Autor Tema: PPCC: Pisitófilos Creditófagos. Veranito 2025 (Leído 462746 veces)

pollo · « **Respuesta #4035 en:** **Hoy** a las 15:43:42 »

Cita de: senslev en Hoy a las 09:52:47

Cita de: pollo en Hoy a las 00:27:12
En la línea de lo que hablábamos, aquí está este artículo muy interesante y con experiencia de primera mano, que viene a confirmar lo que estábamos hablando tanto unos como otros.

Citar
https://www.cerbos.dev/blog/productivity-paradox-of-ai-coding-assistants

The productivity paradox of AI coding assistants

Our development team at Cerbos is split into two camps. One side uses AI coding help like Cursor or Claude Code and sees it as the fastest way to ship. The other side has the typical “meh” reaction you often see on Reddit, arguing that AI assistance is mostly a racket.

Some of us lean on AI coding to push side projects faster into the delivery pipeline. These are not core product features but experiments and MVP-style initiatives. For bringing that kind of work to its first version, the speed-up is real.

AI coding assistants promise less boilerplate, fewer doc lookups, and quicker iteration. From my perspective, they deliver on that promise when building MVPs, automations, and hobby projects. Outside of those use cases, the picture changes. You may feel like you are moving quickly, but getting code production ready often takes longer.

Let’s dig into the data.

[...]

Hay una observación extremadamente interesante sobre la dopamina en el artículo, que viene a decir que el mecanismo de la IA nos engaña dándonos la recompensa de un trabajo hecho, pero sin hacer esfuerzo.

Esto ya es cosecha propia que se desprende de lo anterior: ese mecanismo es una fórmula perfecta para la adicción. No hace falta decir que si esto ya puede ser adictivo para un adulto (y que además induce a no esforzarse porque es lo que te están recompensando de forma inmediata) imaginemos para un adolescente o niño. La tentación de buscar la respuesta sin sufrir el estrés de tener que comerse la cabeza y enfrentarse al problema es demasiado grande.

O se hace algo al respecto o tendrá consecuencias extremadamente graves en el futuro. El conjunto de individuos con potencial intelectual que lo hayan desarrollado de forma correcta y a tiempo (después de los 20 ya es muy complicado) va a reducirse enormemente. Y es algo que ya estaba bajo asedio con consecuencias que pueden verse a nivel civilizatorio desde hace unos años.

https://www.youtube.com/watch?v=Q9DKLuwWepI

Esto se ha dicho también el año pasado y el anterior. Y también iban a tener el coche autónomo total hace 8 años y seguimos exactamente igual. Todo se basa en hype y en maquillar la realidad sacando supuestas nuevas versiones y modelos que son lo mismo que los anteriores con un lavado de cara, pero lo fundamental no cambia.

La realidad es que esta tecnología no da para más y se ha llegado a la parte asintótica de las mejoras.

Para que se cumplan todas estas previsiones sobre la desigualdad, quedar atrás, etc. lo primero que debería cumplirse es que la IA fuese el oráculo infalible que nos venden que es.

Y no lo es. Ni de puta coña.

Si hubiese descubrimientos revolucionarios en el campo ya se sabría hace tiempo porque no habrían parado de hablar de ellos de cara al futuro. Los LLM no dan para más y desde luego nada parecido a una inteligencia.

Es cuestión de tiempo que se vea la IA como el farolazo que es (el runrun mediático no ha parado de crecer al respecto, y llegará un momento que sea vox populi).

Lo que sí que es real es el peligro de que la gente se vuelva aún más imbécil de lo que ya era.

senslev · « **Respuesta #4036 en:** **Hoy** a las 16:27:26 »

Derby · « **Respuesta #4037 en:** **Hoy** a las 16:27:41 »

Cita de: pollo en Hoy a las 15:43:42

Lo que sí que es real es el peligro de que la gente se vuelva aún más imbécil de lo que ya era.

https://www.wsj.com/tech/ai/will-ai-choke-off-the-supply-of-knowledge-8a71cbcd

Citar

Will AI Choke Off the Supply of Knowledge?

More people turn to ChatGPT and other large language models for answers, but they don’t add to the stock of knowledge

Rachel Mendelson/WSJ, iStock

In January, OpenAI Chief Executive Sam Altman and Chief Product Officer Kevin Weil hosted a demonstration of ChatGPT’s soon-to-be-released “deep research” application. A Beltway audience watched as Weil asked ChatGPT to prepare a memo briefing a fictional senator for the confirmation of Albert Einstein to be energy secretary.

ChatGPT soon produced a thorough profile of Einstein, listing his technical and engineering accomplishments, leadership style, strengths (“a globally respected scientist-statesman”) and weaknesses (“never managed a large organization”) plus questions the senator could ask (“You have been an outspoken voice on nuclear issues since WWII. As Energy Secretary, how will you ensure the safety of nuclear power plants and uphold U.S. commitments to nuclear nonproliferation?”).

