Great story and it shows what everybody knows but won't say - normal academics are the frauds. Not just mysterious strangers in foreign countries, paper mills, etc. but normal medium or high status academics in prestigious universities in western countries doing the fraud themselves.
You can say it's not their fault - they're forced into it because it's the only way to succeed. But it's still their fault. I remember a case of a politician in China being arrested for corruption and he said that the only way to reach his position was with corruption. Yea he was probably right but he was still corrupt.
For the curious, the laboratory I'm talking about is the Laboratory of Photonics Interfaces[1] at the EPFL in Switzerland ran by Michael Gratzel[2].
I want to stress out that the lab is great, the people in there are extremely hardworking, Gratzel is a great scientist, but at the end of the day research is what it is and stuff like this can slip under both your lab managers and reviewers. I have never ever seen the slightiest indication that lab staff ever encouraged nor tolerated such stuff, but it's easy for it to happen and there's not enough incentives (nor possibility) to review every single experiment.
But the reproducibility problem does exist and the number of scientists tweaking numbers by tiny percentages here and there to make sure they publish is relevant.
I can understand mistakes happening and going unnoticed. But from what that lady told you, it sounds like they have a culture of sweeping mistakes under the rug. Was the affected paper retracted or at least a correction published? If not, they're still benefiting from the fake success which makes it unethical even if it's not technically academic fraud.
You can say it's not their fault - they're forced into it because it's the only way to succeed. But it's still their fault. I remember a case of a politician in China being arrested for corruption and he said that the only way to reach his position was with corruption. Yea he was probably right but he was still corrupt.