Experience and big amount of data are generated and used in risk and crisis management. Structuring the volume of data and learning from them are still big challenges to be faced to help actors either in decision-making or in operations. Data collection, for instance, is an important aspect, and sometimes, there can be overemphasis on using raw social media data for crisis informatics without adopting appropriate methodologies for cleaning the data and ensuring it is applicable to the situation at hand (i.e. assessing topical relevance). In recent years, this has become even more important with growing recognition that bots can often wield undue influence in social media, especially Twitter. Several techniques have been developed in the last years in Artificial Intelligence (AI) study and Computer Supported Cooperative Work (CSCW) that can be applied to face these challenges. The combination of these tools and methods continue to show promising results in improving sharing of information in crisis and emergency contexts.There are many approaches of AI, such as neural networks and ontologies that can be used to support risk and crisis management.Machine learning, in particular, is an approach that gives "computers the ability to learn without being explicitly programmed" by learning from and making predictions from data. Also, the use of symbolic AI approaches, like ontologies as a knowledge representation mechanism, offers many advantages in information retrieval and analysis.In addition, semantic models of knowledge allow users as well as systems to clearly understand what is happening in a crisis situation and can provide support to decision makers. This special issue mainly addresses the application of semantic models and AI methods and tools trying to answer to users' needs in the scope of risk management, crisis response, prediction, modelling and mitigation. According to a policy forum article in Science in 2016, (Palen & Anderson, 2016) describe crisis informatics as a "multidisciplinary field combining computing and social science knowledge of disasters." The special issue covers a broad range of topics that fall within, and even expand the scope of, crisis informatics. The articles are particularly timely today, as the world grapples with the COVID-19 pandemic that has resulted in hundreds of thousands of deaths, and millions of infections. Although this issue was prepared before the COVID-19 crisis struck, many of the individual topics covered by an international set of authors are highly relevant to the situation unfolding before our eyes.