%45%6c%65%63%74%72%69%63%61%6c%20%45%6e%67%69%6e%65%65%72%69%6e%67
-
KAIST and KIOM Develop a Wearable Microneedle Electroceutical for Personalized Pain Management
A "wearable electroceutical" has been developed that could reduce the need to visit a hospital or rely on painkillers every time pain occurs. When attached to the skin, it modulates pain through electrical stimulation and can be remotely controlled via smartphone even across long distances, such as between Korea and the United States.
KAIST (President Choongsik Bae) announced on September 13 that a joint research team led by Professor Jae-Woong Jeong from the School of Electrical Engineering at KAIST and Dr. Sanghun Lee from the KIOM (Korea Institute of Oriental Medicine, President Sung-Kyu Kho) has developed a wearable electroceutical platform that combines a wireless microneedle (fine needles that adhere to the skin) electroceutical with Internet of Things (IoT)-based remote control technology.
The key feature of this research is that a single small skin-attached device integrates stable electrical stimulation, smartphone-based remote control, and automatic stimulation based on the body's physiological state. Once its efficacy and safety are validated with actual patients, the technology may be used for personalized pain management at home or in daily life.
The need for such an approach is particularly relevant for chronic pain management. Painkillers are widely used to treat chronic pain, but long-term medication use raises concerns about side effects and dependency. In particular, opioid analgesics used for conditions such as cancer pain carry an increasing risk of tolerance and misuse with prolonged use, which has spurred research into "electroceuticals", devices that modulate nerves through electrical stimulation as an alternative to drugs.
However, existing electroceuticals have limitations. Implantable devices require surgery, and skin-attached electrodes may fail to deliver electrical current properly depending on skin conditions such as sweat or dead skin cells. A concentrated current at specific sites can also raise skin temperature or pose a risk of burns.
To address these issues, the research team developed a temperature-responsive, conductive, and adhesive microneedle electrode (a fine needle electrode that conducts electricity well while adhering to the skin).
Because the microneedles penetrate the highly resistive stratum corneum, the device can deliver stable electrical stimulation while reducing the influence of sweat and dead skin cells. The team also coated the electrode with a conductive hydrogel (a gel-like material that retains a large amount of water) so that current spreads evenly rather than concentrating at the needle tips.
In addition to improving electrical performance, the electrode was designed with a built-in thermal safety mechanism. When skin temperature rises abnormally, the electrode's adhesion weakens and it detaches from the skin on its own, helping reduce the risk of skin burns that could occur during electrical stimulation.
Beyond the skin interface itself, the research team integrated the device with IoT-based remote management. Using a smartphone and cloud server, a healthcare provider can control the device's operating time and electrical stimulation in real time or on a scheduled basis, even when located far from the patient. The team confirmed that the device could be remotely controlled even across long international distances, such as between Korea and the United States.
Following future clinical validation, this could develop into a home-based or remote pain management approach in which patients use the electroceutical under medical supervision without needing to visit a hospital.
The platform further extends beyond remote control by enabling automatic operation based on the body's physiological state. Using a photoplethysmography (PPG) sensor (a technology that measures pulse and blood flow changes using light), the team detected pain-related stress states and, based on this, implemented a closed-loop (a method that automatically adjusts treatment while continuously checking the body's condition) therapy function that automatically triggers electrical stimulation.
The researchers evaluated the performance of the platform in both animal experiments and a small-scale human study. In animal experiments, current was delivered more effectively than with conventional gel electrodes, and pain-relieving effects were also confirmed. In a small-scale study involving healthy adults, changes in skin sensory pain thresholds (the level of stimulation at which pain begins to be felt) following electrical stimulation were observed to assess the potential for application in humans.
However, direct analgesic effects in this study were confirmed only through animal experiments. The research team noted that further clinical studies are needed to confirm therapeutic efficacy and the safety of long-term use in actual chronic pain patients.
Professor Jae-Woong Jeong from the KAIST School of Electrical Engineering said, "By combining a stable skin interface with IoT-based remote management, we have expanded the potential for wearable electroceuticals in daily life. We hope that, following future clinical validation, this approach can evolve into a personalized digital healthcare platform that enables pain management tailored to each patient's condition."
Dr. Sanghun Lee from the KIOM said, "We hope this research will be integrated with future wearable acupuncture technologies to contribute to the development of a new non-pharmacological pain management approach that stimulates acupoints through electrical stimulation."
The study, co-first-authored by Heesoo Kim, a PhD student at KAIST, and Dr. Se Kyun Bang from the KIOM, was published in the international journal Nature Communications on August 28th.
Paper title: Wireless IoT-Enabled Microneedle Electroceutical for Personalized and Connected Pain Management, DOI: 10.1038/s41467-026-76527-y
Author information: Heesoo Kim (KAIST, co-first author), Se Kyun Bang (KIOM/UST, co-first author), Sanghun Lee (KIOM, co-corresponding author), Jae-Woong Jeong (KAIST, corresponding author), and 10 others
Demonstration video: https://www.dropbox.com/scl/fo/zrglys1t1c49l5u1xo08e/AJsbboH3sQMFag-6aO6sSSs?rlkey=dnux0jj80bdczws22u0dq7ugx&e=1&dl=0
This work was supported by the National Research Foundation of Korea (RS-2022-NR067853, RS-2025-02218624, RS-2024-00335066), and by the Korea Institute of Oriental Medicine (KSN2511012 and KSN2511013).
2026.09.14 View 93 -
KAIST Skin-Conformable Micro-LED Mask Boosts Skin Rejuvenation, Brightening, and Synergistic Benefits with Polynucleotide (PN) Injections
Home beauty devices that let users care for their skin conveniently at home have grown popular recently, but conventional LED masks are limited not only by their rigid structures but also by their point-emitting LEDs, which must be positioned away from the skin to spread light over a broader area. This inherently prevents close skin contact and increases optical loss.
KAIST (President Chung-Sik Bae) announced on September 2 that a joint research team led by Professor Keon Jae Lee from the Department of Materials Science and Engineering confirmed skin-brightening and elasticity-improving effects using a face-conforming LED mask. The mask combines a flexible surface-emitting micro-LED layer, consisting of a dense micro-LED array and a light-diffusing layer for uniform illumination, with a three-dimensional elastic scaffold that conforms to facial contours.
