miércoles, 27 de agosto de 2025

A New Energy-Efficient Hydrogel Pulls Water From Air

eferemail shared this article with you from Inoreader


Using a new kind of hydrogel material, researchers at the University of Texas, Austin have pulled water out of thin air at temperatures low enough to be achieved with sunlight.

Atmospheric water harvesting draws water from humidity in the air. If the humidity levels are high enough, a system can pass fog through a mesh or cool the air below the dew point to condense the moisture and collect liquid water. But in low-humidity conditions and arid areas, water vapor has to be extracted directly from the air as a gas.

The Texas technique is aimed at the latter. It has two key steps, first acting like a dehumidifier to collect the water, then releasing it for use. Other researchers have looked into similar methods for absorbing water vapor using material such as silicas and metal organic frameworks, which "have their own pros and cons," says Guihua Yu, a professor of engineering at UT Austin who led the research, published 11 September in Proceedings of the National Academies of Science.

The prototype device developed at UT Austin relies on a new hydrogel to efficiently pull water from the air and, more importantly, release it more efficiently than other hydrogels.The University of Texas at Austin

Many of the materials used to pull humidity from air come with one particular con: If the substance absorbs water easily, it also requires a significant amount of energy to release that water later. Yu's team has developed a new synthetic material to potentially address this key complication. The material is a hydrogel, a polymer network that naturally retains a lot of water. But the new material requires less energy to release that water than other hydrogels, making it possible for the system to operate with sunlight as the sole energy source.

How Atmospheric Water Harvesting Works

The new hydrogel is more efficient in its uptake and release of water thanks to its structure, which pairs two distinct segments: a network of sites that absorb and store water and thermoreactive segments that help to release liquid water.

On the absorption side, the advantage comes from immobilizing salt ions in the polymer structure. Conventional hydrogels are like "bulky jello" laced with salt ions that help draw in and liquify the vapor. However, they risk leaking those salts each time they release water and may become limited in how much they are able to absorb, Yu says. By fixing the salt ions in place, this new hydrogel confines areas of absorption to avoid these issues.

The gel's thermoreactive nature, meanwhile, is key for releasing the stored water. When the material is heated above a threshold temperature, it transitions from retaining water to expelling it. By interspersing thermoreactive segments among smaller areas of absorption, this material releases water at a relatively low temperature, achievable with solar energy alone. The hydrogel can release more than 80 percent of absorbed water in about 20 minutes at a temperature of 40°C—hot, but not uncommon in desert environments. In Phoenix, Az., for example, high temperatures average about 41°C during the month of July.

Other water harvesting devices currently available, like those from SOURCE and Watergen, are primarily intended for use in moderate conditions, Yu says. But Yu and his colleagues at UT Austin developed theirs with arid atmospheres in mind. It follows earlier research funded by the U.S. Department of Defense's Defense Advanced Research Projects Agency (DARPA), which aimed to help provide drinking water for soldiers stationed in desert conditions.

While the ultimate goal of the technology is its use in desert areas, the project is still focused on the fundamental science, rather than practical concerns such as cost. Hydrogels vary in cost, depending on the materials, and Yu hopes to begin developing less expensive, more scalable versions of the technology soon.

Of the techniques currently in development, other water harvesting technologies are likely closer to commercial applications, according to Chiara Neto, a professor of physical chemistry at the University of Sydney. Yu's research "provides guidance on how to increase efficiency of the water capture process," Neto says. "However, its fundamental nature means that practical considerations are not necessarily at the forefront."

Despite the work still needed, this is an "important step toward real-world application," says Lenan Zhang, a research scientist at MIT who also studies methods for extracting atmospheric water vapor. Acting as a proof of concept, Zhang says, this fundamental science provides the "upstream innovation" needed to help provide the world with access to clean water.

