Javier Velilla retweeted
Geoff Eldridge « When the code becomes the easy part » nemlog.substack.com/p/when-t…
Javier Velilla retweeted
El CEO de DonWeb dijo recién que "en el 2000 no había carrera que te enseñaran a programar, a diseñar, era todo autodidacta".
Guillermo: esto es FALSO. ¿de dónde sacaste eso? Yo empecé a estudiar una carrera formal de programación en 1998 y en el 2001 empecé a dar clases en una carrera privada. La carrera de computador científico se creó en 1963. Analista de Sistemas arrancó en los '70. Durante los '80 y '90 hubo cursos cortos, carreras terciarias, tecnicaturas y carreras de grado para programación y diseño en muchísimos lugares. Por 1995 ya había cursos de programación en academias privadas. Había revistas. Libros. Existía ya todo un ecosistema armado para la educación en programación.
"Hoy, al contrario del 2000, tenemos todo Internet para preguntar, todo a flor de piel para aprender" 🤌
"Hoy volvemos a tener una necesidad de ser autodidacta" 👎
Vengo combatiendo es mito hace años. Antes era "Internet", después "está todo en youtube", después "tenés las clases gratis de Harvard".
Es un mito elitista y simplista: yo aprendí autodidacta, todos pueden y deben aprender así. Los que venimos dando clases hace años sabemos que eso es FALSO. No todo el mundo es autodidacta: no importa bajo qué paradigma, es con libros, con enciclopedias, con videos de youtube o con una IA. Mucha gente necesita un apoyo docente por diversos motivos. Hasta el día de hoy no hay registro de una herramienta que convierta a todo el mundo en autodidacta. La IA tampoco, al menos por ahora. El proceso educativo no existe por el objetivo de disponibilizar información al destinatario y, por ende, si la información ahora está disponible, el proceso educativo ya no tiene sentido. La información es lo menos relevante.
Javier Velilla retweeted
Just released: my book "AI for Smarties -- Understanding Artificial Intelligence", ai-for-smarties.com (e-book, €29.95, €24.95 until July 15, for an extra discount read this thread to the end). It is a general intro to AI, for people who want to understand... [1/3]🧵
Yann LeCun says you cannot build a reliable agentic system without a world model
LLMs don't have world models. They can't predict the consequences of their actions before taking them
"they just act, and whatever happens next is someone else's problem"
Without that, it's not intelligence
Javier Velilla retweeted
Very useful analysis, thanks. Also look up the technology developed by @Recognyze_AI, recognyze.ai, based on the "don't sue them, charge them!" principle -- monetize LLM access to Web content in a simple and fair way.
Javier Velilla retweeted
Friendly reminder, Hinton is the same guy who told us this:
"We should stop training radiologists now. It's just completely obvious that within five years, deep learning is going to do better than radiologists.".
That was six year ago.
He was right about the AI. Wrong about the job.
AI already reads scans better than any human. Six years after this nonsense prediction we have more radiologists. That's because Hinton and many others just fundamentally misunderstand that tasks are not jobs and that the job of a radiologist is also interacting with patients, being a light in darkness, providing hope and warmth and care and a thousand other things.
Again we need to stop listening to these folks.
I cannot say it enough, just because this fellow is brilliant with AI does not mean he has any clue how it will impact society. None. Zero.
Javier Velilla retweeted
Celebrating Tony Hoare's mark on computer science - a personal assessment and recollection bertrandmeyer.com/2026/03/16…
Javier Velilla retweeted
Over 50 million views for that @MattShumer_ blog (“Something Big is Happening”) and not a shred of actual data.
Nobody thought to ask what are the data? Does this match reality? Is Shumer credible? Did he give a single concrete example?
People gonna believe what they want to believe, but here are 7 critical points most people missed:
• Shumer’s blog post is weaponized hype, filled with vivid narrative and marketing speech, but stumbles on the facts, especially with respect to reliability. He gives no actual data to support this claim that the latest coding systems can (reliably) write “perfect” whole complex apps without making errors.
• When he describes how AI’s are doing longer and longer tasks on METR’s famous task-time benchmark, he neglects to say the criterion on that benchmark is 50% correct, not 100%, and that the benchmark is only about coding and not tasks in general. No AI system can reliably do every five-hour long task humans can do without error, or even close. But you wouldn’t know that reading Shumer’s blog, which largely ignores all the hallucination and boneheaded errors that are so common in every day experience.
• Shumer didn’t cite the new Caltech/Stanford article that reviews a wide range of reasoning errors in so-called reasoning models [or the Apple reasoning paper or the ASU mirage paper, etc]. The picture he sells just isn’t realistic, however much people might wish it were true.
• Shumer is the guy who was once famous for apparently exaggerated claims about a big model of his that didn’t replicate and that many people saw as fraud; he likes to sell big. But that doesn’t mean we should take him seriously.
• Shumer made no reference to a different METR study showed that coders sometimes imagine big productivity gains where they actually lost productivity. (Even though he selectively mentioned their other well-known study).
• He also didn’t acknowledge that other user’s experience is certainly not “it’s usually perfect.” @KelseyTuocreported a few weeks back that Claude Code was sometimes perfect, and other times maddening. (Example: “Sometimes, Claude is absolutely the worst coworker you’ve ever had. At one point, it deleted every single one of the phoneme files of each English sound pronounced [that she was working with in her app] absolutely correctly, which I had personally emailed an English teacher to secure permission to use, and replaced them with AI-generated sounds which were all subtly wrong.”) Shumer glosses over that kind of experience.