In a 2024 study published in Advanced Materials, Professor Lee clinically demonstrated that a flexible surface-emitting micro-LED mask produced up to 340% greater improvement in deep skin elasticity than conventional LED masks.
In the present study, the 3D elastic scaffold adapted to different facial contours, increasing skin-contact area from 46.9% to 78.1% and reducing the light-source-to-skin distance to 1.8 mm, while achieving 93.83% light uniformity across eight facial measurement sites.
The researchers evaluated the mask in a split-face clinical study involving 33 participants. All of the participants received PN injections across the entire face, while the micro-LED mask was applied to only one side for eight weeks. The micro-LED-treated side showed greater improvement across all five skin-brightening indices, including skin brightness, tone uniformity, skin exfoliation, melasma count, and melasma area. Consistent with the clinical findings, human-derived skin tissue treated with micro-LEDs also showed reduced expression of the melanogenesis-related markers MITF and TYR, supporting a direct contribution of the LED treatment to the brightening effect. PN injections are primarily known for skin rejuvenation, with limited evidence of a direct skin-brightening effect when used alone.
The synergy between PN injection and the micro-LED mask was also evident in skin regeneration and post-procedure recovery. Deep skin elasticity improved by 12.8% on the LED+PN side, compared with 3.1% on the PN-only side, representing approximately 4.1 times the improvement observed with PN alone. Skin-barrier recovery was faster, while post-procedure redness was reduced to a greater extent, consistent with the known anti-inflammatory and tissue-repair effects of red-light photobiomodulation. These findings suggest that the LED mask may boost mitochondrial ATP production in the skin and activate regenerative responses that PN alone cannot fully induce.
Professor Lee said, “This study clinically demonstrates that home-use LED masks can extend beyond skin rejuvenation to skin brightening and highlights the importance of delivering light in close contact with the skin. When combined with in-clinic skin-rejuvenation injections, the mask can substantially improve skin elasticity and accelerate post-procedure recovery, creating a new clinic-to-home care platform.”
This study was conducted jointly by researchers from KAIST and AMOREPACIFIC. A related product based on the technology is scheduled to launch in Japan in Q4 2026 and enter the U.S. market in Q1 2027.
The resulting paper, titled “Clinical validation of skin brightening and rejuvenation enabled by a skin-conformable surface-emitting micro-LED mask with injection,” was published in Nano Energy
(Vol. 157, Article 112288; DOI: 10.1016/j.nanoen.2026.112288).
2026.09.02 View 656 -
Neural Implant in Korea Remotely Controlled from the United States, Bringing Brain Research into the IoT Era
A researcher in Chicago remotely controls a miniaturized brain implant in Daejeon, Korea — over the internet. Korean researchers have developed a wireless device that can deliver drugs and light to precisely modulate targeted neurons from anywhere in the world. The technology is expected to overcome the constraints of distance and location, supporting long-term studies of brain disorders and the future development of therapeutic devices.
KAIST (President Choongsik Bae) announced on August 27 that a research team led by Professor Jae-Woong Jeong from the School of Electrical Engineering, in collaboration with Professor Wha Young Kim's team at Yonsei University College of Medicine, has developed an IoT-enabled wireless neural implant that integrates drug delivery, optical stimulation, wireless communication, and internet-based remote control into a single miniaturized device.
Conventional studies involving optical stimulation or drug delivery to the brain often required bulky equipment connected by wires, restricting the natural movement of experimental animals. Even wireless devices had their own limitations, often requiring researchers to operate them at close range, thereby restricting experimental flexibility and introducing the so-called “observer effect”.
To overcome these limitations, the research team developed the brain implant with IoT connectivity. Even without being physically present in the laboratory, researchers can remotely administer drugs or stimulate specific brain neurons with light in real time via the internet. The device can also be programmed to operate automatically at a preset time.
The device is about the size of a sugar cube and is designed not to interfere with the animal's natural behavior. Researchers no longer need to repeatedly approach or handle equipment near the animal, reducing the stress caused by a researcher's presence, which can otherwise affect the animal's behavior and bias experimental results.
The implant contains a microfluidic system that precisely delivers drugs to a targeted region of the brain, as well as a micro-LED that enables optical control of specific neurons. Drug delivery and optical stimulation can be controlled independently, or the two functions can be combined.
The drug reservoir is designed to be magnetically detachable. Even after the drug is depleted, researchers can replace or refill the reservoir without the need for additional implantation surgery, enabling long-term, repeated experiments.
The research team implanted the device in rats and verified its performance over a four-week period. In particular, a researcher in Chicago successfully operated the brain implant in Daejeon, Korea, in real time via the internet, demonstrating that the device can operate reliably over intercontinental distances.
The team also conducted an experiment in which cocaine was wirelessly administered to a rat's brain while specific neurons were simultaneously stimulated with light. The results showed that addiction-related behavioral responses could be suppressed, demonstrating the potential of combining drug delivery and optical stimulation for neural circuit research.
By eliminating the need for researchers to operate equipment directly beside experimental animals, this technology enables long-term studies of the relationship between brain circuits and behavior under naturalistic conditions. It is expected to be useful for studying conditions that involve long-term changes in neural circuit function and behavior, such as addiction, depression, and neurodegenerative diseases.
The technology could ultimately pave the way for intelligent implantable medical devices that combine brain-state sensing with AI to deliver drugs or neural stimulation precisely when needed.
Professor Jae-Woong Jeong from KAIST said, “This technology transforms wireless brain implants that use light and drugs from short-range control tools into IoT-based brain engineering platforms capable of long-term, automated, and remote experimentation.” He added, “In the long term, it could contribute to the development of intelligent implantable medical devices for the diagnosis and treatment of brain disorders.”
Professor Wha Young Kim from Yonsei University said, “This platform allows researchers to remotely and precisely control specific brain circuits over extended periods while animals move freely under naturalistic conditions.” She added, “It is expected to become an important tool for identifying causal relationships between neural circuits and behavior in disease models such as addiction, depression, and neurodegenerative disorders.”
Eun Young Jeong, a doctoral student in KAIST's School of Electrical Engineering, and Jong Woo Park, a doctoral student at Yonsei University College of Medicine, served as co-first authors. The study was published on July 29 in the international journal Science Advances.
