Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Fraud, so much fraud

science.org

Post reply on HN
Print view View original

721–730 of 868 posts

Re: Fraud, so much fraud

#721

Post by jboggan »

mike_hearn wrote: post #402
jboggan wrote: post #144

When I was in my doctoral program I had some pretty promising early results applying network analysis to metabolic networks. My lab boss/PI was happy to advertise my work and scheduled a cross-departmental talk to present my research in front of ~100 professors or so. While I was making a last-minute slide for my presentation I realized one chart looked a little off and I started looking into the raw data. I soon rea…

Coding errors are a really common source of fraud unfortunately. You did the right thing but the vast majority don't. Given a choice between admitting the grant money was wasted, the exciting finding isn't real, everyone who cited your work should retract their papers or just covering it up, the pressure to do the latter is enormous. During COVID I talked to a guy who used to do computational epidemiology. He came to…

>the COVID model that drove lockdowns in the UK and USA was nothing but bugs

I would love a source but I believe this given my experience with coding standards in the computational biology space, especially given that some of my own previous work and teaching touched on those models, and I couldn't believe any of the models that were publicized because they were at odds with what I thought the scientific consensus was about spread and containment models (96 hours after patient 0 it's pointless to attempt to restrict movement).

Re: Fraud, so much fraud

#722

Post by fluidcruft »

dr_kiszonka wrote: post #421
hprotagonist wrote: post #2

> It seems like a strange thing to take someone with a long and respected career and subject them to what would essentially be a Western blot and photomicrograph audit before offering them a big position. This is absolutely something that we should routinely be doing, though.

I agree. To their defense, he had a few hundred papers published before joining, according to PubMed, and was a leader in his field: https://pubmed.ncbi.nlm.nih.gov/?term=Masliah%20E&filter=dat... My concern is that with AI getting better and easier to use (e.g., in Photoshop) fraud will be extremely hard to detect. We might need to re-think how research is done and results verified.

Maybe we need instruments that sign results cryptographically and use a blockchain mechanism to establish provenance. We should have cameras that can establish that published images have not been modified (or at least provide raw and adjusted pairs--digital radiology has the concept of a "presentation state" that I think could work).

In theory at least research should be auditable to a lab notebook. The problem with photos and such is you can't tell if it was modified before it was pasted to the page and large datasets just can't be put in to paper. And electronic notebooks I've used tend to be even more annoying than paper (too rigidly formatted and not adaptive to workflow optimization but it's difficult to explain).

Anyway those sorts of things that establish provenance should also protect against deep learning. You may be able to create fake data, but can you deep fake data that was signed by Nikon device #XYZ with cryptographicly confirmed hashes published to a blockchain 3 years ago at the time the data was generated?

Re: Fraud, so much fraud

#723

Post by Vecr »

themanmaran wrote: post #68
layer8 wrote: post #53

Earlier quoted context omitted.

LLMs might find some specific indications of possible fraud, but then fraudsters would just learn to avoid those. LLMs won’t be able to detect when a study or experiment isn’t reproducible.

Of course, but increasing the difficulty of committing fraud is still good. Fraudsters learn to bypass captchas as well, but they still block a ton of bad traffic.

Won't the scientist use some relatively secure/private model to fraud-check their own work before submitting? If it catches something, they would just improve the fraud.

Re: Fraud, so much fraud

#724

Post by rurban »

He probably refers to Sylvain Lesné's previously detected Alzheimer fraud, a hugely influencal doctored paper. And now the #1 Alzheimer researcher Eliezer Masliah is also a fraudster.

Re: Fraud, so much fraud

#725

Post by rebanevapustus »

I was the victim of a pretty bizarre super in-your-face academic theft.

Someone snooped a half-finished draft of mine off GitHub and...actually got it published in a real journal: https://forbetterscience.com/2024/05/29/who-are-you-matthew-...

In spite of having a full commit log (with GitHub verified commits!!!) of both the code AND the paper, both arxiv and the journal didn't seem to care or bother at all.

Anyhow, I highly recommend reading the for better science blog. It's incredible how rampant fraud truly is. This applies to multiple nobel prize winners as well. It's nuts.