The benefits of such impressive, and now routine, capabilities, were obvious: enormous savings of time and effort. Of course, there were potential costs: How many jobs could researchers, writers and other knowledge workers lose to artificial intelligence?

I wondered about a different cost: How much knowledge will be lost to AI? Large language models (LLMs) such as ChatGPT, Google Gemini and Anthropic’s Claude excel at locating, synthesizing and connecting knowledge. They don’t add to the stock of knowledge.

By contrast, when humans answer questions, such as whether Einstein should be energy secretary, they often pursue novel avenues of inquiry, creating new knowledge and insight as they go. They do this for a variety of reasons: salary, wealth, fame, tenure, “likes,” clicks, curiosity.

If LLMs come to dominate the business of answering questions, those incentives shrivel. There is little reward to creating knowledge that then gets puréed in a large language blender.

Consider the fate of Stack Overflow, a website where software developers ask and answer questions, becoming both a wellspring and repository for knowledge.

But then developers started putting their questions to ChatGPT. Six months after its introduction in November 2022, the number of questions on Stack Overflow had fallen 25% relative to similar Chinese and Russian language sites where ChatGPT wasn’t an alternative, according to a study by Johannes Wachs of Corvinus University of Budapest and two co-authors.

The drop was the same regardless of quality, based on peer feedback, refuting predictions that AI would displace only low-value research.

As of this month, the number of questions is down more than 90%. Why should anyone other than Stack Overflow’s owners care? Because, as tech writer Nick Hodges explained in InfoWorld, “Stack Overflow provides much of the knowledge that is embedded in AI coding tools, but the more developers rely on AI coding tools the less likely they will participate in Stack Overflow, the site that produces that knowledge.”

Stack Overflow may be an extreme case. A different study found no similar decline on Reddit.

But there are signs of similar effects elsewhere. Many LLMs are trained on Wikipedia, a repository of knowledge compiled and curated by humans. Columbia University business professor Hannah Li and five co-authors found that between the year before and the year after ChatGPT’s launch, views fell for Wikipedia pages most similar to what ChatGPT could produce. Edits also dropped, a potential sign of less incentive to contribute, although the data were inconclusive.

Meanwhile, as Google has enabled users to answer queries through AI without clicking on links, web publishers large and small have seen referral traffic from search plummet.

Internet search itself is at risk. Last year, District Court Judge Amit Mehta found Google, a unit of Alphabet, had an illegal monopoly in search. But the dramatic growth of AI prompted him to impose surprisingly light penalties last week. “For the first time in over a decade, there is a genuine prospect that a product could emerge that will present a meaningful challenge to Google’s market dominance,” Mehta wrote.

If LLM output comes to dominate the web, the web will become, well, dumber. Columbia’s Li said in an interview: “What happens when we train LLMs on other LLM outputs? The overall outcomes get worse. The models get worse. This is what they call model collapse.”

There is a parallel in what index funds and other passive strategies have done to the stock market.

They don’t do research and price discovery (the process of negotiation that reveals an asset’s value). Instead, they free ride on the research and price discovery of active investors. In other words, they exploit market efficiency without contributing to it. In the process, they are squeezing out active investing, leaving a market increasingly dominated by algorithms trading against each other.

These are, I’ll admit, dystopian scenarios. I could tell a different story of how AI will help scholars discover connections between otherwise disparate bits of knowledge across the web. Joshua Gans, a University of Toronto economist who has written extensively on AI, thinks that so long as new knowledge has value, it will find a way to be created. He says when AI insights are incremental, humans will pivot to more truly novel research.

Maybe. But instead of pivoting, what if humans lose interest in learning altogether? Reliance on AI can cause critical thinking to atrophy, just as reliance on GPS weakens spatial memory. A study by Nataliya Kosmyna at Massachusetts Institute of Technology and seven co-authors asked three groups of subjects to write essays, one using an LLM, one using internet search, and one just their brains.

Scans later showed the LLM group had the least engagement across brain regions such as for memory recall and executive functioning; the brain-only group had the most.

Mental engagement, the authors argue, is enhanced by “novelty, encountering new or unexpected content.” That resonates. Dissatisfied with OpenAI’s demo, I searched the web for biographies and writings of Einstein.

I learned his father’s company made electrical equipment based on direct current, then went bust when alternating current triumphed; that he was outspoken in support for civil rights in the U.S. and against oppression of Jews in Germany, for which the Nazis put a price on his head; that during the McCarthyite fervor of the 1940s and 1950s he was called a foreign-born agitator spreading communism; that he wasn’t a communist but was a socialist. In a 1949 essay for a socialist journal, he answered a question I often ponder: how economics differs from the physical sciences, like astronomy: “economic phenomena are often affected by many factors which are very hard to evaluate separately.”

I have no idea if any of that bore on his qualifications to be energy secretary. I could have spent the time on more productive work. But then, acquiring new knowledge has never felt like work.

