“The first model fakes the data, then the second model trains on the fake data. Any problems in the synthetic data set are amplified further. Then the second model — based on fake data — is used to treat real patients. This is, of course, all fine.”

Today I want to take David Gerard’s recent post on synthetic data in medical research to talk about synthetic data (which is a fancy term for just making shit up) in general.

As David points out: The idea of just generating data itself is – for most cases – ridiculous.

Often synthetic data is shall supposedly fill in the holes that too little actual data has created. But you are filling those holes with … a mirage. Which for medical cases is patently absurd but even for other cases it’s just … not exactly a rational thing to do? Since when does “we don’t have data so we just fabricated it” pass scientific rigor? It doesn’t? So why does it if I claim “the AI did it”?

I’ve seen this being suggested for helping with social studies (“You can just ask our models instead of actual people“) and other humanities. But the point of studies is not just generating some numbers to put in a paper or a study, the numbers are supposed to be an abstraction, a foundation of an understanding of the actual world and the people in it. By definition you cannot do that with synthetic data. Sure, maybe you can scam a few people in the ad industry with it: “Here we can tell you how people will like the campaign by asking AI” is a product you can sell that means nothing but you might find a few clients to keep the scam going.

But he whole – problematic – narrative of the power of data (“the truth is in the data” and all that data sciency stuff) rests on the data actually representing something real, coming from actual sensors (talking to people being a sensor in that regard as well) in the real world. Even if the data represents something from the real world it is of course still biased and subjective but can at least be meaningfully contextualized: You can look at the way the sensor/the question/ the methodology/etc. work, can analyze its problems and issues. We’ve been doing that for a while and know how to understand data that way.

When you rip away that foundation, those roots in “reality” you also rip away the narrative of data being the supreme, pseudo-objective way of understanding the world.

Which might be one actually useful thing coming from this whole AI bubble (which is just a continuation of the old Big Data bubble): The understanding that the total supremacy of the “data” discourse was always a problematic, neoliberal way of seeing and structuring the world, of legitimizing violence according to the needs of those in power. Data always served those who had the power to give it its structure, who could form the way the data flowed from the world and therefore could form the pathways that that data opened up or closed off.

Data was never your friend or ally.

Medical research means dealing with ethics boards — who keep asking all these pointed questions on how you’re going to use people’s most personal data.

What if we just … fake the data? Sorry, synthesise the data. Feed some real data to machine learning, churn out statistically similar synthetic numbers, then write up this fake data! Just as if you did science!

Remember: it’s not technically data fraud if you list it in your methodology!

Journal articles have pushed the idea of synthetic data for a few decades now. It isn’t actually very popular. But also, they keep pushing it.

You can only dodge the ethics board like this if your institution lets you. Nature spoke to a few institutions who do let researchers use AI-faked data so they don’t have to think about ethics. [Nature]

Here’s the use case for synthetic data:

protecting patient privacy, being more easily able to share data between sites and speeding up research.

The data is literally fake — but they get so many more papers out!

Synthetic data also solves data scarcity — there just isn’t enough real data in the world for all the researchers with papers to write.

So we turn a small real dataset into a huge fake dataset. Then we send the huge fake ethics-laundered datasets around the world! So everyone is using derivatives of the same small original dataset! With any noise in the original treated as the finest A-grade data that tells you things!

Sometimes the synthetic data fans admit this might cause issues: [Nature]

bias amplification, low interpretability, and an absence of robust methods for auditing data quality.

Are you sure the real data set you started with was good? That it doesn’t turn out to be wrong or horribly biased for some reason? Did you actually capture the statistics of the original — or did you oversimplify because you were in a hurry?

None of that matters! You already decided ethics was for dodging!

The other use for synthetic medical data is … training medical machine learning models. [ScienceDirect]

The first model fakes the data, then the second model trains on the fake data. Any problems in the synthetic data set are amplified further. Then the second model — based on fake data — is used to treat real patients. This is, of course, all fine.

The abstract of that paper makes this amazing claim for synthetic data:

unbiased data with sufficient sample size and statistical power.

You’re talking about a fake data set synthesized from a real data set. Even if you’ve duplicated its character completely, you can’t synthesize the statistical power to tell you things about the world which wasn’t in the original. You can’t do a statistical CSI “enhance!”

But sure, you can fake more data to write down a bigger N.

Synthetic data is not widely used in medical research yet because most researchers still actually give a hoot. It’s still at the hype stage — like this effusive bilge in BMJ Evidence Based Medicine in July, about the incredible potential of faking the evidence base for your medicine. Their main use case is: [BMJ EBM]

overcoming technical and regulatory barriers to assembling sufficiently large datasets for modern AI methods is paramount.

At least they’re doing just machine learning, not generative AI. So far.

• Video version

Everything (including piss) in moderation

The post Bluesky Really Doesn’t Want People To Say ‘Rest In Piss’ About Charlie Kirk appeared first on Aftermath.

Are we talking about the same guy, NYT, MSNBC, and Gavin Newsom?

The post OK, But You Do Know You’re Eulogizing Charlie Kirk, Right? appeared first on Aftermath.

Thursday morning, Ezra Klein at the New York Times published a column titled “Charlie Kirk Was Practicing Politics the Right Way.” Klein’s general thesis is that Kirk was willing to talk to anyone, regardless of their beliefs, as evidenced by what he was doing while he was shot, which was debating people on college campuses. Klein is not alone in this take; the overwhelming sentiment from America’s largest media institutions in the immediate aftermath of his death has been to paint Kirk as a mainstream political commentator, someone whose  politics liberals and leftists may not agree with but someone who was open to dialogue and who espoused the virtues of free speech. 

