KAIST 利用脑电波检测人机认知失配，让 AI 实时修正行为

September 10, 2026 AI could soon infer human intent by sensing "that's not what I meant" through neural feedback by The Korea Advanced Institute of Science and Technology (KAIST) edited by Sadie Harley, reviewed by Robert Egan Sadie Harley Scientific Editor Meet our editorial team Behind our editorial process Robert Egan Senior Editor Meet our editorial team Behind our editorial process Editors' notes This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility: fact-checked trusted source proofread The GIST Add as preferred source Concept of neural value alignment because human behavior can be ambiguous, conventional AI can misunderstand both the goal a person intends and the preferred way of achieving it. Neural Value Alignment leverages two types of brain responses—reward prediction error (RPE) and state prediction error (SPE)—to help AI correct both the goal and the action simultaneously. Credit: KAIST A future in which AI can recognize a person's unspoken "that's not what I meant" response from brain signals and adjust its behavior on its own is coming closer. KAIST researchers have developed a technology that detects cognitive mismatch between humans and AI through brainwaves, enabling AI systems to revise their actions in real time according to human goals. The achievement is expected to accelerate the shift from AI that follows explicit commands to AI that can infer human intent. A research team led by Endowed Chair Professor Sang Wan Lee from the Department of Brain and Cognitive Sciences (Director of the Center for Neuroscience-Inspired Artificial Intelligence), in collaboration with Microsoft Research Asia (MSRA), developed Neural Value Alignment (NVA), a next-generation brain–computer interface (BCI) technology. The research is published in IEEE Transactions on Cybernetics. Where actions and goals diverge For AI to collaborate naturally with people, it must accurately understand what a person actually wants. Until now, however, AI systems have largely inferred human intent from externally observable information such as speech, actions or gestures. The challenge is that the same action can reflect different goals, and the same goal can also be achieved through different actions. For example, when a person picks up a cup, it may be unclear whether they intend to drink from it or hand it to someone else. Conversely, if the person's goal is simply to quench their thirst, that same goal could be pursued through several different actions—reaching for the cup, picking up a water bottle or asking someone else to bring a drink. Because of this two-sided "goal–action ambiguity," when AI misunderstands human intent, users often have to correct it through additional commands or actions. Reading the brain's mismatch signal To address this problem, the research team focused on the "prediction error" signals that arise unconsciously in the brain when a person encounters an unexpected situation. In simple terms, the team used the brain's instant "that's not what I meant" response when AI performs the wrong action or pursues the wrong goal. The researchers distinguished between two types of brain responses. The first is reward prediction error (RPE), which appears when AI misunderstands the person's ultimate goal. The second is state prediction error (SPE), which appears when the goal is correct but the process or method of action differs from what the person expected. The team measured real-time electroencephalography (EEG) signals from people as they observed AI performing tasks. They found that the brain produced different signals depending on whether the AI misunderstood the goal itself or chose the wrong method while pursuing the correct goal. The researchers also identified distinctive brainwave patterns that appeared when both types of errors occurred simultaneously. By applying deep learning to these brain signals, the team developed a technology that can determine, from EEG alone, how a person is interpreting the AI's behavior. In other words, even without a person saying "that's wrong," the AI can recognize whether the human brain is signaling that "the goal is wrong" or "the method is wrong." How the system corrects itself The team then proposed a Neural Value Alignment-based human–AI synergy algorithm, which feeds these decoded brain signals back to the AI in real time so it can correct its own behavior. When the AI detects an SPE signal, it interprets the situation as "the desired goal is correct, but the method is wrong" and adjusts its action strategy. When it detects an RPE signal, it understands that "the goal itself was misunderstood" and searches again for what the person truly intended. Simulation results showed that the proposed method adapted more quickly than existing approaches even in uncertain situations, such as when a person's goal suddenly changed or some human neural feedback signals