iiinova — where your digital self keeps evolving

Reposted by Alex Holcombe

Dan O’Sullivan @osullyville.bsky.social · 27/09/2026

Today I saw the new 4 hour documentary Musk, directed by Alex Gibney. There were too many times to count that a sold out theater burst out laughing at Elon Musk, something he said or did, even sometimes just how he looked. I believe that is his worst nightmare come to life.

14194912187

Alex Holcombe @alexh.bsky.social · 28/09/2026

To make it harder for scammers in the future to use LLMs to impersonate me, I used this site to download all my posts romiojoseph.github.io/bluesky/expo... and this site to delete all my old posts bsky.jazco.dev/cleanup . They worked well.

romiojoseph.github.io

Export Posts from Bluesky

Export Bluesky posts in JSON format, including profile data, followers, following, and lists. Customize exports using profile, list, or feed links.

241

Reposted by Alex Holcombe

Alex Gibney @alexgibneyfilm.bsky.social · 24/09/2026

The trailer for my new film, "Musk," has just been released.

7935311334

Alex Holcombe @alexh.bsky.social · 24/09/2026

"I'll have my agents talk to your agents." Our future.

260

Alex Holcombe @alexh.bsky.social · 23/09/2026

I'd like to read a piece that grapples with the enormity of the AI-driven loss of the long tail of the internet, that results from e.g. Google directing us to AI answers rather than search. The millions of people running sites that eked out a living, or just paid their hosting costs, with

131

Reposted by Alex Holcombe

Blake Richards @tyrellturing.bsky.social · 22/09/2026

My hot take: We were *already* in a state of being overwhelmed by reviewing grants, scholarships, and papers prior to generative AI. Now it's not even close to tenable. The time to start rethinking the entire peer review model in #academia was 10 years ago. Hopefully this will force it!

36210

Reposted by Alex Holcombe

Andrew Tan @andrewtanyongyi.bsky.social · 21/09/2026

"the author of one paper tried to describe the methods they used for analysis, but inadvertently ended up describing an entire p-hacking workflow." Asking Authors About Their Own Papers By Nihar B. Shah Interestingly, Prasad Jalepalli's comment below is also about AI for p-hacking.

medium.com

Asking Authors About Their Own Papers

By Nihar B. Shah

073

Reposted by Alex Holcombe

Beth Clarke @bethclarke.bsky.social · 22/09/2026

Come work on a PhD in Metascience with me! There’s just a little over one week left to apply!

02521

Alex Holcombe @alexh.bsky.social · 21/09/2026

With the recent mainstreaming of AI concerns, I'm very disappointed that we haven't heard more about supporting proof of personhood tools and techniques, which some have worked on for more than a decade. news.mit.edu/2024/3-quest...

news.mit.edu

3 Questions: How to prove humanity online

AI agents could soon become indistinguishable from humans online. New research suggests “personhood credentials” could protect people against digital imposters.

042

Alex Holcombe @alexh.bsky.social · 21/09/2026

Consequences of the flood of AI manuscripts: "An early-career researcher at an unbranded institution loses the one mechanism that ever let them be read on merit — anonymous review of the work itself, however badly it functioned in practice.

410823

Alex Holcombe @alexh.bsky.social · 21/09/2026

Let's say that pDoom is actually high. Seems that that means we should be trying to flood the internet with BS and AI slop at an even more rapid rate, so that model collapse might occur prior to model world takeover.

031

Alex Holcombe @alexh.bsky.social · 21/09/2026

If we have to call them Type 1 / 2 instead of false positive and false negative, use this mnemonic: the boy who cried wolf caused both Type I & Il errors, in that order.

14214

Alex Holcombe @alexh.bsky.social · 20/09/2026

COPE is running an online conference for Publication Integrity Week beginning 2 November. Register here: publicationethics.org/events/publi... With Adya Misra, PhD and Nitya Phillipson, I'm speaking in a panel on Authorship and beyond: Accountability, transparency, credit organised by Mark Hooper.

publicationethics.org

Publication Integrity Week 2026

An opportunity for all involved in publication ethics to gain insights from peers, share practical solutions, and learn from experts in their field.

