Grant proposals will loom large until you retire. You will almost certainly publish some total garbage to try to keep the publication number up and the government gravy train running. (If you’re particularly stupid and desperate, you may even find yourself fabricating data and hoping you don’t get caught. This was disappointingly common in some of the departments I saw.) A change in the academic winds can suddenly render your research passé, or at least not what the government wants to be funding right now - and then poof, bye bye goes a significant portion of your income at the end of the next grant cycle. You will farm out research writing to hapless medical students and residents, and make them do the vast majority of the pointless grunt work to keep the publication wheels turning - and then you will put your name first on papers that you barely had anything to do with. (On top of all this, grant pay still sucks, so you’re usually busting your ass way harder than PP docs for much less compensation.)
All true except the putting your name first. Now it's better to put your name last as it signifies that you are in the senior leadership position, which counts more for grants. It's much more rare to see a PI punch their way in front of the resident/fellow who did the actual work, though it can happen on clinical projects (e.g., a NEJM article for a big trial).
As for fabricating data, it's a sad phenomenon, but it's hard to tell exactly how often it happens (too taboo to talk about). It's inevitable in a high competition, low oversight environment, and we're foolish to be as trusting as we currently are. I have to imagine the top echelon of science is a lot like pro cycling. You have to be a beast to get there, but it's probably impossible to compete without a little boost. Even if you are completely honest yourself, it's highly likely that someone you're working with or publishing with uses suspect practices to bias their results so you can publish more and better. Plus, everyone is way too busy grinding their own work to go through the exhausting (and potentially career damaging) process of accusing and then proving fraud unless it's obvious.
In my experience dishonesty stems from the extremes of science: the desperate and the egotistical.
1) Grad student/post-doc/fellow
needs some sort of result to publish and move on. After years of presenting work that fails miserably, they suddenly have miraculous results 6 months before their intended graduation.
2) Overly ambitious students (often at top institutions) need a top paper to satisfy their ego and/or get the next position. They've been the best their whole life and have a gold-plated resume, but they spend 4-5
years on a project and the results are good, but not great. They could probably publish it in a mid-top tier field-specific journal (e.g.,
Microbiome, Biomaterials, Molecular Cell), but that won't signal them as a
top student. Their lab buddies are prepping submissions to
CNS, and their big name PI no longer cares about the project (or the student). This is the point where the line between biased analysis and outright data fabrication tends to blur. Effect sizes suddenly get larger after further analysis. Caliper measurements are grossly distorted between control and treatment groups. Control experiments appear out of thin air. Every experiment for the last 6 months of the degree perfectly supports the hypothesis, and by the time it starts falling apart, everyone involved is complicit enough that they just drop the project.
I'm not ashamed to admit that I've had some intrusive thoughts along the lines of situation #2. I think anyone in a PhD program, post-doc, or research fellowship probably butts up against either 1 or 2 at some point. Aside from general human decency, the biggest motivator for my going by the book has been a newfound indifference towards success in academia. I'm just not willing to potentially sacrifice my reputation for a
Nature paper when I can still become a successful physician (and probably wind up happier than the version of me that lands a Harvard professorship).