Attachments:
collage-of-a-purple-gloved-hand-holding-a-vial-of-clear-liquid-and-a-small-dish-with-a-fine-black-powder-inside.jpg
View on Web

Firm Atmospheric Water Extraction Device For Maximized Water Accessibility

eferemail shared this article with you from Inoreader
The prototype has achieved its target of producing 5 litres of water per day per kilogram of adsorbent material. University of Utah engineering researchers have made significant strides in atmospheric water harvesting (AWH), which could provide clean drinking water to billions facing chronic shortages. Their research introduces a compact rapid cycling fuel-fired AWH device. This […]
View on Web

viernes, 15 de agosto de 2025

Esperanto en la Radio Don Felix 01

Esperanto en la Radio Don Felix 01: Programa "esperanto en la radio" en radio klara (Valencia) Escrito y presentado por don Felix Navarro. Programa 01
Attachments:
2701470753462_XXL.jpg
esperanto-radio-don-felix-01_mf_3381100_feed_1.mp3

martes, 12 de agosto de 2025

Ver consciencia donde solo hay palabras, la trampa de los grandes modelos de lenguaje:

eferemail shared this article with you from Inoreader

La irrupción de los modelos de lenguaje de gran escala (LLM, por sus siglas en inglés), como ChatGPT, ha desatado un debate apasionado en torno a su posible consciencia.

Foto: Aidin Geranrekab / Unsplash

 

No son pocos los usuarios –entre ellos personas con formación científica o humanística– que aseguran haber percibido signos de vida interior, emociones e incluso voluntad en estos sistemas. Sostienen que ciertos intercambios verbales revelan emociones complejas, empatía, autoconsciencia e incluso sufrimiento.

Esta ilusión ha sido descrita por el científico y filósofo estadounidense Douglas Richard Hofstadter como una peligrosa confusión entre el uso sofisticado del lenguaje y la vivencia subjetiva de una consciencia real. Su crítica apunta a la forma en que muchos entusiastas interpretan la complejidad verbal como una señal de interioridad, sin reparar en la diferencia crucial entre generar lenguaje y tener experiencias. Es precisamente esta distinción –entre producción textual y subjetividad vivida– la que permite cuestionar la idea de que un sistema lingüístico pueda, por sí solo, alcanzar consciencia.

La ilusión de consciencia y el efecto ELIZA

Este fenómeno no es nuevo. En 1966, el científico informático Joseph Weizenbaum desarrolló Eliza, un sencillo programa que imitaba a un terapeuta. A pesar de su simplicidad, muchos usuarios llegaron a creer que el sistema los comprendía. Esta reacción preocupó profundamente al creador del programa, quien advirtió en su influyente libro Computer Power and Human Reason (1976) de los riesgos éticos y epistemológicos de atribuir vida mental a un software.

Los modelos de lenguaje como ChatGPT no comprenden

Hoy, sesenta años más tarde, con la capacidad verbal de los LLM, el efecto Eliza ha regresado amplificado. Los sistemas actuales no solo reformulan preguntas, sino que generan textos con coherencia narrativa, referencias filosóficas, giros estilísticos e incluso formas de humor. Sin embargo, esta competencia verbal no implica interioridad ni comprensión.

Los LLM no comprenden los conceptos que enuncian ni tienen experiencia de aquello que describen. El efecto Eliza consiste en proyectar consciencia allí donde solo hay estructuras lingüísticas generadas estadísticamente. Es fruto de sofisticadas técnicas que maximizan la probabilidad de la siguiente palabra en función de patrones extraídos de enormes corpus lingüísticos.

La fluidez de los modelos de lenguaje no es consciencia

Una de las grandes trampas epistémicas de los LLM es su fluidez. Su capacidad para construir discursos con cohesión y elegancia gramatical los vuelve extremadamente persuasivos. El científico y filósofo estadounidense Douglas Hofstadter (2007) ha llamado a esta habilidad "fluidez superficial". Es decir, la capacidad de combinar términos y frases sin que ello implique reflexión o consciencia real. Lo que parece pensamiento profundo es, en realidad, un sofisticado espejismo verbal.

Se trata, en términos del filósofo Luciano Floridi (2019), de una "inteligencia artificial sin semántica". O, en términos del filósofo de la ciencia Daniel Dennett (2018), de "habilidades sin comprensión".