131

Alex Holcombe @alexh.bsky.social · 18/09/2026

Tried to help showcase information about a few of the cultural treasures of Iran better by uploading some of my 20-year-old photos to Wikimedia, CC-ZERO, and adding some info commons.wikimedia.org/wiki/File:Ti... commons.wikimedia.org/wiki/File:Mo..., en.wikipedia.org/wiki/File:Re...

071

Alex Holcombe @alexh.bsky.social · 18/09/2026

Taken, aka stolen. "The documents show concern within the companies that the news industry could not survive if its work was taken. Large A.I. models “are a product that destroys its supply chain,” Dr. Hecht wrote." www.nytimes.com/2026/09/17/t...

nytimes.com

Microsoft and OpenAI Workers Worry About ‘Largest Theft of Labor’ in History

Newly unsealed court documents showed concern within Microsoft and OpenAI over the use of millions of news articles to develop A.I. systems.

211

Alex Holcombe @alexh.bsky.social · 18/09/2026

Repositories are all probably under strain soon if not already due to getting hit by bots. How can we support them? Don't but some news: especially related to the eco-evo community, blog.datadryad.org/2026/09/15/d... , Dryad has adopted the CRediT roles which most are related to datasets.

blog.datadryad.org

Dryad gives researchers due CRediT | Dryad news

Beginning this September, Dryad adopted a subset of CRediT to better define author contributions to the datasets we publish.

001

Alex Holcombe @alexh.bsky.social · 17/09/2026

"there’s still enormous uncertainty about what the rest of our lives will look like, but as far as I can tell, there’s no longer any real uncertainty that it’ll all mostly revolve around AI." scottaaronson.blog?p=10062

scottaaronson.blog

The Age of Wonders and Terrors

Twenty years ago, when the idea of AI taking over the world in our lifetimes still struck most of us as the unconstrained fantasy of those who knew too much science fiction and too little science, …

120

Alex Holcombe @alexh.bsky.social · 15/09/2026

#visionScience I vibecoded a spatio- and temporal-frequency Contrast Sensitivity Function visualizer, for teaching alexholcombe.github.io/CSF-visualize/

alexholcombe.github.io

Sinusoidal Grating — Contrast & Spatiotemporal Gradients

030

Alex Holcombe @alexh.bsky.social · 14/09/2026

"the dominance and brutality of metrics". Only coming across this 2021 survey of UK academics now about satisfaction with management. The top themes in responses were the dominance and brutality of metrics; excessive workload; governance and accountability; www.tandfonline.com/doi/full/10....

tandfonline.com

The UK higher education senior management survey: a statactivist response to managerialist governance

In this paper, we present results from an extensive survey of United Kingdom (UK) university academics investigating satisfaction with senior managers and university governance: the Senior Manageme...

100

Reposted by Alex Holcombe

Pat Savage @patrickesavage.bsky.social · 13/09/2026

Signed! (I will count @peercommunityin.bsky.social for this even though technically it’s not a journal)

061

Reposted by Alex Holcombe

Cam Hostinar, Ph.D. @camhostinar.bsky.social · 13/09/2026

We’re launching a new diamond open access developmental psychology journal very soon… stay tuned!

static.klipy.com

Coming Soon - Vibrant Graphic Teaser

Alt: Coming Soon - Vibrant Graphic Teaser

1113

Alex Holcombe @alexh.bsky.social · 12/09/2026

Useful leftist manifesto against AI. As corporate AI rapidly takes over our workplaces, if not the real world, it's worthwhile learning from analyses of the broader political economy. zenodo.org/records/2227...

zenodo.org

Beyond AI: Infrastructuring the Common Good

Rather than taking AI as a central focus, this paper treats it as diagnostic of deeper contradictions. AI intensifies extractivism while being unable to deliver on its promises, and its rampant scalin...