Paper title: IoT-enabled wireless neural implant for chronic, programmable neuropharmacology and optogenetics,
DOI: 10.1126/sciadv.aee8648
This research was supported by the Mid-Career Researcher Program and Basic Research Laboratory Program of the National Research Foundation of Korea, funded by the Ministry of Science and ICT, as well as the Industrial Technology Alchemist Project of the Ministry of Trade, Industry and Energy.
2026.08.27 View 2091 -
KAIST Solves 3D Memory Reliability Problem with "Oxygen Tunnel" Structure, Boosting AI Chip Performance and Reducing Power Consumption
As AI systems become more advanced, memory is required to transfer larger amounts of data at higher speeds. But conventional planar semiconductor scaling is running out of room. A KAIST research team has now addressed a key weakness in three-dimensional, vertically stacked memory devices, opening a new path to faster, more power-efficient AI semiconductors.
KAIST (President Choongsik Bae) announced on August 25 that a research team led by Professor Jimin Kwon from the School of Electrical Engineering has developed a new multilayer interlayer dielectric structure that reduces defects and significantly enhances the performance of oxide vertical channel transistors (VCTs), a next-generation memory device. The study was conducted in collaboration with researchers from UNIST, Yonsei University, and other Korean institutions.
DRAM, which serves as the main memory in computers, has advanced over the past several decades by scaling down device size while reducing power leakage. More recently, vertical channel structures, in which current flows vertically, have become a key technology for increasing memory density.
The challenge is oxygen vacancies — defects caused by the absence of oxygen atoms in the oxide semiconductor — which destabilizes the material's electrical properties. But oxygen cannot simply be supplied without limit: when oxygen is supplied to suppress oxygen vacancies, some of the oxygen tends to migrate further, reaching the metal electrode and oxidizing it, which degrades device performance instead. The channel needed oxygen; the electrode did not. Therefore, selectively controlling oxygen flow became a key challenge.
The KAIST team developed a new multilayer interlayer dielectric consisting of silicon nitride/silicon dioxide/silicon nitride (SiN/SiO₂/SiN), engineered to function as an "oxygen tunnel" that steers oxygen selectively toward the channel while blocking its path to the electrode. The structure enabled stable compensation of oxygen vacancies in the oxide semiconductor while simultaneously suppressing unwanted oxidation at the electrode, thereby resolving the trade-off.
As a result, the researchers achieved world-class current density and data retention time in oxide vertical channel transistors.
The device also demonstrated outstanding operational stability. Even after more than ten million cycles of harsh electrical stress testing, the threshold voltage shift remained below 50 millivolts (mV), confirming its high reliability as a memory device.
The team further evaluated system-level performance by integrating conventional silicon CMOS technology with the new oxide semiconductor platform. The results suggest that this approach could significantly improve the performance of next-generation compute-in-memory (CIM) systems, intelligent semiconductors that perform AI computation directly inside memory.
Hyeonho Gu, the first author of the study, said, “This research is significant because it goes beyond improving memory density and addresses the long-standing instability problem in 3D devices through a new approach based on oxygen migration control.” He added, “We expect this technology to play a key role in accelerating the commercialization of ultra-low-power, high-performance compute-in-memory systems required for the AI era.”
This study was led by KAIST researcher Hyeonho Gu as the first author and was published on May 20 in Advanced Functional Materials, a leading international journal in materials science. The paper was also selected as a Front Cover article in recognition of its academic significance and originality.
Paper title: Oxygen-Tunnel Indium Tin Oxide Vertical Channel Transistors with Enhanced Current Density and Reliability for Monolithic 3D Compute-In-Memory Systems
DOI: https://doi.org/10.1002/adfm.202531989
Author information: Hyeonho Gu (KAIST, first author); Yongwoo Lee (KAIST, corresponding author); Haksoon Jung (KAIST, corresponding author); Jimin Kwon (KAIST, corresponding author); Hoichang Jeong (UNIST, co-author); Yanfeng Zhao (UNIST, co-author); Heesoo Yang (UNIST, co-author); Minho Park (UNIST, co-author); Hyeonjin Lee (UNIST, co-author); Seunghun Baek (UNIST, co-author); Minju Song (UNIST, co-author); Junghwan Kim (UNIST, co-author); Youngmin Jo (KAIST, co-author); Hyunjin Park (Korea Research Institute of Chemical Technology, co-author); Munhyeon Kim (Seoul National University of Science and Technology, co-author); Jae-Joon Kim (Seoul National University, co-author); Kyuho Jason Lee (Yonsei University, co-author); and Byungjo Kim (UNIST, co-author).
This research was supported by the National Semiconductor Laboratory Program and the Excellent Young Researcher Program of the National Research Foundation of Korea, funded by the Ministry of Science and ICT; the Broadcast and Telecommunications Industry Technology Development Program of the Institute of Information & Communications Technology Planning & Evaluation; and the Super Gap Technology Development Program of the Korea Evaluation Institute of Industrial Technology, funded by the Ministry of Trade, Industry and Energy.
2026.08.20 View 1025 -
KAIST Develops ‘Chameleon AI Semiconductor’ with Programmable Response Speeds
AI semiconductors are becoming more programmable. KAIST researchers have developed a device whose response characteristics can be programmed to process data changing at different speeds. The technology reduced prediction errors for time-varying data by up to 40-fold and is expected to enhance real-time AI performance in autonomous vehicles, robots, and wearable devices.
KAIST (President Choongsik Bae) announced on August 7 that a research team led by Chair Professor Shinhyun Choi from the School of Electrical Engineering and the Graduate School of Semiconductor Technology has developed a programmable dynamic memtransistor (PDM), a semiconductor device whose time-response characteristics can be adjusted to multiple states and retained, as well as an integrated array based on the device.
A memtransistor is a next-generation semiconductor device that combines the information-storage function of memory with the computing function of a transistor. In the developed PDM, the ability to process data while retaining previous information allows its response characteristics to be adjusted and retained for incoming data.
Today’s computers and smartphones require complex software processing to analyze data that changes over time, resulting in large computational loads and high power consumption. To address this, researchers have been studying technologies that allow semiconductor hardware itself to process data directly. However, conventional devices have had fixed response speeds that cannot be changed once the device is fabricated.
