Geunsik Lim

Alexander K. Saeri, Jess Graham, Michael Noetel et al.

Artificial intelligence poses many risks, ranging from familiar present-day harms to unprecedented and potentially catastrophic ones. Effective risk management requires prioritization: we must understand which risks are most severe, who is most vulnerable, and who is most responsible for addressing them. We report results from a three-round Delphi study conducted late 2025 with 272 international AI experts. Experts rated 24 AI risks on harm probability and severity, sector and actor vulnerability, actor responsibility, and overall concern. Experts estimated the five most severe harms in the next 5 years were likely to come from dangerous capabilities, competitive dynamics, weapons & cyberattacks (including CBRNE), power centralization, and false information. In a business-as-usual scenario, experts judged 18 of 24 risks as having a more than 10% probability of catastrophic outcomes (e.g., more than 1 million deaths or more than USD 100B in financial loss) in the next 5 years (2025-2030). In a scenario where pragmatic mitigations are implemented, experts still judged five risks as having a more than 10% probability of catastrophic outcomes: dangerous capabilities, weapons & cyberattacks, environmental harm, inequality & unemployment, and power centralization. All 24 risks were judged as being more than 5% likely to cause catastrophic outcomes. AI users and the general public were judged the most vulnerable to these risks, but experts assigned the highest responsibility for addressing them to general-purpose AI developers and governance actors (including governments, regulators, and standards bodies). Across most risks, experts identified information, finance, and national security as the most vulnerable sectors. These findings can guide AI risk prioritization and clarify expert expectations about who should bear responsibility for mitigation.

MyungJoo Ham, Jijoong Moon, Geunsik Lim et al.

We propose NNStreamer, a software system that handles neural networks as filters of stream pipelines, applying the stream processing paradigm to deep neural network applications. A new trend with the wide-spread of deep neural network applications is on-device AI. It is to process neural networks on mobile devices or edge/IoT devices instead of cloud servers. Emerging privacy issues, data transmission costs, and operational costs signify the need for on-device AI, especially if we deploy a massive number of devices. NNStreamer efficiently handles neural networks with complex data stream pipelines on devices, significantly improving the overall performance with minimal efforts. Besides, NNStreamer simplifies implementations and allows reusing off-the-shelf media filters directly, which reduces developmental costs significantly. We are already deploying NNStreamer for a wide range of products and platforms, including the Galaxy series and various consumer electronic devices. The experimental results suggest a reduction in developmental costs and enhanced performance of pipeline architectures and NNStreamer. It is an open-source project incubated by Linux Foundation AI, available to the public and applicable to various hardware and software platforms.

MyungJoo Ham, Ji Joong Moon, Geunsik Lim et al.

We propose nnstreamer, a software system that handles neural networks as filters of stream pipelines, applying the stream processing paradigm to neural network applications. A new trend with the wide-spread of deep neural network applications is on-device AI; i.e., processing neural networks directly on mobile devices or edge/IoT devices instead of cloud servers. Emerging privacy issues, data transmission costs, and operational costs signifies the need for on-device AI especially when a huge number of devices with real-time data processing are deployed. Nnstreamer efficiently handles neural networks with complex data stream pipelines on devices, improving the overall performance significantly with minimal efforts. Besides, nnstreamer simplifies the neural network pipeline implementations and allows reusing off-shelf multimedia stream filters directly; thus it reduces the developmental costs significantly. Nnstreamer is already being deployed with a product releasing soon and is open source software applicable to a wide range of hardware architectures and software platforms.

Geunsik Lim

As climate-related hazards intensify, conventional early warning systems (EWS) disseminate alerts rapidly but often fail to trigger timely protective actions, leading to preventable losses and inequities. We introduce Climate RADAR (Risk-Aware, Dynamic, and Action Recommendation system), a generative AI-based reliability layer that reframes disaster communication from alerts delivered to actions executed. It integrates meteorological, hydrological, vulnerability, and social data into a composite risk index and employs guardrail-embedded large language models (LLMs) to deliver personalized recommendations across citizen, volunteer, and municipal interfaces. Evaluation through simulations, user studies, and a municipal pilot shows improved outcomes, including higher protective action execution, reduced response latency, and increased usability and trust. By combining predictive analytics, behavioral science, and responsible AI, Climate RADAR advances people-centered, transparent, and equitable early warning systems, offering practical pathways toward compliance-ready disaster resilience infrastructures.

Geunsik Lim, Yonghwi Kwon, Joonbae Park et al.

The two most significant bottlenecks in code merging are the build process and the unit tests. However, as the number of items to be checked in a code review increases, that code review becomes a bottleneck for code merging as well. Because of the dependency structure between code review services, an error in one service affects the entire service. As a result, whenever a service error occurs, it is crucial to have methods for determining which code review service has ultimately caused the error. With the goal of achieving a non-stop & non-breakable code review service, this paper describes an early error detection method along with a case study of the service.

Geunsik Lim, MyungJoo Ham, Jijoong Moon et al.

With the proliferation of IoT and edge devices, we are observing a lot of consumer electronics becoming yet another IoT and edge devices. Unlike traditional smart devices, such as smart phones, consumer electronics, in general, have significant diversities with fewer number of devices per product model. With such high diversities, the proliferation of edge devices requires frequent and seamless updates of consumer electronics, which makes the manufacturers prone to regressions because the manufacturers have less resource per an instance of software release; i.e., they need to repeat releases by the number of product models times the number of updates. Continuous Integration (CI) systems can help prevent regression bugs from actively developing software packages including the frequently updated device software platforms. The proposed CI system provides a portable and modular software platform automatically inspecting potential issues of incoming changes with the enabled modules: code format and style, performance regressions, static checks on the source code, build and packaging tests, and dynamic checks with the built binary before deploying a platform image on the IoT and edge devices. Besides, our proposed approach is lightweight enough to be hosted in normal desktop computers even for dozens of developers. As a result, it can be easily applied to a lot of various source code repositories. Evaluation results demonstrate that the proposed method drastically improves plug-ins execution time and memory consumption, compared with methods in previous studies.

Geunsik Lim, MyungJoo Ham, Jijoong Moon et al.

The complexity and size increase of software has extended the delay for developers as they wait for code analysis and code merge. With the larger and more complex software, more developers nowadays are developing software with large source code repositories. The tendency for software platforms to immediately update software packages with feature updates and bug-fixes is a significant obstacle. Continuous integration systems may help prevent software flaws during the active development of software packages, even when they are deployed and updated frequently. Herein, we present a portable and modular code review automation system that inspects incoming code changes such as code format and style, performance regression, static analysis, build and deployment tests, and dynamic analysis before merging and changing code. The proposed mechanisms are sufficiently lightweight to be hosted on a regular desktop computer even for numerous developers. The resulting reduced costs allow developers to apply the proposed mechanism to many source code repositories. Experimental results demonstrate that the proposed mechanism drastically reduces overheads and improves usability compared with conventional mechanisms: execution time (6x faster), CPU usage (40% lower), memory consumption (1/180), and no out-of-memory occurrence.