Derby · « **Respuesta #4038 en:** **Hoy** a las 17:03:12 »

https://www.baha.com/Trump-calls-for-50percent-100percent-China-tariffs-major-Russia-sanctions/news/details/64810299

Citar

Trump calls for 50%-100% China tariffs, major Russia sanctions

US President Donald Trump called for tariffs of 50%-100% on China until the Russia-Ukraine war is over. In a letter to all NATO countries posted on Saturday, he added that he's ready to impose major sanctions on Russia, but only when all NATO countries agree on the same sanctions and when they cease oil purchases from Russia.

"I am ready to do major Sanctions on Russia when all NATO Nations have agreed, and started, to do the same thing, and when all NATO Nations STOP BUYING OIL FROM RUSSIA," the letter posted on Truth Social read. " I believe that this, plus NATO, as a group, placing 50% to 100% TARIFFS ON CHINA, to be fully withdrawn after the WAR with Russia and Ukraine is ended, will also be of great help in ENDING this deadly, but RIDICULOUS, WAR," he continued.

Derby · « **Respuesta #4039 en:** **Hoy** a las 18:20:49 »

https://www.aol.com/articles/gigantic-price-bubble-famed-economist-091501018.html

Citar

'We are in a gigantic price bubble': Famed economist warns extreme stock valuations point to negative returns ahead

David Rosenberg warns of negative S&P 500 returns due to high valuations.

The Shiller CAPE ratio on the index is at its third-highest level ever.

At the same time, recession risks are rising as the labor market slows, Rosenberg warned.

David Rosenberg isn't always right. The founder of Rosenberg Research, who rose to fame after calling the 2008 recession, regularly expresses a bearish outlook for the economy and markets that often don't come to fruition.

But in a world where bullish forecasts are the consensus among Wall Street's top equity strategists, it can be prudent to heed Rosenberg's warnings. While his predictions usually don't play out, there's no denying that the economist sufficiently shows his work, providing relevant data that ought to give investors pause.

In a recent note to clients, Rosenberg provided some concerning numbers on where the S&P 500's forward returns could be headed, given current valuations.

The index's Shiller cyclically adjusted price-to-earnings ratio is hovering around 37.5. The measure smooths out business cycles by comparing current stock prices to a 10-year rolling average of earnings.

It's the third-most expensive level of all-time, behind peaks in 2021 and 2022.

Valuations are usually reliable predictors of long-term stock market performance. Bank of America data shows that starting valuations can explain about 80% of the market's performance over the following 10 years. Last year, Morgan Stanley and Goldman Sachs strategists said that high valuations would lead to relatively weak returns for the market over the coming decade.

In the short term, valuations are poorer predictors of performance. Rosenberg's data, however, shows that when the market gets this historically expensive — though, granted, it has only happened twice before — one-year forward returns have been negative.

The column on the right in the table below shows forward returns over 1-, 3-, 5-, and 10-year periods when the Shiller CAPE ratio gets above 35.

"It's the only cutoff point where every single time is negative," Rosenberg said in an interview with Business Insider on Thursday.

Valuations alone aren't why Rosenberg is skeptical of the rally. It's the heightened expectations paired with a weakening economic backdrop as the labor market continues to show signs of slowing. Job growth has been below 100,000 per month over the past four months, data from the Bureau of Labor Statistics show. And the economy has added 911,000 fewer jobs than previously thought in the year through this March, the BLS said this week.

Rosenberg believes the outlook will continue to worsen.

He pointed to initial jobless claims rising to 263,000 last week, worse than economists expected and at levels that should trigger downward pressure on payroll growth. All of this adds up to the US economy already being in the midst of or on the precipice of a downturn, he said.

"What we know arithmetically is that the hiring rate today is so low that once you cross above 240,000 on claims, it triggers a negative impulse on nonfarm payrolls, which I think is what we're going to see when the September data roll out early October," he said.

The fact that stocks continue to punch through all-time highs despite these warning signs — showing heightened investor sentiment — is one clue that stocks are in bubble territory, Rosenberg said.

"This is what a euphoric state looks like we're seeing it in real time," he added. "We are in a gigantic price bubble that is ongoing. And you know it's a price bubble when prices move up in the face of negative fundamentals."

Derby · « **Respuesta #4040 en:** **Hoy** a las 19:26:53 »

https://www.baha.com/Trump-said-to-be-doubting-his-ability-to-influence-Putin/news/details/64810391

Citar

Trump said to be doubting his ability to influence Putin

US President Donald Trump has recently started to question his ability to influence Russian President Vladimir Putin, according to a report published on Saturday by Axios.

The outlet also stated that Trump has admitted to his close associates that he misjudged Putin's desire for peace and denied that Israeli Prime Minister Benjamin Netanyahu is trying to influence him.

Over a month ago, Trump threatened to punish Putin severely if he did not implement a ceasefire in Ukraine or make meaningful progress toward peace during their summit in Alaska. Since then, Russia has escalated its attacks on civilians in Ukrainian cities.