The research team overcame this limitation by introducing a dual-layer structure inside the transistor, combining a charge storage layer that accumulates and processes data with an electron trapping layer that controls the response speed in a nonvolatile manner.
In the PDM developed by the research team, incoming data is processed in the charge storage layer, while the electron trapping layer controls, across multiple levels, the recovery speed at which the semiconductor returns to its original state. In experiments, the team succeeded in tuning the current recovery time over an approximately 5-fold range and the characteristic frequency over a range of more than 10-fold.
In particular, in experiments involving the prediction of data in which fast and slow changes are intricately mixed, the PDM reduced prediction errors by as much as 40 times compared with conventional fixed-response semiconductor devices. The PDM enables accurate information processing even when handwriting or object-movement speeds vary, by using response characteristics configured to match different input timescales. Once the response characteristics are set, the device remembers them without requiring a continuous external power supply, and it does not require complex preprocessing of input data. Because it is fully compatible with materials used in widely adopted commercial semiconductor processes, it is also highly advantageous for mass production and commercialization.
The research team fabricated a PDM array and used it to predict complex data, confirming that it achieved accuracy comparable to conventional software-based systems while consuming far less energy.
“This study demonstrates an AI semiconductor whose response characteristics can be programmed to efficiently process data changing at different speeds,” said Chair Professor Choi. “We expect it to become a core technology that improves the performance of AI devices such as autonomous vehicles, robots, and wearables while reducing their power consumption.”
This research was led by KAIST Graduate School of Semiconductor Technology Ph.D. candidate Dae-won Kim as the first author, with Yoonho Cho, Seokho Seo, Yujin Kim, See-On Park, Taehwan Jang, and Chaebin Park participating as co-authors. Young Taek Oh and Fellow Jae-Duk Lee of Samsung Electronics’ Semiconductor R&D Center also participated as co-authors, and Chair Professor Shinhyun Choi served as the corresponding author. The research was published in July in the internationally renowned journal Nature Communications on July 4.
Paper title: Programmable memtransistor array with temporal dynamics modulation for efficient time-series data processing,
DOI: https://doi.org/10.1038/s41467-026-75211-5
This research was supported by the National R&D Program through the National Research Foundation of Korea funded by the Ministry of Science and ICT, the ETRI R&D Support Program of the Institute of Information & Communications Technology Planning & Evaluation, the HRD Program for Industrial Innovation of the Korea Institute for Advancement of Technology funded by the Ministry of Trade, Industry and Energy, Samsung Electronics, and others.
2026.08.07 View 1952 -
KAIST Develops AI That Avoids Hallucinating Even at Night or in Smoke
Multimodal large language models (MLLMs), which process multiple types of sensory information such as text, images, and audio at the same time, are rapidly expanding the range of applications for artificial intelligence (AI). However, in real-world environments, these models can misinterpret the physical characteristics of sensors, mistakenly identify objects, or claim to hear sounds that are not actually present simply because a certain object appears in a video. These errors are known as hallucinations. A KAIST research team has developed a new technology that corrects such information confusion and physical misperceptions in AI.
KAIST (President Choongsik Bae) announced on the 31st of July that a research team led by Professor Yong Man Ro from the School of Electrical Engineering has developed two core technologies that overcome the tendency of existing large language models to rely too heavily on ordinary camera (RGB) images and enable AI to suppress cross-modal hallucinations that occur when different sensory inputs become mixed.
The first technology developed by the research team is the Diverse Negative Attributes (DNA) optimization method, which helps AI accurately understand the physical characteristics of special camera sensors such as thermal, depth, and X-ray sensors. Existing AI models often failed to understand the physical meaning of such images, for example by mistaking bright areas in thermal images for simple light reflection.
The research team built VS-TDX, the first comprehensive benchmark for evaluating diverse vision sensors, and used the types of wrong answers that AI frequently produces as learning signals to help the model internalize the characteristics of each sensor. As a result, the AI gained a “new eye” that allows it to accurately infer the state of objects even in darkness or smoke.
The second technology is Modality-Adaptive Decoding (MAD), a control method that blocks hallucinations caused by confusion between visual and auditory information at the source. This technology prevents AI from mistakenly claiming that it hears a sound that does not actually exist simply because a certain object appears in a video.
MAD works by having the AI self-assess whether vision or audio is more important for a given task, and then increasing the weight of the more relevant modality in real time. A key advantage of this technology is that it can immediately suppress hallucination errors without costly model retraining, as it is training-free.
Instead of retraining AI models at large scale with massive computing resources, the research team maximized cost efficiency by introducing the DNA method, which enables fine adjustment with only a small amount of data, and the
MAD plug-in approach, which requires no additional training at all.
These technologies can be applied to autonomous vehicles operating at night or in bad weather, robots performing missions in smoke-filled environments, and unmanned aerial vehicles using thermal cameras. They are also expected to be useful in fields that process multiple types of sensor information together, such as airport X-ray security screening and medical image analysis.
Professor Yong Man Ro said, “This research is significant because it reduces AI’s sensory bias and misperceptions without large-scale retraining,” adding, “It will serve as a foundation for building multimodal AI that can be trusted in real-life and industrial settings.”
This achievement was notable for its continuity, with Sangyun Chung, a doctoral student in KAIST’s School of Electrical Engineering, participating as first author in both studies. Dr. Youngjun Yoo also participated as co-first author in the DNA study.
Among the related papers, the MAD study was presented in June at the Conference on Computer Vision and Pattern Recognition (CVPR), the world’s leading international conference in AI and computer vision. The DNA study was published in IEEE Transactions on Image Processing, a leading international journal in the field of image processing.
Paper title: Enhanced Vision-Language Models for Diverse Sensor Understanding: Cost-Efficient Optimization and Benchmarking,
DOI: 10.48550/arXiv.2412.20750 Author information: Sangyun Chung (KAIST, co-first author), Youngjun Yoo (KAIST, co-first author), Se Yeon Kim (KAIST, third author), Youngchae Chee (KAIST, fourth author), Yong Man Ro (KAIST, corresponding author)
Paper title: MAD: Modality-Adaptive Decoding for Mitigating Cross-Modal Hallucinations in Multimodal Large Language Models,
DOI: 10.48550/arXiv.2601.21181
Author information: Sangyun Chung (KAIST, first author), Se Yeon Kim (KAIST, second author), Youngchae Chee (KAIST, third author), Yong Man Ro (KAIST, corresponding author)
Related demo video: https://youtu.be/VuP9i6Vfk8o
This research was supported by the Institute of Information & Communications Technology Planning & Evaluation’s (IITP’s) Human-Centered AI Core Technology Development Program and by a Center for Applied Research in Artificial Intelligence (CARAI) grant funded by the Defense Acquisition Program Administration (DAPA) and the Agency for Defense Development (ADD).
