As AI shifts from leveraging information provided by humans to making decisions on their behalf, tech leaders must weave an intelligent data fabric to unlock the full potential of agentic AI while shoring up enterprise-wide resilience. Digital resilience--the ability to prevent, withstand, and recover from digital disruptions--has long been a strategic priority for enterprises. With the rise of agentic AI, the urgency for robust resilience is greater than ever. Agentic AI represents a new generation of autonomous systems capable of proactive planning, reasoning, and executing tasks with minimal human intervention. As these systems shift from experimental pilots to core elements of business operations, they offer new opportunities but also introduce new challenges when it comes to ensuring digital resilience. That's because the autonomy, speed, and scale at which agentic AI operates can amplify the impact of even minor data inconsistencies, fragmentation, or security gaps.

Machine anomalous sound detection (ASD) has emerged as one of the most promising applications in the Industrial Internet of Things (IIoT) due to its unprecedented efficacy in mitigating risks of malfunctions and promoting production efficiency. Previous works mainly investigated the machine ASD task under centralized settings. However, developing the ASD system under decentralized settings is crucial in practice, since the machine data are dispersed in various factories and the data should not be explicitly shared due to privacy concerns. To enable these factories to cooperatively develop a scalable ASD model while preserving their privacy, we propose a novel framework named CoopASD, where each factory trains an ASD model on its local dataset, and a central server aggregates these local models periodically. We employ a pre-trained model as the backbone of the ASD model to improve its robustness and develop specialized techniques to stabilize the model under a completely non-iid and domain shift setting. Compared with previous state-of-the-art (SOTA) models trained in centralized settings, CoopASD showcases competitive results with negligible degradation of 0.08%. We also conduct extensive ablation studies to demonstrate the effectiveness of CoopASD.

In this work, we share our experience on tele-knowledge pre-training for fault analysis, a crucial task in telecommunication applications that requires a wide range of knowledge normally found in both machine log data and product documents. To organize this knowledge from experts uniformly, we propose to create a Tele-KG (tele-knowledge graph). Using this valuable data, we further propose a tele-domain language pre-training model TeleBERT and its knowledge-enhanced version, a tele-knowledge re-training model KTeleBERT. which includes effective prompt hints, adaptive numerical data encoding, and two knowledge injection paradigms. Concretely, our proposal includes two stages: first, pre-training TeleBERT on 20 million tele-related corpora, and then re-training it on 1 million causal and machine-related corpora to obtain KTeleBERT. Our evaluation on multiple tasks related to fault analysis in tele-applications, including root-cause analysis, event association prediction, and fault chain tracing, shows that pre-training a language model with tele-domain data is beneficial for downstream tasks. Moreover, the KTeleBERT re-training further improves the performance of task models, highlighting the effectiveness of incorporating diverse tele-knowledge into the model.

Changing climatic conditions, the shortage of skilled workers, the use of pesticides – a wide range of factors have an impact on the quality and flow of agricultural processes. Researchers at the Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute, HHI are aiming to make this more efficient and sustainable by means of cloud and AI technologies. As part of the "NaLamKI" project, they are working with partners to establish a software-as-a-service platform that collects device and machine data to form a data basis for forecasts and decision-making aids. The agricultural sector is facing major challenges: German farmers are already feeling the far-reaching effects of climate change and will have to adapt to this to a greater extent in the future. Rising temperatures and changes in precipitation affect all agricultural variables, ranging from crop growth to crop rotations right through to tillage.

Changing climatic conditions, the shortage of skilled workers, the use of pesticides--a wide range of factors have an impact on the quality and flow of agricultural processes. Researchers at the Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute, HHI are aiming to make this more efficient and sustainable by means of cloud and AI technologies. As part of the "NaLamKI" project, they are working with partners to establish a software-as-a-service platform that collects device and machine data to form a data basis for forecasts and decision-making aids. The agricultural sector is facing major challenges: German farmers are already feeling the far-reaching effects of climate change and will have to adapt to this to a greater extent in the future. Rising temperatures and changes in precipitation affect all agricultural variables, ranging from crop growth to crop rotations right through to tillage.

To take one example from the above, there is enormous potential in using industrial knowledge graphs to enhance AI models by combining different datasets. "Knowledge graphs add context to the data you're analyzing," explains Norbert Gaus, Head of R&D in Digitalization and Automation at Siemens. "For example, machine data can be analyzed in the context of design data, including the tasks the machine is made for, the temperatures it should operate at, the key thresholds built into the parts, and so forth. To this we could add the service history of similar machines, including faults, recalls and expected inspection outcomes throughout the machine's operational life. Knowledge graphs make it much easier to augment the machine data we use to train AI models, adding valuable contextual information."

Predictive maintenance is one of the most exciting applications of digital technology in manufacturing. Simply put, predictive maintenance is the use of new and historical machine data to understand and, ideally, anticipate performance problems before they happen. Using sophisticated machine learning and AI techniques to analyze the data generated in the modern factory, predictive analytics can decrease downtime, optimize asset performance, and increase the lifespan of machines. The promises made on behalf of predictive maintenance (PdM) are big. Smart machines that flag performance issues before they happen.

SAP has caught the AI frenzy. Artificial intelligence is supposed to make everything better. There's nothing definitive yet, but machine and deep learning are supposed to optimize many SAP processes and make SAP's vision of the intelligent enterprise become reality. Our company is interested in AI as well, which is why we have already developed some prototypes together with Siemens. And they definitely have potential!

Modern software systems emit a tremendous amount of "machine data" (logs, metrics, etc.) that can be crucial to identifying and understanding misbehavior, but the quantity and complexity of this data is outpacing the human ability to do the required analysis and take timely action. For this reason, I think we will see a lot of opportunities to build automated systems that analyze (and even act) on this machine data in order to improve the security, performance, and reliability of economically critical software services. That said, there's also a lot of exciting research around "ML on code": automatically identifying risky pull requests, automated bug localization, intelligent IDE assistance, and so on. Given the well-known challenges of building and operating software systems, there is likely to be plenty of room for improvement across the entire lifecycle. Overall, I think we're heading into a really interesting time for the application of ML techniques to software development, security, and operations.

In a recent manufacturing industry insights survey on artificial intelligence (AI), 44 percent of respondents from the automotive and manufacturing sectors classified AI as "highly important" to the manufacturing function in the next five years, while almost half--49 percent--said it was "absolutely critical to success." Yet, in many cases, AI is hard to comprehend for manufacturers, as the technology industry has painted it with such a wide brush that few actually understand how it becomes instantiated--beyond some omnipotent source delivering better business results. Manufacturers may actually view AI as highly complex and expensive, requiring end-to-end systems throughout their whole company to work properly, and this translates to a costly overhaul of their entire IT/OT operation. The reality is, AI is much more focused and achievable. AI can work on factory floors with minimal construction and get connected to machines via the Industrial Internet of Things (IIoT).