Fil T.

Brain drained. Adding 🧠 to 🤖.

[nature @Nature]

Rather than racing to replace people, we should build AI tools that amplify human expertise and expand opportunity, says Daron Acemoglu

https://t.co/RBH9QkaKbV

[The Economist @TheEconomist]

This week we published an article critically assessing the work of @DAcemogluMIT. In response, the Nobel prize-winning economist has robustly defended his research and core ideas. Read his letter in full: https://t.co/9LXCR6T1L3

[Fil T. @hplessromantic3]

Think in systems. Like Uncle Bob thinks of systems.

[Uncle Bob Martin @unclebobmartin]

I'm in Inception. I'm using agents to build a system that harnesses agents to build systems...

[More Luigis Needed @LisaA311]

What the fuck

[Fil T. @hplessromantic3]

@anamorfosis Didn't want to sound critical 🙂. Again, cool use case!

[Fil T. @hplessromantic3]

@vivian_e Καλημέρα, ωραίο ποστ άξιο συζήτησης. Όντως η αποπληρωμή είναι θέμα πολιτικό όπως και η επικοινώνηση του. Νομίζω ότι προς αυτήν τη μεριά (πρέπει να) κινείται η κουβέντα.

Επίσης κατ'εμε ο "ο λόγος που πτωχεύσαμε είναι γιατι .." είναι μια λανθασμενη υπεραπλούστευση.

Καλημέρα σας!

[Fil T. @hplessromantic3]

Social media 101: Never feed the trolls.

[elapoueisai🔴⚪️ @xa_xa_xa_xo_xo]

ΑΥΤΟΣ Ο ΤΥΠΟΣ

ΗΤΑΝ ΥΠΟΥΡΓΟΣ ΟΙΚΟΝΟΜΙΚΩΝ ΤΗΣ ΧΩΡΑΣ

ΛΙΓΟ ΝΑ ΞΕΡΕΙΣ ΑΠΟ ΟΙΚΟΝΟΜΙΚΑ

ΔΕΝ ΘΕΛΕΙ ΠΟΛΥ

ΚΑΤΑΛΑΒΑΙΝΕΙΣ ΤΙΣ ΑΡΛΟΥΜΠΕΣ ΠΟΥ ΛΕΕΙ

ΕΝΑ ΠΟΤ ΠΟΥΡΙ ΨΕΚΑΣΜΑΤΟΣ

ΑΣΥΝΔΕΤΩΝ ΕΠΙΧΕΙΡΗΜΑΤΩΝ

ΠΟΥ ΦΥΣΙΚΑ ΣΤΑ ΜΑΤΙΑ ΤΩΝ ΟΙΚΟΝΟΜΙΚΑ ΑΝΑΛΦΑΒΗΤΩΝ ΣΥΝΤΡΟΦΩΝ

ΦΑΙΝΕΤΑΙ WOW 😂🤡

[Fil T. @hplessromantic3]

@kimmonismus No. No benefit over other models of the same class neither on quality nor price

[Andrew Akbashev @Andrew_Akbashev]

Terence Tao, one of the most well known mathematicians, speaks up on AI in mathematics in his new paper:

“What if an AI tool generates a lengthy proof that is verified to be correct, but which nobody — 𝘯𝘰𝘵 𝘦𝘷𝘦𝘯 𝘵𝘩𝘦 𝘩𝘶𝘮𝘢𝘯𝘴 𝘸𝘩𝘰 𝘱𝘳𝘰𝘮𝘱𝘵𝘦𝘥 𝘵𝘩𝘦 𝘵𝘰𝘰𝘭 — understands? This is no longer hypothetical. Sites devoted to collecting mathematical problems already contain dozens of AI-generated proof submissions. Many of these are likely to be correct; but in a substantial number of cases no human expert has yet volunteered to verify and vouch for them, and in several cases the human submitters have themselves declared that they are not qualified to do so.

We may soon be faced with the very real possibility of a verified proof of a major result that NO HUMAN understands well enough to explain.

For a proof to actually contribute to its field, then, it is NOT enough for it to be correct, and NOT enough for it to be readable. It also needs to be accepted and valued by the community: other mathematicians need to 𝗱𝗶𝗴𝗲𝘀𝘁 𝘁𝗵𝗲 𝗿𝗲𝘀𝘂𝗹𝘁 𝗮𝗻𝗱 𝗶𝗻𝗰𝗼𝗿𝗽𝗼𝗿𝗮𝘁𝗲 𝗶𝘁 𝗶𝗻𝘁𝗼 𝘁𝗵𝗲𝗶𝗿 𝗼𝘄𝗻 𝘄𝗼𝗿𝗸.

Our current publication infrastructure relies on human editors and referees to provide this acceptance, voluntarily and largely without credit. This work is routinely regarded as less prestigious than the work of generating proofs in the first place; but it is an essential component of the profession, and it is precisely the mechanism by which the individual achievements of mathematicians are converted into collective progress and understanding.

Finally, even publication is not the last stage. Key results should ultimately become part of the definitive textbooks and reference material of their subject, in the form in which 𝘁𝗵𝗲𝘆 𝗮𝗿𝗲 𝘁𝗮𝘂𝗴𝗵𝘁 𝘁𝗼 𝘁𝗵𝗲 𝗻𝗲𝘅𝘁 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝗼𝗻 𝗼𝗳 𝘀𝘁𝘂𝗱𝗲𝗻𝘁𝘀. This process of canonicalization is the slowest stage of all. It requires broad, deliberative consensus, and it is the stage least amenable to optimization by AI tools.”

📍 Terence Tao concludes:

“We will transition from an era of proof scarcity to an era of proof abundance. Most of our institutions — journals, priority conventions, hiring and promotion criteria, prizes, the very notion of a research program — were designed under the assumption of scarcity, and it should not surprise us if they behave poorly under abundance.

