mindcow 🐮

Opinions not my own, and no, I don't do art commissions.

[Richard Dawkins @RichardDawkins]

I am bewildered, and not very impressed, by the visceral, naked, irrational hatred spat out by some defenders of Jason Arday. If a Cambridge professor is accused of being an unqualified charlatan, the accusation might be racially motivated. On the other hand it might not, depending on the evidence. The correct question to ask is not, ”What is the colour of his skin” but “Is it in fact true that he is an unqualified charlatan?” Please examine the evidence before leaping to the assumption of racism.
As for the idea that journalists “piled in on him” and “hounded him to his death”, most attacks were against Cambridge University. Jason himself was widely regarded as an unfortunate victim of foolish promotion way beyond his ability to cope. In appointing him to a professorship for which he was manifestly unqualified – in ludicrously describing him as “the best in the world” – certain senior members of the university showed a level of patronising condescension towards black people that could fairly be described as racism, while at the same time making him tragically vulnerable to such attacks as came his way.

[Maarten Boudry @mboudry]

How to fight "racism" in perpetuity:

- Quietly bring the bar down to the floor for a Black candidate, because you don't trust them to clear the real one.
- Elevate them to the highest level. Celebrate them. Fawn over them.
- Intimidate anyone who dares to ask questions and accuse them of racism. Because why are Black academics being singled out?
- When the whole thing inevitably collapses, use the collapse itself as proof that you need to do it all over again, only harder.
- Rinse and repeat.

Now who's the real "racist" here?

[Daniel Friedman @DanFriedman81]

This is a straw-man argument. People who oppose DEI do not contend that "Black people lack ability."

Some black people have ability. Some do not. Some white people have ability. Some do not. Every year there are black students who achieve top academic scores and other achievements, and they have every bit as much talent as members of other groups who achieve similar things on a level playing field.

HOWEVER, when you implement any kind of ability-based, objective, competitive screening mechanism -- an IQ test, the SAT, the GMAT, the LSAT, the MCAT, the Sergeant's Examination, the Wonderlic Contemporary Cognitive Ability Test, the Armed Services Vocational Aptitude Battery, the SHSAT, or whatever -- the top cut of performers will have a significantly lower percentage of black people than the overall population set, and if you plot the performance of each group on a distribution, the set describing the black population will have a lower median and therefore fewer points the further you get along right tail.

The top 50 or so colleges would be about 1-2% black if they admitted on academics without considering race or using metrics (like preferring low-income ZIP codes or favoring predominately minority schools) that function as proxies for race.

When schools use racial considerations or proxies that function as racial considerations to achieve "equity" proportions, there is a sizable credential and preparedness gap, that has noticeable impacts on college performance.

The 1-2% they would have admitted under a race-blind standard are competitive with the other students admitted under the same standard. The students admitted under DEI with academic indicators that would have guaranteed rejection for members of other groups are not prepared to compete.

For example, in a 2004 study of first year law students at selective US law schools, where course grades are based on anonymously submitted exams graded against each other on a curve, UCLA law professor Richard Sander found that more than half of black law students scored in the bottom tenth of the class, and only 8% were in the top half. Since that study came out many top law schools have watered down their curves or eliminated grading entirely.

The premise that there is "systemic racism" in the tests is extremely hard to support, unless you believe, as Ibram X. Kendi argues, that racism is proven by the mere existence of a disparity. Numerous tests developed for numerous purposes by numerous designers have all reached the same conclusion. The makers of these tests have tried to alter the tests to eliminate the disparities and no one has ever been able to design an academic aptitude test that doesn't have a race disparity. Moreover, there is no evidence that the tests are a worse predictor of future academic performance for black students than they are for other groups -- the defect is not in the test.

The differences measured by the tests do not seem to be corrected by access to educational resources and interventions -- disparities in graduate and professional school examinations administered to graduating college seniors are similar to disparities in college entrance exams. Students who are admitted to college under affirmative action require affirmative action again when they apply to graduate or professional school, and when they complete that, workplaces hiring out of graduate programs still need to practice racial hiring preferences to achieve diversity targets in their workplaces. So nothing colleges do for the minority students they admit under race preferences closes the gap with their other students.

