If you read the papers, the claims are not very well backed by independent experiment. (The claims are 'replicated' meaning lots of people repeat them!) There's

If you read the papers, the claims are not very well backed by independent experiment. (The claims are 'replicated' meaning lots of people repeat them!) There's little data and what there is is right at the limit of measurement error. Results from different groups conflict wildly. Tech Review is a PR rag, not a scientific publication.