Inteligencia artificial
Cómo los Memristores están haciendo que la IA sea más parecida a la humana
El auge del hardware de IA neuromórfico, similar al cerebro
As AI becomes the center of the tech industry, a growing problem has emerged: the massive computing and energy demand of AI when performed using CPUs and GPUs.
As a result, researchers are working hard on Neural Processing Units (NPUs), also called neuromorphic chips, a type of AI hardware that mimics the brain’s neurons.
“No es que nuestros chips o computadoras no sean lo suficientemente potentes para lo que hacen. Es que no son lo suficientemente eficientes. Consumen demasiada energía.”
The shift toward brain-inspired hardware could reshape how we approach artificial intelligence. Neuromorphic designs offer three major advantages over conventional chips:
- Arquitectura adaptativa: circuitería que puede reconfigurarse según los datos de entrenamiento.
- Eficiencia energética radical: en algunos casos, usando tan solo 1/100th de la potencia de una GPU.
- Menor emisión de calor: reduciendo los costosos requerimientos de refrigeración que aquejan a los centros de datos de IA actuales.
“Poder desarrollar microchips que imiten la actividad neuronal real significa que no se necesita mucha energía en modo de espera o cuando la máquina no está en uso.
Eso puede ser una gran ventaja computacional y económica potencial.
Researchers are testing several promising methods for creating neuromorphic chips. One approach involves leveraging incipient ferroelectricity—a still poorly understood phenomenon that could allow materials to spontaneously switch their electric polarization under the right conditions. Another focuses on active substrates made from vanadium or titanium, materials that can dynamically change their electrical properties to mimic brain-like signaling.
Perhaps the most widely discussed path is the use of memristors—a revolutionary class of electronic components capable of storing information through resistance changes. These devices can perform AI tasks at as little as 1/800th the normal power consumption, making them one of the most energy-efficient solutions under development.
Cómo los Memristores imitan las sinapsis
Desliza para desplazarte →
| Característica | CPU | GPU | NPU / Chip de Memristor |
|---|---|---|---|
| Arquitectura | Secuencial, propósito general | Paralelo, enfocado en matrices | Inspirado en el cerebro, adaptativo |
| Uso de energía | Alto | Moderado a alto | Extremadamente bajo (1/100–1/800 de la energía) |
| Eficiencia de aprendizaje | Lento, memoria externa | Entrenamiento rápido, memoria externa | En memoria, auto‑adaptativo |
| Mejor caso de uso | Computación general | Entrenamiento de modelos de IA | IA en el borde, robótica, IA de bajo consumo |
Los memristores son componentes electrónicos que imitan las sinapsis que conectan neuronas al recordar qué estado eléctrico tenían después de que se apagara la energía.
Esto puede reducir enormemente la energía y el tiempo perdidos al trasladar datos de ida y vuelta entre procesadores y memoria.
Una de las principales fortalezas de los memristores es su capacidad para un aprendizaje in situ eficiente y auto‑adaptativo, lo cual es crítico para aplicaciones en robótica y vehículos autónomos.
Además, el bajo consumo de energía de los memristores es particularmente beneficioso en robótica y vehículos autónomos, donde la eficiencia energética es primordial.
Many paths are being explored on how to create the best memristors, from relatively conventional titanium oxide memristors to using actual human neurons (organoids) or even mushrooms.
The idea of using organic material, including actual neurons, to mimic the activity of neurons makes sense on a theoretical level. However, in practice, interfacing such a “computer” to traditional IT systems can be challenging.
“Nuestros sistemas informáticos existentes nunca fueron diseñados para procesar enormes cantidades de datos o para aprender por sí solos a partir de solo unos pocos ejemplos.
Una forma de impulsar tanto la eficiencia energética como la de aprendizaje es construir sistemas artificiales que operen según los principios observados en el cerebro.”
This lack of efficiency is staggering when compared to the human brain. A young child can learn to recognize handwritten digits after seeing only a few examples of each, whereas a computer typically needs thousands to achieve the same task.
And the human brain performs this feat while consuming around 20W of power, while the latest AI data centers are looking at GW-scale, or almost a hundred million times more power.
A new intermediary option could be to create artificial chips that act like neurons in their basic principle. This is the path taken by researchers at the University of Southern California, University of California, University of Massachusetts, Syracuse University, Air Force Research Laboratory, and NASA Ames Research Center.
They published their results in Nature Electronics1, under the title “Una neurona artificial de disparo basada en un memristor difusivo, un transistor y una resistencia”.
Replicando cómo las neuronas disparan usando memristores difusivos
¿Cómo funcionan las neuronas?