2026.08.03 View 1349 -
KAIST Develops AI That Finds Its Own Hidden Weaknesses, Paving the Way for Safer Generative AI Models
KAIST researchers have developed a safety verification technology that uncovers roughly seven times more hidden vulnerabilities in AI than existing methods. The technology is expected to serve as a foundation for developing safer, more trustworthy AI.
KAIST (President Choongsik Bae) announced on the 30th of July that a research team led by Professor Junmo Kim from the School of Electrical Engineering has developed a new framework called Stable-GFlowNet (S-GFN), which overcomes the limitations of red-teaming—a safety verification process that deliberately attacks large language models (LLMs) to expose hidden weaknesses.
Red-teaming for generative AI is the process of crafting attack prompts designed to probe an AI's vulnerabilities and induce the AI to produce harmful or dangerous responses before the program is deployed. Since discovering a wider variety of attack methods allows more vulnerabilities to be addressed in advance, both the success rate and diversity of attacks are critical.
Previous approaches primarily relied on reinforcement learning—an AI technique trained to maximize reward—to generate attack prompts. However, these methods frequently suffered from mode collapse—a phenomenon where the model repeatedly converges on a narrow set of high-reward attack prompts rather than generating diverse outputs, thereby limiting its ability to uncover various vulnerabilities.
Generative Flow Networks (GFlowNets)—an AI generation technique trained to produce diverse outputs in proportion to their reward—were proposed as a solution. Yet GFlowNet training is computationally complex and unstable, and noisy reward signals can assign high rewards even to meaningless sentences, often causing training to collapse.
To address these issues, the research team developed three core techniques that help the model learn effective attacks more reliably while filtering out flawed ones.
First, much like comparing several paths to choose the best one, the team introduced Contrastive Trajectory Balance (CTB), which reduces computational complexity and stabilizes training by directly comparing pairs of generated attack trajectories.
Second, akin to filtering out background noise to focus on a single voice, the team applied Noise Gradient Pruning (NGP) to eliminate minor reward fluctuations and ensure the model learns exclusively from meaningful signals.
Third, the team applied the Min-K Fluency Stabilizer (MKS), which guides the model to generate attack prompts resembling text that a real user would write—just as a human reader naturally prefers coherent sentences to gibberish.
As a result, Stable-GFlowNet discovered 134 unique attack types—about seven times more than the 17 unique attack types found by the existing GFlowNet-based technique—while maintaining a high attack success rate of 92%.
Defense models trained using attacks generated by Stable-GFlowNet also demonstrated strong generalization, effectively defending against a wide range of attacks in cross-attack tests, which evaluate performance using attack techniques different from those used during training.
The team further demonstrated that CTB and NGP achieve faster and more stable performance than existing methods—not only in AI safety verification, but also in other distribution-matching tasks such as molecular generation for drug discovery.
Professor Kim said, "This technology is significant in that it can reliably uncover a wide range of AI vulnerabilities even in realistic conditions with limited data and high noise." He added, "Because it allows a broader range of risk factors to be identified and defended against before generative AI is deployed in real-world services, we expect it to become a core foundational technology for developing safer, more trustworthy AI."
The study was led by first author Minchan Kwon, a Ph.D. candidate from the School of Electrical Engineering, and was selected as a Spotlight paper—placing it in the top 2.2% of submissions—at the International Conference on Machine Learning (ICML) 2026, one of the world's most prestigious AI conferences.
※ Paper title: Stable-GFlowNet: Toward Diverse and Robust LLM Red-Teaming via Contrastive Trajectory Balance
arXiv: https://arxiv.org/abs/2605.00553
This research was supported by the Institute of Information & Communications Technology Planning & Evaluation’s (IITP) SW Star Lab program, funded by the Ministry of Science and ICT.
2026.07.30 View 1560 -
KAIST Opens a New Era of Webtoons: From “Viewing” to “Experiencing”
Webtoons are coming to life in the physical world, ushering in a new era in which comics are not merely viewed, but experienced.
A KAIST research team has developed the world’s first next-generation extended reality (XR) comics platform that enables a wide range of readers to enjoy immersive, three-dimensional comics in physical space. By expanding webtoons beyond the screen and into the real world, the team has opened up new possibilities for the future of comics.
KAIST (President Choongsik Bae) announced on the 21st of July that a research team led by Professor Ian Oakley from the School of Electrical Engineering has proposed core design principles and future directions for next-generation extended reality (XR) comics through a systematic user study involving 15 participants, including human-computer interaction (HCI) experts, professional webtoon creators, and readers.
The research team developed ComiXR, a new platform that enables users to both read and create comics in XR environments. Participants used the platform to transform a conventional print comic into an XR comic and explored how different visual, auditory, haptic, and interactive features could be combined.
Comics, which originated in printed books and newspapers, have evolved dramatically with the rise of smartphones. The vertical-scrolling format of webtoons has become particularly successful by adapting comics to the interaction methods of mobile devices.
The research team viewed XR devices as a potential next stage in this evolution. To explore how spatial depth, three-dimensional rendering, spatial audio, eye tracking, and facial expression tracking could be incorporated into comics, the team built ComiXR using a Meta Quest Pro headset.
While wearing the headset, participants freely positioned 3D characters, speech bubbles, sound effects, and other comic elements throughout a physical room. They were able to construct comic environments that they found comfortable, engaging, and immersive.
The results showed that readers strongly preferred designs that actively used the depth of physical space over simply displaying flat comic pages in a virtual environment. Immersion increased significantly when characters were positioned at a different depth from the background and speech bubbles were separated into distinct layers. In particular, an eye-tracking feature that revealed the next line of dialogue only when the reader looked at a specific character proved effective in preventing spoilers.