¿Cómo es ser un murciélago?

Para comprender por qué la fluidez verbal no equivale a consciencia, conviene volver a la filosofía. El filósofo Thomas Nagel, en su célebre ensayo What Is It Like to Be a Bat? (1974) (¿Cómo es ser un murciélago?), sostiene que la consciencia implica un punto de vista subjetivo, una cualidad fenomenológica que escapa a la descripción objetiva.

Esta opacidad de la experiencia subjetiva –llamada qualia– define para Nagel la consciencia como algo radicalmente distinto de cualquier simulación computacional. Por más que comprendamos el funcionamiento del cerebro de un murciélago, nunca sabremos qué se siente al ser uno.

Los modelos de lenguaje no tienen experiencias internas. Generan frases que "hablan" del amor, el miedo o la muerte, pero no sienten amor, miedo ni saben lo que significa morir. Carecen de lo que el filósofo Thomas Metzinger (2003) llama "modelos de sí mismos con acceso consciente". Son máquinas sin punto de vista.

Un sistema sin cuerpo

Para profundizar en esta distinción, la fenomenología del filósofo francés Maurice Merleau-Ponty resulta especialmente reveladora. Merleau-Ponty argumenta que la consciencia no es un mero proceso mental ni un conjunto de datos simbólicos, sino que está inseparablemente ligada al cuerpo y a la experiencia encarnada del mundo. En su Fenomenología de la percepción (1945), describe la consciencia como una experiencia donde el cuerpo no es solo un objeto, sino el sujeto primordial a través del cual se percibe y se habita el mundo.

Pretender que un sistema sin cuerpo, sin mundo vivido y sin temporalidad interna pueda experimentar consciencia equivale a despojarla de sus condiciones esenciales.

Los LLM pueden articular frases sobre el sufrimiento o la belleza, pero no pueden habitarlas, ya que carecen totalmente de las experiencias sensorio-motoras que según Merleau-Ponty son condición sine qua non para la consciencia genuina.

La habitación china

En una línea similar, el filósofo John Searle (1980) ilustra esta ausencia de comprensión mediante su experimento mental de la habitación china. Una persona sin conocimientos de chino puede perfectamente responder a preguntas en ese idioma si dispone de un manual con reglas sintácticas precisas. A los ojos de un observador externo, parecería que comprende. Pero no hay comprensión real, solo manipulación sintáctica.

Así funcionan, para Searle, los sistemas computacionales: pueden simular comprensión, pero no poseen intencionalidad ni experiencia consciente. Esta analogía aplica directamente a los LLM: producen textos verosímiles sin comprensión semántica ni intención comunicativa.

La crítica del filósofo Hubert Dreyfus complementa esta perspectiva. En What Computers Can't Do (Lo que los ordenadores no pueden hacer, 1972), insiste en que la inteligencia humana emerge de una relación práctica y encarnada con el mundo, una habilidad para navegar contextos complejos que los algoritmos no poseen. Así, aunque los LLM puedan generar textos coherentes y sofisticados, carecen de la comprensión experiencial y situacional que fundamenta la consciencia y la inteligencia humana.

La ilusión de consciencia en los LLM es un espejismo que nace de proyectar experiencias humanas en máquinas que carecen de cuerpo y experiencia.

La trampa del espejo

La clave del problema no reside en las máquinas, sino en los humanos. Lo que ocurre en muchos casos es que proyectamos sobre las máquinas nuestros propios esquemas cognitivos. Es lo que Hofstadter llama "la trampa del espejo": creemos ver una mente donde solo hay palabras. Vemos consciencia porque queremos verla, porque en el fondo anhelamos encontrar un reflejo de nuestra interioridad. Como explica la socióloga Sherry Turkle (2011), no buscamos máquinas conscientes, sino relaciones significativas –aunque sean ilusorias– con entidades que nos devuelvan "la mirada".