The way neurons interact with each other, and ultimately process information, is by using both electrical and chemical signals.
If a signal is strong enough, it generates an electrical impulse called an action potential by allowing positively charged sodium ions to flood into the cell.
Source: Nature Electronics
When this electric signal is received, it causes the release of neurotransmitters.
Until now, electronic memristors and complementary metal–oxide–semiconductor (CMOS) circuits have used electric signals to virtually simulate such functioning, requiring hundreds of transistors to simulate a neuron.
Instead, the researchers developed a device called a “diffusive memristor”, which also uses actual chemical interactions to start computational processes.
¿Qué son los memristores difusivos y cómo funcionan?
While traditional silicon systems rely on electrons to perform computations, diffusive memristors use the motion of atoms instead. They use silver ions embedded in oxide materials to generate electrical pulses that mimic natural brain functions.
Source: Nature Electronics
Of course, this does not replicate exactly how a neuron works, but the principle is very similar.
“Even though it’s not exactly the same ions in our artificial synapses and neurons, the physics governing the ion motion and the dynamics are very similar.”
In part, this similarity comes from the fact that silver ions are easy to diffuse in this memristor system, similar to how sodium ions can move in organic cells.
Besides silver, the memristor also uses palladium, silicon, titanium, and hafnium. The research could visualize in real time the diffusion of silver in response to an electric stimulus.
Desliza para desplazarte →
| Capa / Material | Rol en el dispositivo | Por qué es importante |
|---|---|---|
| Iones de plata (Ag) | Especie móvil para picos | Se difunde fácilmente, permitiendo pulsos impulsados por iones similares al disparo neuronal |
| Matriz de óxido (p.ej., HfO2) | Hospedaje de iones / medio de conmutación | Controla el movimiento de iones y la formación de filamentos para los estados memristivos |
| Paladio (Pd) | Electrodo / interfaz catalítica | Contacto estable y química de interfaz favorable |
| Titanio (Ti) | Capa de adhesión/barrera | Mejora la estabilidad del electrodo y la integridad de la pila |
| Silicio (Si) | Sustrato / integración CMOS | Permite integración vertical dentro del huella de un transistor |
El futuro: Chips neuromórficos y IA energéticamente eficiente
A key advantage of this new type of memristor is that it fits within the footprint of a single transistor, whereas older designs required tens or even hundreds.
The initial test used only a handful of such diffusive memristors, demonstrating that they can be used for building the typical multi-level neural network used by almost all AI systems today.
Source: Nature Electronics
The next step will be to assemble many more of such systems to test how efficient they can be at actually performing AI tasks.
“We are designing the building blocks that eventually led us to reduce the chip size by orders of magnitude, and reduce the energy consumption by orders of magnitude.
So it can be sustainable to perform AI in the future, with a similar level of intelligence without burning energy that we cannot sustain.
Finding out if other ions can be used could also be useful, as silver ions are not commonly used in semiconductor manufacturing, which could limit the speed of adoption of this design by the industry.
Another effect of diffusive memristors is that they could help better understand how biological brains work.
In the long run, they are likely to be especially useful for so-called “edge computing”, where computation is done directly on site, like for example with a robot or self-driving car having to make a decision without connection to an AI data center.
Invertir en fabricantes de chips neuromórficos
Intel (INTC )
Intel is a giant in the semiconductor sector and has evolved over the years from a founder of the industry to a scientific and innovation leader, losing the top spot of manufacturing volume to companies like Taiwan’s TSMC.
Intel is a leader in neuromorphic computing, including through its Loihi 2 chip.
It also created the Intel Neuromorphic Research Community, which includes Pennsylvania State University, involved in vanadium dioxide research, as well as many other potential neuromorphic designs, and 75+ other research groups.
(INTC )
Intel is also very active in mimicking biological sense through replicating the way our brain works (itself a branch of neuromorphic computing), something we discussed further in our article “Biomimetic Olfactory Chips: Are Artificial Intelligence and E-Noses the Next Canary in a Coal Mine?”.
Overall, research from Intel Lab is at the forefront of semiconductor innovation, including AI, quantum computing, neuromorphic computing, etc. (We discussed Intel advances in quantum computing in our article “The Current State of Quantum Computing”).
You can also read more about Intel’s current business and R&D programs in our dedicated investment report.
Últimas noticias y desarrollos de acciones de Intel (INTC)
Estudio Referenciado
1. Zhao, R., Wang, T., Moon, T. et al. Una neurona artificial de disparo basada en un memristor difusivo, un transistor, y una resistencia. Nature Electronics (2025). https://doi.org/10.1038/s41928-025-01488-x