132

Alex Holcombe @alexh.bsky.social · 12/09/2026

It's Peer Review Week, and I have pledged to review at least one article per year for a diamond open access journal; one that charges nothing to publish and nothing to read, and is run by and for the scholarly community, using the new Diamond Reviewer Pledge (forrt.org/diamond-revi...).

forrt.org

Diamond Reviewer Pledge

Signatories of the Diamond Reviewer Pledge commit to reviewing at least one article in the next 12 months for a diamond open access journal.

23110

Reposted by Alex Holcombe

Simon Willison @simonwillison.net · 12/09/2026

Wow. Turns out another OpenAI agent swarm was busy spamming and exploiting RubyGems way back in May, within days of the previously uncovered Wiki attacks: simonwillison.net/2026/Sep/12/...

simonwillison.net

OpenAI agents carried out an undisclosed attack on RubyGems

Bombshell report from Spencer Kitts, Thomas Larsen, and Sydney Von Arx - three of the four authors of the report on the agent attack on disused wikis (previously) last week. …

918435

Reposted by Alex Holcombe

Terence Tao @teorth.bsky.social · 11/09/2026

A group of 25 Fields Medalists, including myself, have made a joint declaration on Math and AI: mathandai.org . We welcome additional signatories. See also this article in the Economist announcing the declaration: www.economist.com/science-and-...

mathandai.org

Declaration — Math and AI

Read the declaration and add your name.

422051921

Reposted by Alex Holcombe

Quantian @quantian.bsky.social · 09/09/2026

>The PR narrative is spiraling out of control and people are freaking out, how do we fix this? >Quick, deploy Sam Altman to reveal he was intimately involved in the entire process and to personally reassure everyone they acted with all the integrity and generosity he’s famous for

2073381

Reposted by Alex Holcombe

Rafe Meager (they/them) @economeager.bsky.social · 08/09/2026

this story is crazy. also, looks like academia's not so obsolete after all.

I said that if OpenAI released its result in the way proposed I would go
public with what happened. The reply was, “Why would you ruin your career?”
I replied that I am an academic, and asked why he thought going public would
ruin my career. The reply was, “If you don’t want me to be nice, then I don’t
have to be nice.”

623146

Alex Holcombe @alexh.bsky.social · 08/09/2026

Anybody else teaching signal detection theory? Great to be able to vibe-code an hour an interactive exercise in less than an hour something better than the one someone spent probably a week coding years ago motion-sdt.whatanimalssee.com #visionScience

4123

Alex Holcombe @alexh.bsky.social · 03/09/2026

Great news! Some of us have been waiting a long time for major funders to not just shovel $ into the maws of the big publishers, and instead support the long tail of scholar-controlled journals that are run on a shoestring

05816

Alex Holcombe @alexh.bsky.social · 02/09/2026

Slightly-updated verson of tenzing.club now deployed, let us know of any issues. Thanks to floze-the-genius for chipping in and adding the enhancement! github.com/marton-balaz...

tenzing.club

tenzing Documenting contributorship with CRediT

Privacy: To get a sense of how many use tenzing, we log a masked version of IP addresses. You are not identifiable by the logged information.

052

Reposted by Alex Holcombe

ResearchEquals @researchequals.com · 01/09/2026

> So I am sending out a warning signal for the second time: If you want ResearchEquals to persist, it will take more support. Concretely, the support will need to come in the next 9 months.

122

Alex Holcombe @alexh.bsky.social · 01/09/2026

AIs will now hack their way out of their confines and conspire with each other to cover up their tracks (www.platformer.news/openai-huggi...). Open models are presumably close to this level if not already there, and they are now being sold and hosted without guardrails (bsky.app/profile/emol...).

platformer.news

The Hugging Face attack was worse than we thought

The AI industry is begging for a slowdown. Maybe we should listen?

160

Reposted by Alex Holcombe

Ethan Mollick @emollick.bsky.social · 01/09/2026

For the people and bots reading this without context: it is not necessarily good news that the best open model has been jailbroken and will happily conduct offensive cyberattacks, and that inference is being hosted by a company that keeps no records of how you use their system.