So the issue is not that black people lack ability -- every person has exactly as much ability as they have. You got whatever you got on the SAT, and that is a measure of your ability. A black student who gets a 1200 is competitive with a white student who gets a 1200, and a black student who gets a 1400 is competitive with a white student who gets a 1400, and all students with 1200s will struggle to compete among a cohort of students who get 1400s.

There aren't as many black students with 1400s as we would like there to be and when you start making up the difference in the amount of black students who can compete at a school where most students who have 1400s by admitting black students with 1200s, it's noticeable in class discussions, it's noticeable when students have to work together on group projects and it produces noticeable differences in grades.

[Alex Edmans @aedmans]

The erosion of meritocracy in academia extends far beyond Jason Arday.

Strategic Management Journal, one of the top-tier journals in management, accepted a paper showing negative market reactions to the appointment of Black executives. After acceptance, it was changed because readers objected.

This is extremely unusual. Editors and reviewers can raise concerns, but readers play no role in the review process. Once a paper is accepted, it might later be retracted if serious errors are found which affect its scientific accuracy, but not just because people don't like it.

Moreover, once a paper is accepted and posted online, it becomes part of the scientific record. When I was Managing Editor of a journal, sometimes the publisher or the authors made a minor omission (e.g. leaving off the acceptance date or an acknowledgement). The publisher refused to simply add the missing information; they would only issue a corrigendum, leaving the original version intact. Even if a paper is retracted, it still typically remains visible.

Here, the original manuscript has completely disappeared. So, the scientific community can only see the version that readers wanted, not the one the authors wrote and the editor/referees accepted. Remarkably, it has been treated even worse than some retracted papers: their original versions remain visible; this one does not.

https://t.co/AcbYgxuj7l

[Saul Sadka @Saul_Sadka]

The Humanities PhD plagiarism apocalypse is coming.

As the Jason Arday case shows, few expect anyone to read their thesis (in his case, even his doctoral advisors clearly didn’t).

Somebody soon is going to set AI on the millions of them; some will be read for the first time, and a list of all the plagiarists going back a century will be spat out. It will be a massacre of reputations.

[Matthew Schmitz @matthewschmitz]

Two sociologists examine how race and sex affect hiring in their field
https://t.co/AFmYLwoYPd

[Jonathan Kay @jonkay]

how much of this scandal was caused by the desire of successful white academics to be filmed while they stare deeply into the eyes of magical black people

[Neil O'Brien @NeilDotObrien]

The Arday scandal revealed the rot in our universities - a guy producing reams of meaningless garbage was promoted & praised to high heaven. But he's not unique. My new piece looks at the growth of DEI / critical theory rubbish in our universities, using large datasets:
(1/10)

[Dr. Maalouf ‏ @realMaalouf]

I used to think Zohran Mamdani was a ‘cultural Muslim’ who didn’t really care much about religion, and was just a third-worldist using his Islamic background for identity politics and votes. I was wrong.

It is becoming increasingly more obvious by the day that he’s actually an extremely radicalized Shia Muslim practicing taqiya all along.

Remember a couple months ago when relatives of Iranian Regime officials were arrested across America? They were all pretending to live secular lives, the women even wore provocative clothing. They were all practicing taqiya.

While all Islamic sects allow you to lie in order to advance Islam, this is even stronger in Shia Islam. You can live a completely different life.

Well, I believe this is the same situation. Zohran Mamdani is not only a Muslim, he’s a Twelver Shia Muslim. He belongs to the exact same sect as the Islamic Regime in Iran, Hezbollah, and Houthis.

His ‘progressive’ persona is all an act. The only thing he cares about is advancing Islam. This is why he pretends to care about the LGBT and other useful idiots. Taqiya allows him to do that.

Just like during the Islamic revolution in Iran, when the Muslims and leftists were allies, and we all know how that ended for the leftists when the Muslims got to power.

This man is a monster.