Re: Fraud, so much fraud

#726

Post by 2OEH8eoCRo0 »

molave wrote: post #338
AndrewKemendo wrote: post #333

If I have learned anything over 40 years, is that the number of people who actually live in a way consistent with hypothesis testing, data collection, evidence evaluation framework required to have scientific confidence in future action or even claims is effectively zero That includes people who consider themselves professional scientists, PhD‘s authors, leaders etc. The only people I know who live “scientifically” c…

It's disheartening to think that the virtues you are told to have as a kid are considered "weak sauce" once you are an adult.

It's a dilemma- do you want to be virtuous or do you want to maximize your money? I get a sense around here that only the law matters (morals be damned) and we do whatever work pays best.

Re: Fraud, so much fraud

#727

Post by fluidcruft »

asdf123qweasd wrote: post #326
lupire wrote: post #318

Earlier quoted context omitted.

How so? The fraction of academic science that is applied to anything, anywhere, with clearly identifiable impact is both a tiny fraction of academic science, and also often detrimental quality of impact.

Thats not true entirely. There is research, with huge impacts and the money leveraging- keeps it brutally honest. Nothing makes it out of a lab and into a fab, without certainty of the method working at least in lab conditions reproduceable. There are many billions, but not that many billions.

> nothing makes it out of a lab and into a fab, without certainty of the method working at least in lab conditions reproducible.

The article describes multiple full clinical drug trials both completed (inconclusive==can't prove harm, effect is so small that benefit cannot be excluded) and ongoing that are fundamentally built on the fabricated results which literally do represent many billions of private investment.

Ultimately research falsification is a con game, and you seem to have faith in something magical about "money leveraging" that smokes out cons. I do think shit eventually hits the fan in the market since reality affects the market, but it's not because the market is more rigorous than "science". Ultimately, investors are not experts and the are listening to the same people who do "peer review" and didn't notice (or ignored) the fraud in the first place.

Re: Fraud, so much fraud

#728

Post by fluidcruft »

rebanevapustus wrote: post #725

I was the victim of a pretty bizarre super in-your-face academic theft. Someone snooped a half-finished draft of mine off GitHub and...actually got it published in a real journal: https://forbetterscience.com/2024/05/29/who-are-you-matthew-... In spite of having a full commit log (with GitHub verified commits!!!) of both the code AND the paper, both arxiv and the journal didn't seem to care or bother at all. Anyhow,…

This guy sounds like someone who would be fun for Javier Leiva's PRETEND podcast [0]. You should reach out to Javier.

[0] https://pretendradio.org/

Re: Fraud, so much fraud

#729

Post by AndrewKemendo »

physPop wrote: post #542
AndrewKemendo wrote: post #333

If I have learned anything over 40 years, is that the number of people who actually live in a way consistent with hypothesis testing, data collection, evidence evaluation framework required to have scientific confidence in future action or even claims is effectively zero That includes people who consider themselves professional scientists, PhD‘s authors, leaders etc. The only people I know who live “scientifically” c…

I would just like to point out the irony of claiming that people live in a way inconsistent with scientific rigour, based solely on personal experience.

I think you’re suggesting that I’m making a conclusion without sufficient evidence - hence the “irony”

Recall I’m discussing how people live, namely that they don’t live based on their own claims as to how to live. You’d have to evaluate my behaviors to derive if my claim is ironic.

However I’m Happy to provide that epistemological chain if requested

Re: Fraud, so much fraud

#730

Post by ellyagg »

jprete wrote: post #705
Aerroon wrote: post #535

Earlier quoted context omitted.

I don't think that's really going to work. People won't list all their incentives, because some of them are implicit and others are embarrassing or "creepy". Others will absolutely judge you for what incentivizes your actions, therefore hiding them is the status quo. If you say that your incentive for working out is to look good and be popular with the ladies then people will judge you for it, even if it's exactly th…

At this point I'm so used to this that I mentally translate "for health" into "to look good for the opposite sex".

It’s funny how those incentives are so well aligned.
Post reply on HN