The implications of John Stewart Bell’s theorem are far from settled, and technological advances keep providing us with new ways of looking at it, often by resolving problems — or ‘loopholes’ — associated with the conclusions of existing experiments. A particular pesky one is the ‘freedom-of-choice loophole’. Bell’s tests require randomized measurements, but one could imagine (not without some creativity) that the random measurement choice and the system in question may be somehow subtly correlated, conspiring to fool us. This isn’t totally crazy: so far the randomness has essentially always come from another quantum system.
But now, the members of the BIG Bell Test Collaboration have carried out a Bell test of unprecedented scale, setting this loophole aside for good. Their tactic was to use human free will as the random element of the test. Almost 100,000 participants across the world played a game that generated millions of random choices, which were used to select the measurement setting in 13 different experiments. Bell’s inequality was simultaneously violated everywhere.
https://www.nature.com/articles/s41567-018-0176-y