“You can dislike much of what Kirk believed and the following statement is still true: Kirk was practicing politics in exactly the right way. He was showing up to campuses and talking with anyone who would talk to him,” Klein wrote. “He was one of the era’s most effective practitioners of persuasion. When the left thought its hold on the hearts and minds of college students was nearly absolute, Kirk showed up again and again to break it.”

“I envied what he built. A taste for disagreement is a virtue in a democracy. Liberalism could use more of his moxie and fearlessness,” Klein continued.

Kirk is being posthumously celebrated by much of the mainstream press as a noble sparring partner for center-left politicians and pundits. Meanwhile, the very real, very negative, and sometimes violent impacts of his rhetoric and his political projects are being glossed over or ignored entirely. In the New York Times, Kirk was an “energetic” voice who was “critical of gay and transgender rights,” but few of the national pundits have encouraged people to actually go read what Kirk tweeted or listen to what he said on his podcast to millions and millions of people. “Whatever you think of Kirk (I had many disagreements with him, and he with me), when he died he was doing exactly what we ask people to do on campus: Show up. Debate. Talk. Engage peacefully, even when emotions run high,” David French wrote in the Times. “In fact, that’s how he made his name, in debate after debate on campus after campus.”

This does not mean Kirk deserved to die or that political violence is ever justified. What happened to Kirk is horrifying, and we fear deeply for whatever will happen next. But it is undeniable that Kirk was not just a part of the extremely tense, very dangerous national dialogue, he was an accelerationist force whose work to dehumanize LGBTQ+ people and threaten the free speech of professors, teachers, and school board members around the country has directly put the livelihoods and physical safety of many people in danger. We do no one any favors by ignoring this, even in the immediate aftermath of an assassination like this.

Kirk claimed that his Turning Point USA sent “80+ buses full of patriots” to the January 6 insurrection. Turning Point USA has also run a “Professor Watchlist,”and a “School Board Watchlist” for nearly a decade. 

“America’s radical education system has taken a devastating toll on our children,” Kirk said in an intro video posted on these projects’ websites. “From sexualized material in textbooks to teaching CRT and implementing the 1619 Project doctrine, the radical leftist agenda will not stop … The School Board Watch List exposes school districts that host drag queen story hour, teach courses on transgenderism, and implement unsafe gender neutral bathroom policies. The Professor Watch List uncovers the most radical left-wing professors from universities that are known to suppress conservative voices and advance the progressive agenda.”

These websites have been directly tied to harassment and threats against professors and school board members all over the country. Professor Watchlist lists hundreds of professors around the country, many of them Black or trans, and their perceived radical agendas, which include things like supporting gun control, “socialism,” “Antifa,” “abortion,” and acknowledging that trans people exist and racism exists. Trans professors are misgendered on the website, and numerous people who have been listed on it have publicly spoken about receiving death threats and being harassed after being listed on the site.

One professor on the watchlist who 404 Media is granting anonymity for his safety said once he was added to the list, he started receiving anonymous letters in his campus mailbox. “‘You're everything wrong with colleges,’ ‘watch your step, we're watching you’ kind of stuff,” he said, “One anonymous DM on Twitter had a picture of my house and driveway, which was chilling.” His president and provost also received emails attempting to discredit him with “all the allegedly communist and subversive stuff I was up to,” he said. “It was all certainly concerning, but compared to colleagues who are people of color and/or women, I feel like the volume was smaller for me. But it was certainly not a great feeling to experience that stuff. That watchlist fucked up careers and ruined lives.” 

The American Association of University Professors said in an open letter in 2017 that Professor Watchlist “lists names of professors with their institutional affiliations and photographs, thereby making it easy for would-be stalkers and cyberbullies to target them. Individual faculty members who have been included on such lists or singled out elsewhere have been subject to threats of physical violence, including sexual assault, through hundreds of e-mails, calls, and social media postings. Such threatening messages are likely to stifle the free expression of the targeted faculty member; further, the publicity that such cases attract can cause others to self-censor so as to avoid being subjected to similar treatment.” Campus free speech rights group FIRE found that censorship and punishment of professors skyrocketed between 2020 and 2023, in part because of efforts from Professor Watchlist.

Many more professors who Turning Point USA added to their watchlist have spoken out in the past about how being targeted upended their lives, brought years of harassment down on them and their colleagues, and resulted in death threats against them and their loved ones. 

At Arizona State University, a professor on the watchlist was assaulted by two people from Turning Point USA in 2023. 

“Earlier this year, I wrote to Turning Point USA to request that it remove ASU professors from its Professor Watchlist. I did not receive a response,” university president Michael Crow wrote in a statement. “Instead, the incident we’ve all now witnessed on the video shows Turning Point’s refusal to stop dangerous practices that result in both physical and mental harm to ASU faculty members, which they then apparently exploit for fundraising, social media clicks and financial gain.” Crow said the Professor Watchlist resulted in “antisemitic, anti-LGBTQ+ and misogynistic attacks on ASU faculty with whom Turning Point USA and its followers disagree,” and called the organization’s tactics “anti-democratic, anti-free speech and completely contrary” to the spirit of scholarship.  

Kirk’s death is a horrifying moment in our current American nightmare. Kirk’s actions and rhetoric do not justify what happened to him because they cannot be justified. But Kirk was not merely someone who showed up to college campuses and listened. It should not be controversial to plainly state some of the impact of his work.