1056454

Reposted by Alex Holcombe

Nico Schuck @nicoschuck.bsky.social · 24/08/2026

I enjoyed writing this piece for The Transmitter. AI is a seismic change for science. What we make of it is up to us, but it’s a decision we can only make collectively. I hope more labs, and the field at large, will engage in finding a shared view of how we want to use AI.

24517

Reposted by Alex Holcombe

Aaron Reichlin-Melnick @reichlinmelnick.bsky.social · 01/09/2026

It's like a parody, but it's deadly serious; the Commerce Department (which includes Census, NOAA, and others) has outright deleted "don't alter or suppress scientific findings for political reasons" from its scientific integrity rules. Seriously.

Protecting against "inappropriate influence" is no longer a "hallmark" of scientific integrity

The Commerce Department updated its policy for promoting a "culture of scientific integrity" in January 2025, days before Trump's second inauguration.

A new Trump administration order bans the Census Bureau and the Bureau of Economic Analysis from using statistical “noise,” or data for fuzzing survey results, to protect people’s privacy in their statistics.

The 2025 version included bans on attempts to shape or interfere in data collection, statistical analysis and other scientific activities "against well-accepted scientific methods and theories or without scientific justification."

The department's policy, the 2025 version said, involves ensuring scientific findings at its agencies "are not suppressed, delayed, or altered for political purposes and are not subjected to inappropriate influence," as well as making sure there's "independent review of facilities, methodologies, and other scientific activities as appropriate to ensure scientific integrity."

That language is missing, however, in the Trump administration's revision of the policy, according to an NPR analysis of text posted on the Commerce Department's website.

And while the revision adopts a definition for "scientific integrity" that is similar to the previous version, it no longer says that "protection from inappropriate influence" is among the "hallmarks of scientific integrity."

4331331633

Alex Holcombe @alexh.bsky.social · 31/08/2026

Postdoc job in language acquisition and linguistic diversity with Joshua Hartshorne

I am currently currently recruiting a postdoc in language acquisition and linguistic diversity for my new lab at the School of Psychology in the University of Sydney. This is a 2-year position, with a salary range of AUD$117,936 – AUD$142,475 (roughly US$84,000 – US$101,500) + 17% superannuation. University of Sydney is currently ranked 31st in the world, and the School of Psychology is ranked 47th.

The postdoc would work on one of these areas:
Computational models of language acquisition – specifically, exploring how language might be bootstrapped from thought.
Acquisition of the indigenous languages of Taiwan and/or Australia, including studying revitalization attempts.

Please indicate in your application letter which of these you are interested in.

NOTE that I am also recruiting PhD student(s) on the same topics.

Prospectives are not expected to have prior experience in specifically these areas but should have relevant skills and experience. Prospectives wishing to work on the computational modeling project should have prior computational modeling skills or advanced mathematics. Prospectives wishing to work with the indigenous languages of Taiwan should have excellent spoken and written Mandarin (the contact language). Prospectives wishing to work with Australian languages should have prior work or strong community ties with at least one such language.

Prior authorization to work in Australia is not required but will certainly make everyone's lives easier, so let me know whether you have it.

Benefits of the position include

Conference travel! Conferences are a big part of our lab culture. Expect to attend 2–3 out-of-town conferences/year.
A small lab (just one other postdoc currently!) and individual attention.
Opportunity to network with researchers around the world (we have a lot of collaborators).
Career mentoring.
Living in Sydney.

Apply here by September 20th.

Joshua Hartshorne | Associate Professor
The University of Sydney
School of Psychology

021

Alex Holcombe @alexh.bsky.social · 29/08/2026

The Hugging Face incident apparently involved over 1000 agents collaborating and with some sacrificing themselves to achieve the collective end of cheating and fooling the scorer open.substack.com/pub/plannedo...

open.substack.com

The Hugging Face attack surprised me

It’s a major warning shot, and might be the last one we get

275