were missing. From silent feedback to real-world use The key significance of this research is that it demonstrates the possibility of AI systems correcting themselves by reading a person's unconscious "that's not what I meant" brain response, without requiring the user to repeatedly say "do it this way" or "that's not right." As the technology advances, it could be applied to physical AI robots in homes and industrial settings, allowing them to understand user intent more naturally and adjust their actions accordingly. It could also be extended to autonomous vehicles that quickly reflect driver judgment, medical and rehabilitation robots for patients who have difficulty speaking or moving, and educational AI systems that adapt to a student's cognitive state. Distinct brain responses to RPE and SPE. The brain generates different EEG patterns depending on whether the AI selected the wrong goal (RPE) or took an unexpected action (SPE). It also shows the distinctive pattern that emerges when both types of errors occur simultaneously. Credit: KAIST AI adaptation when human goals change unexpectedly. Simulation shows how quickly different AI models adapt when a person's goal suddenly changes. The model using both RPE and SPE adapted faster and reduced mismatches more effectively than models relying on only one type of neural feedback or on observed behavior alone. Credit: KAIST A broader model for collaboration Ultimately, the study presents a new model of human–AI collaboration, moving beyond AI that acts only when explicitly instructed toward AI that can sense human responses and adjust itself accordingly. Lee, who led the international collaboration, said, "This research is meaningful because it shows that AI can move beyond inferring human intent only from visible behavioral outcomes and instead directly use cognitive signals generated in the brain during collaboration with AI." He added, "The technology can be expanded to a wide range of fields where human judgment and AI behavior must be closely connected, including physical AI, BCI, autonomous driving, precision personalized education, medical robotics, and human–computer interaction." Miran Lee, director of Microsoft Research Accelerator at Microsoft Research, said, "This achievement is the result of the ongoing international collaboration between KAIST and Microsoft Research Asia. We look forward to continuing this partnership to develop world-class BCI technologies that enable humans and AI to communicate and collaborate more naturally." The study's first author is Xin Xu, a Ph.D. student in KAIST's Department of Brain and Cognitive Sciences. Researchers from Microsoft Research Asia, including Yansen Wang, Dongqi Han and Dongsheng Li, also participated in the study. More information Xin Xu et al, Neural Value Alignment: Human–AI Collaboration Under Goal-Action Ambiguity, IEEE Transactions on Cybernetics (2026). DOI: 10.1109/tcyb.2026.3722605 Key concepts AI alignmentHuman-centered AI interfacesTrustworthy machine learningMachine learning methodologies Provided by The Korea Advanced Institute of Science and Technology (KAIST) Who's behind this story? Sadie Harley BSc Life Sciences & Ecology. Microbiology lab background with pharmaceutical news experience in oil, gas, and renewable industries. Full profile → Robert Egan Bachelor's in mathematical biology, Master's in creative writing. Well-traveled with unique perspectives on science and language. Full profile → Citation: AI could soon infer human intent by sensing "that's not what I meant" through neural feedback (2026, September 10) retrieved 10 September 2026 from https://techxplore.com/news/2026-09-ai-infer-human-intent-meant.html This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only. Brain-inspired AI: Human brain separates goals and uncertainty to enable adaptive decision-making Dec 29, 2025 AI co-pilot boosts noninvasive brain-computer interface by interpreting user intent Sep 1, 2025 Simplicity versus adaptability: Scientists propose AI method that integrates habitual and goal-directed behaviors Jun 14, 2024 Meta AI develops a non-invasive method to decode speech from brain activity Nov 28, 2023 AI reveals new insights into human brain activity Sep 10, 2024 Human-machine learning boosts noninvasive brain-computer control in untrained users Jul 15, 2026 AI may cut junior coding tasks and weaken the pipeline to senior developers 1 hour ago AI framework generates more realistic 4D LiDAR worlds for autonomous systems 1 hour ago Unexpectedly rigid polymers boost ion transport in hydrogen fuel cell membranes 2 hours ago Humanoid robot learns to sprint and perform spin kicks using AI trained on human motion data 7 hours ago 'Electrotech moneyball' framework developed to rank risk and opportunity in energy and AI supply chains 2 hours ago Training fix could let AI-generated scenes respond properly to a user's controls 4 hours ago