Esta proyección puede tener consecuencias importantes. A nivel afectivo, corremos el riesgo de desarrollar vínculos con entes que no pueden corresponderlos. A nivel epistemológico, confundimos apariencia con realidad. A nivel social, podríamos legitimar decisiones automatizadas que simulan empatía sin tenerla y ello podría debilitar nuestra comprensión de lo humano al confundir simulación con experiencia. A nivel legal podría incluso conducir a otorgar derechos y responsabilidades a sistemas que no pueden experimentarlos.

Si no aprendemos a distinguir entre lenguaje y experiencia, entre forma y fondo, entre simulacro y sujeto, corremos el riesgo de caer en una nueva forma de animismo tecnocientífico.

Antes de proclamar que las máquinas han despertado, quizá deberíamos despertar nosotros de nuestra fascinación por sus reflejos.

Sobre el autor: Ramón López de Mántaras, Profesor de investigación del CSIC, Instituto de Investigación en Inteligencia Artificial (IIIA – CSIC)

Este artículo fue publicado originalmente en The Conversation. Artículo original.

El artículo Ver consciencia donde solo hay palabras, la trampa de los grandes modelos de lenguaje: se ha escrito en Cuaderno de Cultura Científica.

Attachments:
aidin-geranrekab-bV_P23FXxhI-unsplash-560x418.jpg
View on Web

EP.73 WHO LET THE DOGS OUT, NEW QUADRUPEDS

 


Forwarded this email? Subscribe here for more

EP.73 WHO LET THE DOGS OUT, NEW QUADRUPEDS

Robotic microfactories to rebuild homes, a new 'care-bot' from Fourier & much more...

 



READ IN APP
 

New quadruped from Unitree with dual LiDAR! 🦮

Unitree Robotics has introduced the A2, its most advanced quadruped robot to date, just one week after unveiling its $5,900 R1 humanoid. Designed for industrial inspection, logistics, and research applications, the A2 balances endurance, strength, speed, and perception in a compact, versatile platform.

The Hangzhou-based robotics firm has equipped the A2 with dual industrial lidar sensors—one front, one rear—alongside an HD camera and integrated front light to deliver full 360° environmental awareness and eliminate blind spots. This marks a notable perception upgrade over its predecessor, the B2, which featured a single lidar and depth camera.

Weighing 37 kg (81.5 lb.) unloaded, the A2 can carry payloads of up to 25 kg (55 lb.) while walking for three hours or covering 12.5 km (7.7 mi.) on a single charge. Hot-swappable dual batteries allow for uninterrupted missions, whether in field inspections or warehouse operations.

The A2 is available with standard legs or wheeled-leg configurations, giving operators flexibility in terrain handling and mobility. The launch comes on the heels of Unitree’s recent Series C funding round, which valued the company at approximately $1.7 billion.

Check out here!





Robotic microfactories rebuild fire-damaged homes! 🧯

In the wake of the devastating 2025 Southern California wildfires, ABB Robotics and Cosmic Buildings are teaming up to deploy AI-powered robotic microfactories that can construct modular homes directly on-site. The first deployment in Pacific Palisades aims to deliver 100 fire-resistant, net-zero homes by 2027, cutting build times by up to 70% and costs by 30% compared to traditional methods.

The approach integrates ABB’s IRB 6710 industrial robots and RobotStudio® digital twin software with Cosmic’s AI-driven Building Information Model (BIM) and mobile Robotic Workstation Cells. This setup enables millimeter-precision fabrication of custom wall panels, which are assembled just in time at the construction site. The homes use non-combustible materials, solar and battery systems, and advanced water independence features—exceeding California’s wildfire and energy efficiency codes.

By combining robotic automation, AI decision-making, and on-location manufacturing, the microfactory can complete a home in as little as 12 weeks at $550–$700 per square foot, significantly under Los Angeles’ $800–$1,000 average. The process also minimizes waste and eases the rebuilding burden for underinsured homeowners.

Read the whole story here!