The platform also demonstrated new sensory experiences that are not possible in conventional comics. Special effects could be triggered in response to readers’ facial expressions, while haptic feedback could convey sensations such as a character’s heartbeat or the impact represented by an onomatopoeic effect.
Based on the study, the research team also proposed four key design concepts for XR comics. The first, “The Panel Gallery,” transforms the walls of a room into a gallery for displaying comic panels. The second, “The Pop-Up,” presents comics like pop-up books on desks or walls. The third, “Around Comic,” places 3D characters and other comic elements in outdoor spaces. The fourth, “Inclusive ComiX,” improves accessibility for a wide range of readers.
The research team expects XR comics to complement, rather than replace, existing smartphone-based webtoons. They could be used for special exhibitions and educational content that allow audiences to experience fictional worlds more vividly, as well as platforms that improve access to cultural content for a wider range of users.
Ammar Al-Taie, a postdoctoral researcher at the KAIST Information and Electronics Research Institute, participated as the first author, while Hyunyoung Han, a doctoral student in the School of Electrical Engineering, participated as a co-author.
The research was presented at the ACM Designing Interactive Systems Conference 2026, or ACM DIS 2026, one of the leading international conferences in human-computer interaction and design. The ComiXR platform has also been released as open-source software for public use.
Paper title: ComiXR: Exploring Comic Layouts in eXtended Reality
DOI: https://doi.org/10.1145/3800645.3812857
Related Video: https://drive.google.com/drive/folders/1D9Efp3T0biDbUm1K5Gu6HLSSy89Uaq8R?usp=sharing
Open-source platform: https://github.com/ammarjamal/ComiXR
The research was supported by the KAIST Jang Young Sil Fel¬lowship Program (Excellence Track). The authors acknowledge support from the IITP (Institute of Information & Communications Technology Planning & Evaluation)-ITRC (Information Technology Research Center) grant funded by the Korean government (Ministry of Science and ICT) (IITP-2026-RS-2024-00436398).
2026.07.21 View 1869 -
KAIST Develops Key Technology to Make Personalized AI Safer
“Create an AI assistant trained only on our company’s documents.”
The era of building “personalized AI” by training AI models on individual or corporate documents and data is beginning. However, while such customization can improve task performance, it can also weaken the model’s existing safety safeguards. KAIST researchers have developed a core AI technology that preserves customized performance while further strengthening safety.
KAIST (President Choongsik Bae) announced on the 15th of July that a research team led by Professor Changick Kim from its School of Electrical Engineering has developed “Buffer-and-Reinforce,” a training framework for safe fine-tuning that prevents safety degradation when large language models (LLMs), such as
ChatGPT, are retrained on data from individuals or companies to better suit their needs.
Until now, one of the biggest challenges in the era of personalized AI has been that fine-tuning improves a model’s ability to perform new tasks, but can also weaken its existing safety rules. The research team focused on prior findings showing that, counterintuitively, fine-tuning an AI model while it is in a temporarily jailbroken state — a state in which it may respond even to dangerous requests it would normally refuse — does not significantly compromise its safety.
The team then devised a new approach in which this jailbroken state is not used in actual services, but is applied only temporarily during the fine-tuning process through a buffering module called “BufferLoRA,” which is removed after training.
The research team was the first to clarify why this phenomenon occurs. They found that, in the temporarily jailbroken state, the AI model becomes less easily influenced by harmful information, while still effectively learning the new task abilities desired by the user. In other words, the model can continue learning useful knowledge without additionally absorbing harmful behaviors.
Based on this insight, the team developed a two-stage learning method consisting of “buffering” and “safety reinforcement.”
First, the temporary buffering module, BufferLoRA, is applied to the AI model during user fine-tuning, where it acts as a protective layer that prevents harmful data from directly affecting the base model. Once fine-tuning is complete, this module is removed.
Next, a safety reinforcement module called “ReinforceLoRA” is applied to restore and strengthen the model’s safety. In this process, the team used QR decomposition, a mathematical technique that separates different types of information and selectively reflects only the necessary components. This allowed the model to retain the new functions learned from user data while selectively reinforcing safety.
Simply put, the researchers first placed a temporary protective layer, BufferLoRA, over the AI model so that harmful data could not directly affect it, while allowing the model to learn the necessary task. They then removed the protective layer and applied ReinforceLoRA to strengthen the model’s safety safeguards. As a result, the model maintained its customized performance while achieving even stronger safety.
In experiments, the AI model maintained high safety even in an extreme setting where all user data consisted of harmful questions and answers. After fine-tuning, the rate at which the AI generated harmful responses was about 8%, lower than the roughly 18% observed in the original model that had not been fine-tuned at all. The framework also achieved strong customized performance and state-of-the-art safety without requiring additional safety data during user fine-tuning or significantly increasing computational cost, suggesting its practical applicability to real-world personalized AI services.
Professor Changick Kim stated, “This research provides a key foundational technology that allows anyone to build customized AI with their own data while using it more safely,” adding, “We expect it to contribute significantly to building a trustworthy AI service environment in the era of personalized AI and AI agents.”
This research was led by Seokil Ham, a doctoral student in KAIST’s School of Electrical Engineering, as first author. The paper was selected as a Spotlight presentation at the International Conference on Machine Learning (ICML) 2026, one of the world’s most prestigious conferences in artificial intelligence, an honor given to only about the top 2.2% of all submitted papers, drawing international attention.
※ Paper title: Jailbreak to Protect: Buffering and Reinforcing via Temporary Jailbreaking for Safe Fine-Tuning in Large Language Models
DOI: 10.48550/arXiv.2605.24550
※ Author information: Seokil Ham (KAIST, first author), Jaehyuk Jang (KAIST, second author), Wonjun Lee (KAIST, third author), Changick Kim (KAIST, corresponding author)
※ Related video: https://drive.google.com/file/d/1gfok06dE8699qtiUR7gVsRoVmBGADaWQ/view?usp=sharing
This work was supported by Institute of Information & Communication Technology Planning & Evaluation (IITP) grant funded by Ministry of Science and ICT(MSIT) (No. RS-2025-02215344, Development of AI Technology with Robust and Flexible Resilience Against Risk Factors).