•LLMs write code wicked fast, but some coders are starting to report burnout, and only relatively modest gains relative to that burnout, as a new story from Connie Loizos at Tech Crunch just reported.
• Shumer is actually correct that something has changed recently. You really can let things rip more in the most recent systems. Bit, quoting from a developer friend “sometimes! [it] will come out with the right answer. … Sometimes… [But] ultimately, I think this makes it more dangerous …. Generally, the closer these systems are to appearing right, the more dangerous they become…”
• The security of the autogenerated code is very much in question.
The bottom line is this: LLMs are certainly coding more, but it’s not clear that the code they are creating is secure or trustworthy. Shumer’s presentation is completely one-sided, omitting lots of concerns that have been widely expressed here and elsewhere.
A lot of people may have taken his post seriously, but they shouldn’t have.
Adapted from my newsletter, Marcus on AI (nearly 100k subscribers).
Javier Velilla retweeted
GPT-5 hot take
• took almost 3 years, many billions of dollars
• good progress on many fronts
• but still part of the pack, not a giant leap forward (e.g. Grok 4 beats it on ARC-AGI-2 results)
• not the huge leap forward people long expected
• lots of questions TBD about real-world performance
• obviously not AGI
Also:
• pricing is good, but profits may continue to be elusive; still no clear technical moat.
• what was up with those graphs (at least two) where the bars didn’t match the numbers? Did GPT-5 make them??
Quote of the day: “I dont wanna read anything about exponential progress anymore” - @flowersslop
Javier Velilla retweeted
“Do AI Models Help Produce Verified Bug Fixes?”
A lot less than you might think, mostly helping beginners, sometimes sending people off in tangents. In many cases AI-debugging actually makes things worse.
new study, link below:
Javier Velilla retweeted
How o3 and Grok 4 Accidentally Vindicated Neurosymbolic AI is now live!
Worked very hard on this one, because the issues are absolutely essential for a clear understanding of where we are in AI now and how we got here.
Link below. Featuring a lot of history, a critique of @geoffreyhinton’s less generous side, and some sharply-observed bits from @_KarenHao’s new book.
Adiós Nonino de #AstorPiazzolla interpretado por este magnífico y brillante octeto liderado por Nico Sorín para La Librería, Música Argentina.
youtu.be/99WMoQQUgvY?si=ochU…
101
Javier Velilla retweeted
Incredible how @yLeCun’s chief talking point is *exactly* what I said in Deep Learning is Hitting a Wall, over three years ago.
The whole company steals.
"we will not get to human-level AI by just scaling up LLMs... it's not going to happen"
Meta AI chief scientist, Yann LeCun:
scaling up LLMs alone will not lead to human-level AI
...what we’ll have instead are systems with huge memory and retrieval abilities, but still not able to invent solutions to new problems
Let's help Universidad Nacional del Sur!reconstruir.uns.edu.ar/
42
Javier Velilla retweeted
I'm teaching databases this semester at Berkeley. My students all seem unusually brilliant. Not many go to office hours, and not too many folks post on the course forum asking project questions.
Weirdly, the exam had the lowest recorded average in my 10 semesters teaching it.
Javier Velilla retweeted
Some people today are discouraging others from learning programming on the grounds AI will automate it. This advice will be seen as some of the worst career advice ever given. I disagree with the Turing Award and Nobel prize winner who wrote, “It is far more likely that the programming occupation will become extinct [...] than that it will become all-powerful. More and more, computers will program themselves.” Statements discouraging people from learning to code are harmful!
In the 1960s, when programming moved from punchcards (where a programmer had to laboriously make holes in physical cards to write code character by character) to keyboards with terminals, programming became easier. And that made it a better time than before to begin programming. Yet it was in this era that Nobel laureate Herb Simon wrote the words quoted in the first paragraph. Today’s arguments not to learn to code continue to echo his comment.
As coding becomes easier, more people should code, not fewer!
Over the past few decades, as programming has moved from assembly language to higher-level languages like C, from desktop to cloud, from raw text editors to IDEs to AI assisted coding where sometimes one barely even looks at the generated code (which some coders recently started to call vibe coding), it is getting easier with each step.
I wrote previously that I see tech-savvy people coordinating AI tools to move toward being 10x professionals — individuals who have 10 times the impact of the average person in their field. I am increasingly convinced that the best way for many people to accomplish this is not to be just consumers of AI applications, but to learn enough coding to use AI-assisted coding tools effectively.
One question I’m asked most often is what someone should do who is worried about job displacement by AI. My answer is: Learn about AI and take control of it, because one of the most important skills in the future will be the ability to tell a computer exactly what you want, so it can do that for you. Coding (or getting AI to code for you) is a great way to do that.
When I was working on the course Generative AI for Everyone and needed to generate AI artwork for the background images, I worked with a collaborator who had studied art history and knew the language of art. He prompted Midjourney with terminology based on the historical style, palette, artist inspiration and so on — using the language of art — to get the result he wanted. I didn’t know this language, and my paltry attempts at prompting could not deliver as effective a result.
Similarly, scientists, analysts, marketers, recruiters, and people of a wide range of professions who understand the language of software through their knowledge of coding can tell an LLM or an AI-enabled IDE what they want much more precisely, and get much better results. As these tools are continuing to make coding easier, this is the best time yet to learn to code, to learn the language of software, and learn to make computers do exactly what you want them to do.
[Original text: deeplearning.ai/the-batch/is… ]

