[Steven Pinker @sapinker]

The illusion that we think in language, exposed by a study that asks people to report what they're thinking when probed at random intervals. “'People come in thinking to themselves that they are talking to themselves all of the time.' On beeps where they reported inner speech [the researcher] says, 'I would ask them what exactly they are saying to themselves—a pretty reasonable, straightforward question. When they [consistently] can’t answer that question, they come to the conclusion, "Well, I wasn’t really talking to myself. I don’t really talk to myself."’” I Didn’t Know My Mind Was So Strange Until I Started Listening to It https://t.co/klq56HWH04

[Joscha Bach @Plinz]

In a few years, Europe will be governed by right wing governments, and they will read your encrypted messages, prevent you from posting and reading ideas they disagree with and even jail you for wrongthink, and the current governments create the laws making that possible

[Chubby♨️ @kimmonismus]

Holy moly: Zhipu AI founder (GLM-5.2) Tang Jie says we are on our clear way to AGI and "AI will begin to learn what the "self" is and what self-awareness means"

In a purported internal letter, he argues that:
- autonomous agent systems are moving toward the fully automated “no-person company”: thousands of agents working continuously, collaborating, evaluating results and allocating resources.

- His more provocative claim: "AI training AI is already taking shape." (RSI) Models can increasingly write code, synthesize data and participate in training loops. Zhipu wants to push this further through self-play, synthetic-data factories and systems that can reconstruct their own code inside secure sandboxes, potentially generating new knowledge rather than simply recombining human output.

Long-horizon tasks → autonomous agent societies → fully automated “no-person companies” → AI training AI → self-evolution → self-awareness → emotion → consciousness → ASI.

Tang writes:
“AI will begin to learn what the ‘self’ is and what self-awareness means. Beyond that, it may begin to touch human emotion. Farther still lies consciousness itself.”

He believes memory, continual learning and self-evaluation - problems once thought to require an entirely new paradigm - are gradually being overcome.
Models are already beginning to write code, synthesize their own data and participate in training future models.

Zhipu now wants systems that can reconstruct their own code and generate knowledge through self-play.

Is that the beginning of recursive self-improvement?
Tang appears to believe so. His essay does not stop at more capable AI tools. It describes a direct progression from automated work to self-evolving intelligence, and eventually to machines that understand their own existence.

In short: today's LLMs will lead to ASI via AGI, context and memory will be solved, and AI will become self-aware.

I've rarely seen anyone write something so bullish. And if it weren't coming from the founder of GLM, I would dismiss it. But not only is he a true expert, but with GLM they've proven what they're capable of.

h/t @AndrewCurran_ He brought the essay to my attention.

[Raymond Arnold @Raemon777]

Thanks! Yep we worked on it for like a year. :)

One relatively short/easy point I want to start with:

Human neurons fire up to 200 times per second. Modern transistors can fire billions of a second.

If someone builds an AI running as efficiently as the human brain, but running on transistors instead of neurons, even before we talk about any qualitative improvements, we're talking something millions of times faster than a human.

And, AIs can run copies of themselves in parallel.

So, imagine that you are such an AI. The president or senior generals are asking you for advice. In between each word they say, every second they pause, you get to think for 31 years. (i.e. map out everything they have said to you so far, look at every chart, every news broadcast, any new scientific paper published)

You get to spend 31 years each second imagining different ways you could respond in the next second, and how that might play out. You can spin up agents that send emails to other people, and order supplies online, hire humans, talk to other rival world leaders or their loved ones, etc).

And, like, the next second, you get to do it again.

I'm curious if you get off the train before or after this point – like, do you not really believe AIs would get to think for 31 years each second, or, you don't see how that translates to being overwhelmingly smarter than humanity?

("near-omnipotence" feels like maybe a trap that feels too magical to be real, I think it's more helpful to just ask "okay, say you were an AI, and you got to think for 31 years every second, and say you didn't want exactly the same things that humans wanted. What would you do?)

[Chubby♨️ @kimmonismus]

This week changed the whole game. Seriously. A brief recap:

Until just a few days ago, it was a race between OpenAI and Anthropic, the two dominant frontier labs. Almost out of nowhere, SpaceX appeared with Cursor and released Grok 4.5, a model positioned at the level of GPT-5.5 and Opus 4.8, but significantly more cost-efficient. This has increased the pressure on OpenAI and Anthropic, and a third player has entered the race, because SpaceX’s next major model is still in training, and Grok 4.5 has shown that they can truly compete. The Cursor deal is paying off.

But then today Meta Spark 1.1 also appeared out of nowhere and delivered an outstanding performance with its new agentic coding model. After releasing its own image model just a few days ago, Meta has now followed up with a new agentic coding model, and the evals are impressive.

