iiinova — where your digital self keeps evolving

Hey, ProPublica reporter Jake Pearson here. I’m taking over the account to tell you about my new story. For months, I posed as a problem gambler on DraftKings to test the company's "responsible gaming" safeguards. Spoiler alert: They made me a VIP — and I lost $10K. Here's how it all went down. 🧵

Selfie taken by ProPublica reporter Jake Pearson, a man wearing a backwards baseball cap and an orange T-shirt.

19595903526

Reposted by Chris Jackson

John Stott @jpsastro.bsky.social · 24/09/2026

Had a UKRI funded grant proposal rejected at the "AI triage stage". "This AI-based triage was used only to identify approximately 50% of the strongest proposals to take forward to full human review." 🧪 #academicsky

I am writing to let you know that, unfortunately, your application was not successful.

Review process

As part of the initial triage of the 179 proposals submitted to this call, each was assessed using an AI-based review. Every proposal was read independently by several different AI models against the same seven criteria used to shape this call (including cyber security relevance, research quality and novelty, importance of the problem addressed, feasibility of the project plan, likely outputs and impact, value for money, and responsible research practice), with each model asked to give a score and a written justification with supporting quotations from the text. To guard against any one model's idiosyncrasies, we also ran a second, independent process in which different AI models debated the merits and weaknesses of each proposal before reaching a judgement, again against the same criteria. Scores from both processes were combined to produce an overall ranking.

This AI-based triage was used only to identify approximately 50% of the strongest proposals to take forward to full human review. Projects selected for funding were drawn from those that progressed to the full human review stage. We recognise that AI-assisted assessment is a new and evolving part of the review process, and we are continuing to evaluate how well it aligns with the judgements of human reviewers on this call. We will soon produce a document detailing our method, so that others can build upon it and improve it.

Feedback on your application

Your application was not selected to progress beyond the AI triage stage. To provide transparency on the outcome of this assessment, we are sharing below the scores (out of 5) from each of the two AI review processes described above, together with the mean score across the two processes.

40414220

Chris Jackson @cajackson.org · 24/09/2026

Do grayscale images before ML even exist for this kind of superresolution microscopy? If they were using one of the subcelluar things as evidence for something I'd be more worried, but they aren't

000

Chris Jackson @cajackson.org · 24/09/2026

Ehh I'm reluctant to call "we threw color on this but just to make it pretty, not for science" image manipulation Lotta colorized or optical stained DIC images that are pretty much the same thing

100

Chris Jackson @cajackson.org · 23/09/2026

Yeah that's why I say 'looks like', all the splitters are oriented wrong I'm pretty sure I know what their physical setup is based on the diagram; the next step is probably some insane superresolution blackbox ML signal processor, that's the voodoo part

110

Chris Jackson @cajackson.org · 23/09/2026

That part looks like a standard confocal setup Don't ask me about the superresolution parts, every time it sounds like voodoo to me

100

Chris Jackson @cajackson.org · 23/09/2026

I don't think those are cells, it looks like colorized DIC to me. Most of that stuff is probably intracellular structure.

150

Chris Jackson @cajackson.org · 22/09/2026

This exact thing is why I just started setting rasterized=True for everything in matplotlib

000

Chris Jackson @cajackson.org · 21/09/2026

time to fight fire with fire my startup has a model that can tell you which embryo is smartest (+28 iq points) yes its a rng no i wont be around in 20 years to give you a refund

010

Chris Jackson @cajackson.org · 21/09/2026

There have been several Very NYC Specific conversations happening that are not lined up with how things are done in Not NYC Feels good, like we're back to 2015 twitter

010

Chris Jackson @cajackson.org · 21/09/2026

'we should add G&T programs to underperforming schools' is a NYC opinion that starts a fight

020

Chris Jackson @cajackson.org · 21/09/2026

You may or may not be right about that but NYC has a 'successful' G&T tracking & specialized school system, and if you want to make changes to it a very large number of very angry people will tell you how much they dont like you

120

Chris Jackson @cajackson.org · 21/09/2026

You dont really get tracked within a school you get tracked by going to a different school

110

Chris Jackson @cajackson.org · 21/09/2026

NYC also does tracking but it's done at the school level

110

Chris Jackson @cajackson.org · 21/09/2026

yes, the plot is awful, but in fairness you also have to consider that the writing is awful too

010

Chris Jackson @cajackson.org · 21/09/2026

fair reaction to terry goodkind

110

Reposted by Chris Jackson

André Marques @amarques.bsky.social · 16/09/2026

Out today in Nature: sex without crossovers. Rhynchospora tenuis makes pollen, fertilises and sets seed, yet 0 crossovers in 10,997 pollen nuclei. Chromosome drive and ~87% seed abortion rebuild the mother's genotype, so the offspring are clones. doi.org/10.1038/s415...

916369

Reposted by Chris Jackson

The Onion @theonion.com · 16/09/2026

Gas Station Price Sign Using Scientific Notation

273655744

Chris Jackson @cajackson.org · 16/09/2026

Yeah, I've published and reviewed there in the past but I just flat out don't understand how it works now and don't have the energy to figure it out

030

Chris Jackson @cajackson.org · 16/09/2026

Temu Nature Science-ish reports Trends in Things That Have Cells

020

Chris Jackson @cajackson.org · 14/09/2026

oh no you can't get me again with 'the jets are good (n=1 game)' I have been on this ride before

000

Reposted by Chris Jackson

Michael Hendricks 🇨🇦 @michaelhendricks.bsky.social · 13/09/2026

Thread. Apologies in advance, I've been grinding away for weeks on a grant with maybe a 10% success rate. Canadian researchers work in buildings held together with duct tape, run labs on shoestrings, and spend more and more time competing for lower and lower success rate operating programs.

