Search

A WRF‐based high‐resolution reanalysis of the Irish climate (1981–2010) is used to create proxy daily precipitation observations at the locations of climatological sites used for precipitation monitoring; the data are statistically representative of the real precipitation climate both for mean (over monthly, seasonal and annual periods) and extreme values. The proxy observations are spatially interpolated to the original WRF grid using a typical gridding package and compared against the original data to assess gridding errors. The errors are more complex than the estimates provided by the gridding software; systematic biases are evident which by the inclusion of strategically placed additional observing sites are shown to be greatly reduced. There is also evidence of systematic differences in trend analyses of extreme precipitation over the period. The method provides independent estimates of the errors that arise from actual gridding applications. It also facilitates the testing of the optimality of a network by highlighting possible inadequacies in an existing station layout and suggesting new observing site locations to fill gaps. Uncertainties regarding the errors in real precipitation observations, and possible spurious impacts linked to temporal changes in the real observing network, are avoided by this method. [ABSTRACT FROM AUTHOR]

Some encouraging uses for AI in medicine will lead to potentially novel legal liability issues. Complex algorithms involve an opacity that creates problems for the medical and legal professions alike. As iatrogenic injury is common in medical malpractice, the medical profession is understandably concerned when AI is introduced in diagnostic and therapeutic devices and events and outcome cannot be fully explained due to the "black box" effect. A concern about machine learning algorithms is the black box issue and understanding how conclusions or outcomes are reached. The deployment of AI devices in healthcare will require an increase in a clinician's understanding of AI to increase the transparency of their use. An important aspect of medical treatment is the notion of "therapeutic privilege". This will only arise in limited circumstances and requires the clinician to make a judgment, based on reasonable grounds, that the patient's physical or mental health might be seriously harmed by providing the information. Given the complexity of AI and the black box effect, could too much AI transparency possibly overwhelm a patient, such that it may dissuade them from giving consent in circumstances where treatment is necessary and essential? In other words, too much AI transparency and information may inadvertently hinder treatment and progress. [ABSTRACT FROM AUTHOR]

The introduction of novel medical technology, such as artificial intelligence (AI), into traditional clinical practice presents legal liability challenges that need to be squarely addressed by litigants and courts when something goes wrong. Some of the most promising applications for the use of AI in medicine will lead to vexed liability questions. As AI in health care is in its relative infancy, there is a paucity of case law globally upon which to draw. This article analyses medical malpractice where AI is involved, what problems arise when applying the tort of negligence -- such as establishing the essential elements of breach of duty of care and causation -- and how can these can be addressed. Product liability under Australian Consumer Law is beyond the scope of this article. In order to address this question, the article: (1) identifies the general problems that black box AI causes in the health care sector; (2) identifies the problems that will arise in establishing breach and causation due to the "black box" nature of AI, with reference to the Civil Liability Act 2002 (NSW) and common law through two hypothetical examples; and (3) considers selected legal solutions to the problems caused by "black box" AI. [ABSTRACT FROM AUTHOR]