In some areas, particularly in education and in the training of young mathematicians, it will be crucial to emphasize 𝘁𝗵𝗲 𝗶𝗿𝗿𝗲𝗱𝘂𝗰𝗶𝗯𝗹𝘆 𝗵𝘂𝗺𝗮𝗻 𝗮𝘀𝗽𝗲𝗰𝘁 of our work, and to restrict the use of AI tools quite tightly; the goal of training a mathematician is NOT achieved by producing correct homework.

In other areas, we will need to take the initiative on AI usage, and define best practices for incorporating these tools into our workflows on our own terms rather than on terms set for us by vendors.”

__

[I highlighted & capitalized words in the text for clarity]

[Fil T. @hplessromantic3]

“For what shall it profit a man, if he shall gain the whole world, and lose his own soul?”

— Mark 8:36

[signüll @signulll]

started as an ai safety research lab because openai was getting too commercialized.

now pushing a $65b revenue run rate straight into an ipo with goldman.

peak character development.

[Fil T. @hplessromantic3]

@petergostev Most random people do not care for fable-level quality. They want a "cool" chat style daily assistant for low/mid "intelligence" stuff. Fable is amazing as productivity enhancer but does not have a friendly/cool vibe for normal people.

[Erik Brynjolfsson @erikbryn]

.@TheEconomist is a great publication, and I’ve learned a lot from it over the years. But I think it missed the mark with this article on @DAcemogluMIT.

Daron and I publicly and privately disagree on a number of important issues, most notably on the likely magnitude of AI’s productivity effects in the coming years. We’ve had many lengthy discussions about our differences, including as recently as last month.

That said, Daron is not merely one of the most intelligent, interesting, and hardworking economists I know. He is also a person of great integrity.

Since many people may form their impressions of him from this article or other public commentary, let me offer three anecdotes from my own experience.

1. In 2017, @tommmitchell and I co-chaired a @theNASciences committee on information technology and the workforce, bringing together a dozen leading scholars. Committee members were unpaid, and there was little extrinsic reward for doing more than the minimum. Everyone contributed, but unquestionably the person who worked hardest, and produced the most high-quality substantive material, was Daron. I came away enormously impressed by his seriousness and work ethic.

2. The article suggests that Daron’s stature may discourage others from criticizing him. That has not been my experience. I remember an MIT seminar where he presented an early version of his work on robots and employment. He was subjected to exceptionally tough criticism from colleagues and graduate students, especially over his instrumental variables and other aspects of his empirical strategy. It was one of the toughest seminars I can remember. Nobody was pulling punches. If anything, the criticism was blunter because everyone knew Daron was perfectly capable of handling it.

3. Along with @professor_ajay, @akorinek, and @testingham, I helped lead the recent “We Must Act Now” statement on AI (https://t.co/BjiX5fGGhJ). One of the first people we approached was Daron precisely because he had disagreed with me about AI capabilities and their likely economic effects. Daron and I went back and forth over the detailed wording for about a week. We ultimately found common ground. In the process, I was reminded of both his openness to new data and his attention to detail.

The Economist correctly highlights some important weaknesses in the economics profession. It should certainly scrutinize the work of prominent economists, presenting arguments and counterarguments.

In my view, what undermined their case was the article’s shift from criticizing his arguments to innuendo about Daron's personality. The suggestion that economists are unwilling to challenge him does not match what I have observed over 3 decades of knowing him.

Criticize Daron’s ideas and conclusions, as I often do. But the personal insinuations struck me as unfair.

[Fil T. @hplessromantic3]

@pasokoprophet Τόλμησε να γράψει κάτι κακό για την silicon valley και όρμηξαν.

[Fil T. @hplessromantic3]

One of the reasons I canceled my Economist Subsription. Targeted articles to anyone questioning the Valley's approaches.

[The Economist @TheEconomist]

Daron Acemoglu’s ideas can seem underwhelming for a titan of social science. Register for free to learn why economists are questioning the Nobel prizewinner’s reputation https://t.co/dJzSRAVXzz

[Fil T. @hplessromantic3]

@provigianos Εξαιρετικά επιφανειακή οπτική γωνία. Λες και οι "ελίτ" ξεφυτρώνουν σαν ροδάκινα από ένα υγιές οικονομικό σύστημα και δεν είναι αποτέλεσμα χρόνιων οικονομικών και κοινωνικών επιλογών που *δεν* γίνονται στην βόρεια Ελλάδα. Ας κατηγορήσουμε άλλη μια φορά το θερμόμετρο για τον πυρετό

[Fil T. @hplessromantic3]

@AlbertCamus_AM Το δεξιά για μένα ειναι must read. Το αριστερά είναι λιγο πιο light read που δεν "πάει σε πολύ βάθος". Depends on your time and effort willing to invest.

[Fil T. @hplessromantic3]

@thsottiaux Link chatgpt voice on projects and codex (remote). So we can discuss ideas on chatgpt voice (don't burn codex tokens) and once ready, spin codex agents to implement. Might be available already via some workflow which I can't find for EU/android (no sites yet) though.

[Fil T. @hplessromantic3]

@signulll I am glad the world cup is over and you can focus solely on this kind of commentary 🫡

[Fil T. @hplessromantic3]

Η αντίληψη των εγχώριων Ελλήνων «experts» για τη γεωπολιτική διάσταση της ΤΝ είναι ό,τι χρειάζεται να γνωρίζει κανείς για το πού μπορεί να καταλήξει η έρμη αυτή χώρα μετά από ακόμη ένα τεχνολογικό ανακάτεμα της παγκόσμιας τράπουλας.