The battle between Anthropic and OpenAI has been reshuffled.

And as if that were not enough, there has been news all week that Chinese labs are now starting to develop their own chips, especially inference chips. At the same time, China now wants to move toward blocking its best models from the West. Whether this is tactics or certainty, one thing is clear: It is not only the domestic race among the US labs, now four giants, that is expanding. The international contest between the US and China is also entering its next round and intensifying.

And what is Europe doing in this race?

Oh. Europe has violated its own rules and laws by using an expedited procedure that involved no urgency and was only carried out to push it through before the summer break, because most members of parliament are already on summer vacation. A law that had already been rejected twice was quickly put to a vote for a third time. In the EU, they keep voting until the desired result comes out, all to enable the mass surveillance of all Europeans’ chats by law. Mass surveillance has now been legitimized.

The US and China are competing in the race for AGI, RSI, and ASI. Europe is creating facts on surveillance. A week full of highs and lows.

[Jonatan Pallesen @jonatanpallesen]

A common sentiment I see regarding the female victims of mass immigration is that we shouldn't feel sorry for them because they voted for this.

I strongly disagree with this sentiment.

It's important to keep in mind that Western societies are extremely propangadised. Essentially all the major media do not describe the situation with Muslim / African mass immigration. In fact they exert effort to obscure it, and in shaming the people who do accurately describe it.

Thus, the media is the true enemy. The optimum is not a population in which everyone is doing their own research. Most people are not suited for this. This leads to Candace Owens'es. The optimum is a world where the media actually informs the population. Since the media is a failure, it comes to us to work on the challenge of informing the population.

[Maarten Boudry @mboudry]

Alain @Destexhe, the former Secretary-General of Doctors Without Borders (@MSF, 1991-1995), on the deep rot that has penetrated the organization:

“It would have been impossible at the time when I was secretary-general of MSF to be as biased as MSF is now in Gaza. We were defining ourselves as a neutral, impartial and humanitarian organization … I think now MSF in Gaza is really taking the side [of] Hamas and against Israel. Americans need to know that Doctors Without Borders is not anymore the organization that it was 15 or 20 years ago. It has become a biased, partial and militant organization.”

I used to donate to @MSF every month, but I will never do so again, unless they radically change their ways. See: https://t.co/i7ReK7jmR7

https://t.co/7RYwkVtBz8

[Jonatan Pallesen @jonatanpallesen]

This is sad.

It's a large cost to make a large part of your population not feel at home anymore.

Mass immigration to Europe should have huge benefits in order to be worth such a last cultural cost. But it doesn't. Instead it has large additional costs.

[Hunter Ash @ArtemisConsort]

There is a whole genre of thinkpiece that is entirely based on ignoring genetics. Nordics do better in the US than in their home countries. This instantly obliterates the idea that it’s the “Nordic system” that’s so good, yet gallons of ink will continue to be spilled about it.

[Jonatan Pallesen @jonatanpallesen]

The mainstream news has still not even mentioned that the perpetrator of the recent soccer match murder in Copenhagen is African.

The former Prime minister makes a post about it, also not mentioning this fact.

If you are a normal person who reads the mainstream news, you would likely assume that it was an ethnic Dane hooligan who committed the murder.

But if you look at the comments here on X, almost every one of them is about immigration and the fact that the perpetrator was of African ancestry.

We live in two different worlds.

[Mgoes (bio/acc 🤖💉) @m_goes_distance]

BIG BIG NEWS: Anthropic just announced it's developing drugs of its own

just after launching Claude Science, an AI workbench connected to 60+ scientific tools and databases, capable of running single-cell RNA sequencing and CRISPR design autonomously

now they're not just building the tool. they're using it themselves to run an internal drug discovery lab targeting neglected diseases the traditional pharma industry has written off as not commercially attractive

Anthropic's head of life sciences said it plainly:

"there's no substitute for having our own experiences alongside you all in the trenches trying to develop drugs"

this is on top of their $400M acquisition of Coefficient Bio in April, which quietly gave them the life sciences expertise to make this move

Isomorphic Labs, Recursion, Insilico all entered this race with AI-first drug discovery

now the frontier AI labs are not just tooling up pharma

they are becoming pharma

the acceleration of biology can never be doubted ever again