2026.07.15 View 1540 -
KAIST Automates the Search for “Dream Semiconductor” 2D Semiconductors
The era of researchers manually searching for two-dimensional semiconductors, which are drawing attention as next-generation AI semiconductors, is coming to an end. KAIST researchers have automated semiconductor screening and device fabrication, analyzed thousands of devices, and revealed the relationship between thickness and performance that had long been difficult to identify. This achievement is expected to shift next-generation semiconductor research toward a data-driven approach and accelerate the commercialization of AI semiconductors and ultra-low-power semiconductors.
KAIST (President Choongsik Bae) announced on the 9th that a research team led by Professor Jimin Kwon of the School of Electrical Engineering and the Department of AI System has developed a technology that automatically identifies two-dimensional semiconductors from optical microscope images alone and connects the process to transistor fabrication, through joint research with UNIST, Hanbat National University, Hanyang University, and Washington University in St. Louis in the United States.
Two-dimensional semiconductors are ultrathin semiconductors only a few atomic layers thick. They are called “dream semiconductors” because they can enable smaller semiconductors that consume less electricity than conventional silicon semiconductors. Today’s silicon semiconductors are approaching physical limits, as continued miniaturization of circuits leads to greater power loss and heat generation. Two-dimensional semiconductors, which are attracting attention as next-generation materials to overcome these limits, are expected to be used in a wide range of future technologies, including AI semiconductors, smartphones, data centers, wearable devices, foldable or stretchable electronics, and ultra-small medical sensors.
However, in two-dimensional semiconductors made through solution processing, the position, size, and thickness of each small semiconductor flake all differ, requiring researchers to find the desired samples one by one under a microscope. They then had to manually design electrodes according to the identified positions, requiring substantial time and effort, and making it practically difficult to analyze thousands or more devices at once.
The research team used molybdenum disulfide (MoS₂), a representative two-dimensional semiconductor material. By using the fact that the RGB red, green, and blue brightness values seen under a microscope change depending on thickness, the team enabled a computer to automatically identify the desired semiconductor and automatically design the electrodes. Verification using atomic force microscopy (AFM) confirmed that even subtle thickness differences of three to eight layers could be accurately distinguished.
Through this approach, the team successfully selected suitable samples automatically from more than 120,000 semiconductor flakes and fabricated and analyzed 1,615 transistors.
The large-scale analysis also produced meaningful results. The team statistically clarified for the first time that as the semiconductor becomes thicker, current flows more easily, but the ability to switch electricity on and off actually decreases. This characteristic had been difficult to confirm previously because only a small number of samples could be analyzed, but the team revealed it through large-scale data.
The greatest significance of this study is that it did not simply automate the fabrication process, but transformed two-dimensional semiconductor research, which had relied on human experience, into data-driven research. Going forward, the technology is expected to enable researchers to fabricate and analyze more semiconductors more quickly, identify high-performance materials, and ultimately expand into research in which AI designs new semiconductors.
This study was conducted with Professor Jimin Kwon, Dr. Haksoon Jung, and Dr. Yongwoo Lee of KAIST as co-corresponding authors, and Sanghyun Lee of UNIST as the first author. The research results were published on April 3 in Advanced Functional Materials, a leading international journal in materials science, and were also selected as an Inside Back Cover article in the field of 2D Materials & Electronics.
※ Paper title: Statistically Resolving Thickness-Dependent Electrical Characteristics in Multilayer-MoS₂ Transistors, DOI: 10.1002/adfm.202532204
※ Author information: Professor Jimin Kwon (KAIST, corresponding author), Dr. Haksoon Jung (KAIST, corresponding author), Dr. Yongwoo Lee (KAIST, corresponding author), Sanghyun Lee (UNIST, first author), and participating researchers from partner institutions: Sumin Hong (UNIST), Minho Park (UNIST), Professor Seongju Kim (Hanbat National University), Professor Sang-Hoon Baek (Hanyang University), Professor Joonki Suh (KAIST), Seonguk Yang (KAIST), Professor Sang-Hoon Bae (Washington University in St. Louis), and Dr. Chang-Soo Lee (TDS)
This research was supported by the Individual Basic Research Program of the National Research Foundation of Korea (NRF), funded by the Ministry of Science and ICT (MSIT), and by the Advanced Strategic Industry Super-Gap Technology Development Program of the Korea Planning & Evaluation Institute of Industrial Technology (KEIT), funded by the Ministry of Trade, Industry and Energy (MOTIE).
2026.07.09 View 1908 -
KAIST Develops Core Display Technology That Prevents Image Distortion Even When Stretched
Beyond bendable and foldable displays, the era of stretchable displays, whose screens can expand freely like rubber, is now emerging. KAIST researchers have developed a core technology that allows text, images, and other on-screen information to retain their original shape even when the screen is stretched by up to 15%. The achievement is expected to help solve the problem of image distortion and accelerate the commercialization of next-generation high-quality stretchable displays.
KAIST (President Choongsik Bae) announced on the July 8 that a research team led by Professor Seunghyup Yoo of the School of Electrical Engineering, in collaboration with Professor Hanul Moon’s team at Dong-A University (President Hae Woo Lee), has successfully implemented an auxetic-based stretchable display platform. Auxetic structures expand in both width and length when pulled, allowing the display to stretch uniformly at the same ratio in all directions without distorting the image on the screen.
Conventional stretchable displays are generally made by forming light-emitting devices on a stretchable substrate, which serves as the base layer of the display. However, when such a substrate is stretched in one direction, it tends to shrink in the opposite direction, causing letters and images on the screen to become flattened or distorted. Auxetic structures have been used to address this problem, but most previous approaches were limited to maintaining the overall horizontal-to-vertical ratio of the screen, while the letters and images within the screen still remained vulnerable to distortion.
Instead of bonding the auxetic structure and the stretchable substrate across the entire surface, as in conventional methods, the research team proposed a new design approach that uses computational analysis to selectively connect only the necessary points that ensure isotropic expansion throughout the substrate.
In the conventional approach, the twisting deformation that occurs as the auxetic structure stretches is directly transferred to the substrate, distorting the image inside the screen. In contrast, the platform developed by the research team was designed so that each region moves evenly outward from its original position. This allows not only the entire screen but also small areas such as letters and images to expand together while maintaining their original shapes.