26722

Reposted by Chris Jackson

Matt Bors @mattbors.bsky.social · 14/09/2026

It was exactly ten years ago today, on my 33rd birthday, that I posted this comic online.

484297717660

Chris Jackson @cajackson.org · 11/09/2026

Oh sorry it's from contextlib import suppress Thats what I get for posting python with a phone

011

Chris Jackson @cajackson.org · 11/09/2026

I think this works as a one liner, but I just write the except pass block from contextmanager import suppress with suppress(KeyError): return d[k]

100

Chris Jackson @cajackson.org · 11/09/2026

I would probably do try: return d[k] except KeyError: pass # other stuff to do

100

Reposted by Chris Jackson

Nick Stroustrup @nstroustrup.bsky.social · 05/09/2026

Did you know that the cells of old animals contain much less mRNA than the cells of young ones? Let me share with you some results from an ongoing project in the lab, where we are finding that a progressive depletion of total mRNA abundance is a crucial part of aging 1/22

620983

Reposted by Chris Jackson

Data Colada @datacolada.bsky.social · 02/09/2026

Colada[139] Artificial Deadlines Part 2 Evidence of fraud in the 25 year old gradesheet behind the key finding in an influential paper on deadlines and procrastination. datacolada.org/139

816857

Reposted by Chris Jackson

Boston Review @bostonreview.bsky.social · 27/08/2026

Biomedical scientist (and former editor-in-chief of @science.org) @jeremymberg.bsky.social defends undirected science and notes, “It is hard for scientists to support a party whose members seem to think that bringing a snowball to the floor of the Senate refutes the concept of climate change.”

bostonreview.net

In Defense of Undirected Science

Jeremy Berg responds to Zeynep Pamuk in our forum, “How to Save Science.”

1118

Chris Jackson @cajackson.org · 28/08/2026

Right wing media is going all in on anti-canada, it's a right wing US position moving forward The new york post ran like a week of articles about the Maple Leaf Menace Real 'didnt get the joke in Canadian Bacon' stuff, but it's happening

080

Reposted by Chris Jackson

Fietser Brandon @americanfietser.bsky.social · 27/08/2026

The American mind could never comprehend such an enabling and accommodating system of transportation. 📍Finland 🇫🇮

922535

Chris Jackson @cajackson.org · 26/08/2026

Don't think this guy would sign a letter for me if I got fired for teaching about climate change tbh

Throughout this time, Kalkstein tried to get a letter to Defense Secretary Pete Hegseth through his chain of command, but Sussman, who shared the letter with Stars and Stripes, said it never left campus. The professor argued that exaggerated positions of climate change made under President Joe Biden politicized the field, “making it difficult to separate fact from fiction.”

3110

Chris Jackson @cajackson.org · 24/08/2026

Is there any way to do that without driving up the price of refined products accordingly in Canada?

100

Reposted by Chris Jackson

Zack Labe @zacklabe.com · 23/08/2026

🚨 Preliminary daily data shows we just broke the all-time highest global sea surface temperature in this record... Download high-resolution graphic at zacklabe.com/climate-chan.... This data (extrapolar ocean) is provided by @copernicusecmwf.bsky.social ERA5. 🌊

Graph showing daily global mean sea surface temperatures for 2026. This is also compared to the daily data from the 1981-2010 average and 1991-2020 average. Annotations are also shown to the current year with the other years from 1940 to 2026. There is a seasonal cycle. There is a long-term warming trend. 2023 and 2024 are also indicated, which were the warmest years on record. Data is from ECMWF ERA5 reanalysis and updated through 21 August 2026.

14456257

Reposted by Chris Jackson

Jeremy Berg @jeremymberg.bsky.social · 21/08/2026

Let the shit show begin (actually continue)... www.nih.gov/news-events/... No NIH experience at all as far as I can tell. No significant management experience. Vaccine injury trial expert witness.

nih.gov

Pediatric neurologist Dr. John Gaitanis to lead NIH’s Eunice Kennedy Shriver National Institute of Child Health and Human Development

Dr. Gaitanis is set to begin his role on August 23, 2026.

13288130

Reposted by Chris Jackson

Grace Lindsay @neurograce.bsky.social · 20/08/2026

Teaching "AI use" in college (not related to teaching how AI is built) is silly because 1.) the tools are meant to be user friendly with minimal training and 2.) they change so often your education would be good for about a semester

3203

Reposted by Chris Jackson

Dhruv Mehrotra @dmehro.bsky.social · 19/08/2026

NEW: Flock has said for years its cameras don’t identify or track people. We obtained the code for its police AI and rebuilt parts of it. The tool can find witnesses, surface associates, search people by race or physical description, and flag vehicles by how they move. W/ @dell.bsky.social

wired.com

Flock Has a Powerful New AI Tool for Police. We Got Its Code

Flock’s surveillance cameras have already sparked outrage. WIRED reconstructed its next-generation AI system, already in use by some police, to confirm it goes much further than tracking license plate...

141580784

Chris Jackson @cajackson.org · 19/08/2026

Moderna doesn't sell chemo drugs so they don't give a shit about how profitable they are They do sell mrna therapeutics and if they're profitable, they'll sell them

1240