15 challenges for AI: or what AI (currently) can’t do

Isaksson

Journal of Mechanical Design

et al. 2020

In the era of digitalization, manufacturing companies expect their growing access to data to lead to improvements and innovations. Manufacturing engineers will have to collaborate with data scientists to analyze the ever-increasing volume of data. This process of adopting data science techniques into an engineering organization is a sociotechnical process fraught with challenges. This article uses a participant observation case study to investigate and discuss the sociotechnical nature of the adoption data science technology into an engineering organization. In the case study, a young data scientist/statistician interacted with experienced production engineers in a global automotive organization to mutual satisfaction. However, the case study highlights the mis-aligned expectations between engineers and data scientists and knowledge in what is necessary to successfully benefit from manufacturing process data. The results reveal that the engineers had an initially romantic and idealistic view on how data scientists can bring value out of dispersed and complex information residing in the multisite manufacturing organization’s datasets in a “magic” way. Conversely, the data scientist had not enough engineering and contextual understanding to ask the right questions. The case reveals important shortcomings in the sociotechnical processes that undergo changes as digitalization is brought into mature engineering organizations and points to a lack of knowledge on multiple levels of the data analysis process and the ethical implications this could have.

Section: Discussion: Lack Of Understanding On Multiple Levelsmentioning

confidence: 99%

“…This also explains the current distrust for human experts who bring understanding and insights into the context to the analysis of problems. In a sense, the experts are "in the way"; they block the direct representation of the world that the data analysis seems to promise [72].…”

Section: Discussion: Lack Of Understanding On Multiple Levelsmentioning

confidence: 99%

Data Fairy in Engineering Land: The Magic of Data Analysis as a Sociotechnical Process in Engineering Companies

Isaksson

Journal of Mechanical Design

et al. 2020

Proceedings of the 2019 2nd Artificial Intelligence and Cloud Computing Conference

“…The results demonstrated that follow-up questions may not be desired when asking for general information (e.g., weather information), but are preferred in decision-making processes (e.g., purchase choices). This aspect is essential, and must be carefully considered in the design process of a chatbot as it may have serious practical implications, for instance for users who may ask medical questions and receive recommendations based on their input [4,18].…”

Section: Discussionmentioning

confidence: 99%

Artificial Intelligence in Conversational Agents

Svenningsson

Faraon

2019

Arti cial intelligence (AI) is gaining traction in service-oriented businesses in the form of chatbots. A chatbot is a popular type of social AI that uses natural language processing to communicate with users. Past studies have shown discrepancies in terms of whether or not a chatbot should communicate and behave like a human. This article aims to explore these discrepancies in order to provide a theoretical contribution of a list of factors related to perceived humanness in chatbots and how these may consequently lead to a positive user experience. The results suggest that a chatbot should have the following characteristics: avoiding small talk and maintaining a formal tone; identifying itself as a bot and asking how it can help; providing speci c information and articulating itself with sophisticated choices of words and well-constructed sentences; asking follow-up questions during decision-making processes and; providing an apology when the context is not comprehensible, followed by a question or a statement to dynamically move a conversation forward. These results may have implications for designers working in the eld of AI as well as for the wider debates and the broader discourse around the adoption of AI in society. CCS CONCEPTS • Human-centered computing → Interaction design → Interaction design theory, concepts and paradigms • Computing methodologies → Artificial intelligence → Intelligent agents; Discourse, dialogue and pragmatics

“…A thought-provoking paper by Hagendorff and Wezel ( 2019 ) classifies what AI can and cannot do. Some of the authors’ concerns, such as measurement, completeness and quality of the data, and problems with transparency of algorithms, have already been discussed, so I will describe those that I feel are most relevant to mental health services.…”

Section: Problems and Limitations With Aimentioning

confidence: 99%

Improving Mental Health Services: A 50-Year Journey from Randomized Experiments to Artificial Intelligence and Precision Mental Health

Bickman

2020

Adm Policy Ment Health

101

This conceptual paper describes the current state of mental health services, identifies critical problems, and suggests how to solve them. I focus on the potential contributions of artificial intelligence and precision mental health to improving mental health services. Toward that end, I draw upon my own research, which has changed over the last half century, to highlight the need to transform the way we conduct mental health services research. I identify exemplars from the emerging literature on artificial intelligence and precision approaches to treatment in which there is an attempt to personalize or fit the treatment to the client in order to produce more effective interventions. Keywords Mental health services • Artificial intelligence • Machine learning • Precision mental health • Randomized clinical trials (RCTs) • Precision medicine "AI will bring many wonders. It may also destabilize everything from nuclear détente to human friendships. We need to think much harder about how to adapt... it is changing human knowledge, perception and reality-and, in so doing, changing the course of human history. We seek to understand it and its consequences and encourage others across the disciplines to do the same."-Kissinger et al. 2019, pp. 24-26 "AI is one of the most profound things we're working on as humanity. It's more profound than fire or electricity."-Alphabet CEO Sundar Pichai, in Thompson and Bodoni 2020. This article is part of a special issue, Festschrift for Leonard Bickman: The Future of Children's Mental Health Services.