The research team verified the platform’s performance by repeatedly stretching a substrate patterned with letters and images in both the horizontal and vertical directions. In the conventional method, the patterns underwent local deformation, whereas in the new platform, the shapes of the letters and images remained intact. This demonstrates that not only the whole screen but also fine images on-screen can expand uniformly without distortion.
The team also integrated an LED array, a structure in which multiple LEDs are arranged at regular intervals, onto the platform to verify its performance as an working stretchable display. Even when stretched by up to 15% in both the horizontal and vertical directions, stable electrical operation and the screen brightness were maintained. After repeated stretching to 15%, the decrease in brightness remained below 2%, confirming the platform’s potential for practical display applications.
This technology is expected to serve as a core platform for next-generation electronics with freely changeable shapes, including wearable electronic devices, electronic skin, or e-skin, which refers to electronic devices that stretch like skin while sensing and displaying information, medical biosensors, soft robots, and curved displays for automobiles and aircraft.
Professor Seunghyup Yoo of KAIST said, “For stretchable displays to be used as actual information display devices, they must not only stretch well, but also preserve on-screen information accurately during stretching,” adding, “This platform enables uniform expansion from small areas of the screen to the entire display, and will serve as a key foundational technology for accelerating the commercialization of high-quality stretchable displays.”
This study was led by KAIST Dr. Su-Bon Kim and Dr. Junho Kim as co-first authors, with Professor Hanul Moon of Dong-A University and Professor Seunghyup Yoo of KAIST as co-corresponding authors. The research was published in the international journal Nature Communications on June 10.
※ Paper title: Hybrid auxetic metamaterial platforms enabling multiscale isotropic expansion for distortion-free stretchable displays, DOI: 10.1038/s41467-026-74141-6
This research was supported by the National Research Foundation of Korea (NRF) Mid-Career Researcher Program, the Future Display Strategic Research Laboratory Program, the Korea Planning & Evaluation Institute of Industrial Technology (KEIT), and the Korea Institute for Advancement of Technology (KIAT) HRD Program.
2026.07.08 View 1881 -
KAIST Identifies the “Hidden Energy Cost” of AI Agents for the First Time
As the era of AI agents—systems that can reason and act autonomously—begins, the power consumption of data centers is emerging as a critical challenge. A KAIST research team has, for the first time, analyzed the computational cost and energy consumption of AI agents, finding that they can consume up to 136.5 times energy per query than conventional generative AI. The study shows that competitiveness in the AI era is expanding beyond model performance to include the efficiency of data centers and power infrastructure.
KAIST announced that a research team led by Professor Minsoo Rhu of the School of Electrical Engineering has systematically analyzed, for the first time, how much computational resources and power AI agents require in real-world service environments.
Large language model (LLMs) powered applications such as ChatGPT have rapidly evolved beyond simply answering questions. They are now developing into AI agents: next-generation AI systems that can plan, use external tools such as web search, calculators, and code execution environments, and solve complex tasks by coordinating multiple steps on their own.
Although AI agents are increasingly being adopted in areas such as software development, research, and workplace automation, little has been known about the amount of electricity and operational cost required to run them in practice.
The research team defined AI agents not merely as software programs, but as a new type of workload that must be continuously processed by data-center servers and graphics processing units, or GPUs—high-performance chips used for large-scale AI computation. The team then analyzed the computational load and energy consumption incurred during actual AI agent execution.
The analysis found that AI agents perform, far higher volumes of LLM invocations than conventional chain-of-thought reasoning. Chain-of-thought, or CoT, refers to a method in which an AI model breaks down its reasoning process step by step to reach an answer, while an LLM invocation refers to each computational request made to a language model to generate a new judgment or response.
Because AI agents repeatedly call language models during execution, their response latency also increases significantly. The team found that response time can increase by up to 153.7 times, while GPUs remain idle for as much as 54.5 percent of the total execution time as external tools perform their tasks. In other words, as AI systems take on more complex tasks, a new form of inefficiency emerges in which expensive GPUs cannot be fully utilized.
The research team also analyzed the power consumption of AI agents at data-center scale. An AI agent using a 70-billion-parameter LLM—a scale comparable to current commercial AI services—consumed an average of 348.41 watt-hours per query. This is 136.5 times higher than the energy consumed by a conventional generative AI system performing simple question answering.
In addition, the team projected a future scenario in which 13.7 billion AI agent requests are generated per day — a volume equivalent to current Google search traffic. Under this scenario, data-center power demand would reach approximately 198.9 gigawatts, a level far exceeding the scale of AI data centers currently under development (which are in the range of a few gigawatts) and equivalent to roughly half of the average power consumption of the United States.
This study demonstrates that the focus of competition in the AI era is shifting from “smarter AI” to “optimally efficient AI.” Going forward, it will be essential not only to advance AI models, but also to jointly optimize AI semiconductors, data centers, and power infrastructure through co-design. Such an approach is expected to become a key strategy for reducing the operating cost of AI services and building sustainable AI infrastructure.
“This study is the first to quantitatively show not only how AI is becoming more intelligent, but also how much electricity and cost are required to implement and sustain that intelligence,” said Professor Rhu. “As AI agents become widespread, it will become increasingly important to take an integrated co-design approach that optimizes not only AI data-center infrastructure, but also AI agent models and power infrastructure.” He added, “Research and investment in this direction will be essential to dramatically reduce the cost for end users to access AI services while building sustainable AI infrastructure.”
The study was conducted with Jiin Kim, a Ph.D. student in the KAIST School of Electrical Engineering, as the first author. The paper was presented in February at the 32nd IEEE International Symposium on High-Performance Computer Architecture, or HPCA, one of the most prestigious international conferences in computer system design. The research team has also released the AI agent implementations and benchmarks used in the paper as open source to support follow-up studies by researchers worldwide.
Paper title: “The Cost of Dynamic Reasoning: Demystifying AI Agents and Test-Time Scaling from an AI Infrastructure Perspective”
Open-source repository: 10.1109/HPCA68181.2026.11408569
This research was supported by the Institute of Information & Communications Technology Planning & Evaluation (IITP) through the SW Starlab program, the K-Cloud Technology Development Program using AI semiconductors, and the Leading Technology Development Program for Advancing AI-Semiconductor-Based Data Centers, as well as by the Samsung Electronics Future Technology Incubation Center.
2026.07.07 View 2179