these types of photo-detectors take the risk of implosion seriously and have mitigation built in to the design. Super-K has been running continously since then with some interruptions for upgrades & maintenance, and is still taking data today (as far as I'm aware).

Construction is underway on the next version of the experiment "Hyper-Kamiokdande" which is similar in design but significantly bigger. If I recall correctly Hyper-K will be two 200 kilo-tonne detectors, compared to Super-K which is a measly 50 kilo-tonne detector.