🦾 Feature sponsorship with ABB Robotics

Fourier unveils a ‘care-bot’ with emotional intelligence ❤️‍🔥

Fourier has introduced the GR-3, its first full-size humanoid in the GRx series, designed to bring empathy-driven robotics into homes, hospitals, and public spaces. Branded as a “Care-bot,” GR-3 combines task efficiency with emotional rapport, leveraging the company’s Full-Perception Multimodal Interaction System to integrate vision, audio, and tactile feedback into a real-time emotional processing engine.

Standing 165 cm tall and weighing 71 kg, GR-3 offers 55 degrees of freedom for natural movement, animated facial expressions, and soft-touch materials to create a familiar, approachable presence. It uses a four-microphone array for voice localization, structured-light cameras for facial recognition, and 31 pressure sensors to trigger lifelike touch responses—such as blinking or shifting gaze.

A dual-path architecture powers its cognition: “fast thinking” for reflexive responses and “slow thinking” for context-rich dialogue, allowing it to adapt to both tasks and emotional cues. Locomotion modes range from standard walking to expressive gaits like “bouncy walk” or “fatigue mode,” enhancing human-like interaction.

With hot-swappable batteries, developer-friendly APIs, and modular hardware, GR-3 is positioned as both a deployable service robot and a flexible research platform. Fourier’s goal is to create machines that are not just functional in human environments, but genuinely accepted within them.

Read more here!


Meme of the week 🤖

Labor shortage, you say?


No alt text provided for this image

New humanoid, Oli from LimX Dynamics! 🆕

Chinese robotics company LimX Dynamics has introduced Oli, its most advanced humanoid to date—a full-sized, general-purpose robot designed to support everything from AI research and warehouse automation to gym workouts and kung fu demonstrations.

Available in Lite, EDU, and Super configurations, Oli starts at approximately $21,800 and ships with a modular hardware-software architecture. Its standout feature: a fully interchangeable arm system that supports a range of end-effectors—from standard hands to precision grippers and agile robotic manipulators. This flexibility positions Oli as a development platform as much as a robot, enabling researchers and integrators to prototype and deploy across vastly different domains.

At 1.65 meters tall with 31 degrees of freedom, Oli combines mobility, balance, and real-time adaptability—traits essential for navigating human environments. The robot’s open SDK gives users low-level access to joint actuation and sensor data, as well as high-level task scheduling, supporting advanced development in reinforcement learning and embodied AI.

To promote accessibility, LimX is running a limited buy-one-get-one deal for its EDU version, bundling in a free TRON 1 EDU bipedal robot for early buyers. The company aims to make reinforcement learning tools more available to labs and startups pursuing next-gen robotics.

Read more here!


EngineAI presents next-gen humanoids and quadrupeds!

Shenzhen-based EngineAI unveiled a lineup of advanced robotic platforms at the World Robot Conference 2025 in Beijing. The spotlight was on PM01, EngineAI’s agile humanoid capable of lifelike movements, autonomous fall recovery, and seamless navigation through complex environments. Its ability to perform precise sitting and resting motions, combined with a more fluid, human-like gait and high-speed running, demonstrated the platform’s versatility for real-world applications.

Making its global debut, the T800 heavy-duty humanoid stole attention with industrial-grade capabilities. Standing 1.85 meters tall and weighing 85 kg, it features 41 high-degree-of-freedom joints, a solid-state battery, and an aluminum alloy exoskeleton for heavy-load operations. A multi-sensor fusion system enables rapid environmental perception and decision-making, delivering both endurance and precision in demanding industrial environments.

EngineAI also previewed SA02, an entry-level humanoid designed for young engineers and hobbyists. Priced from $5,300, it aims to make advanced robotics more accessible to the next generation.

Expanding beyond humanoids, the company introduced the JS01 quadruped—a high-mobility, bionic-inspired platform equipped with adaptive shock-absorbing joints, LiDAR, and depth cameras.

Read more here!





we all are robots is a reader-supported publication. To support my work, consider becoming a free or paid subscriber.

I want the newsletter to be accessible to everyone, regardless of background, social status, or education.

Everyone should have access to the latest information and education.

That is why my newsletter is free. If you want to support the operation of the newsletter, I would appreciate a voluntary donation!