![]() ![]() In the field of artificial intelligence, however, this is difficult due to the extensive and complex possible state space. For example, during verification and validation activities in the software life cycle, the influences of different input values in the system are investigated, but these can be relatively easily mapped by boundary value analyses. Įstablished measures of risk reduction in the development of software are limited in their ability to mitigate these risks, and existing safety standards are hardly applicable to AI systems as they do not take into account their technical peculiarities. Even today, we already face an increasing number of accidents in systems that utilise artificial intelligence, including various reports on fatal accidents due to AI-related failures in automated vehicles. This is because, despite the rapid, positive progression of this technology and the new prospects for occupational safety, the increasing application of this technology will also produce new risks. ![]() However, for the benefit of human safety and health, safe and trustworthy artificial intelligence is required. Furthermore, they enable accidents to be prevented by assistance systems capable of recognising hazardous situations. Protective devices and control systems based on artificial intelligence have already enabled fully automated vehicles and robots to be created. These capabilities enable entirely new smart systems and applications, which is why artificial intelligence is often seen as the key technology of the future. Thus, by using artificial intelligence, concepts such as learning, planning, perceiving, communicating and cooperating can be applied to technical systems. ĪI systems are engineered systems that build, maintain, and use a knowledge model to conduct a predefined set of tasks for which no algorithmic process is provided to the system. An example would be the use of convolutional neural networks in the field of image processing. These neural networks are optimised and trained for specific tasks, and they can differ fundamentally in terms of their architecture and mode of operation. A significant subcategory of machine learning is deep learning, which deals with the development and application of deep neural networks. Today, the term artificial intelligence is mainly used in the context of machine learning, such as decision trees or support vector machines, but also includes a variety of other applications, such as expert systems or knowledge graphs. Driven by success in the fields of image recognition, natural language processing and self-driving vehicles, in the coming years, the fast-growing market of artificial intelligence (AI) will play an increasingly significant role in occupational safety. The importance of artificial intelligence is constantly increasing due to ongoing research successes and the introduction of new applications based on this technology. AI methods do not only allow significantly higher levels of automation to be achieved, but they also open up completely new fields of application. These new sources of risk should be taken into account in the overall risk assessment of a system based on AI technologies, examined for their criticality and managed accordingly at an early stage to prevent a later system failure.Īrtificial intelligence (AI) methods are mainly used to solve highly complex tasks, such as processing natural language or classifying objects in images. On this basis, a taxonomy could be created that provides an overview of various AI-specific sources of risk. For this purpose, the differences between AI systems, especially those based on modern machine learning methods, and classical software were analysed, and the current research fields of trustworthy AI were evaluated. This work objects to contribute hereto by identifying relevant sources of risk for AI systems. Risk management for systems that for systems using AI must therefore be adapted to the new problems. However, established risk mitigation measures in software development are only partially suitable for applications in AI systems, which only create new sources of risk. Artificial intelligence can be used to realise new types of protective devices and assistance systems, so their importance for occupational safety and health is continuously increasing.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |