Human-Centered Artificial Intelligence: Reliable, Safe & Trustworthy

By Ben Shneiderman. International journal of human computer interactions, 2020

1,318 citations recorded by the source corpus.

ABSTRACT Well-designed technologies that offer high levels of human control and high levels of computer automation can increase human performance, leading to wider adoption. The Human-Centered Artificial Intelligence (HCAI) framework clarifies how to (1) design for high levels of human control and high levels of computer automation so as to increase human performance, (2) understand the situations in which full human control or full computer control are necessary, and (3) avoid the dangers of excessive human control or excessive computer control. The methods of HCAI are more likely to produce designs that are Reliable, Safe & Trustworthy (RST). Achieving these goals will dramatically increase human performance, while supporting human self-efficacy, mastery, creativity, and responsibility.

Subfields: Radiological and Ultrasound Technology, Social Psychology, Surgery

View Paper: https://www.semanticscholar.org/paper/e49f67fa5c946ad24afcf59699a9cacf1ca53924

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Radiological and Ultrasound Technology

In Social Psychology

In Surgery

Citations in the corpus

Human-in-the-loop machine learning: a state of the art (2022, background): “Useful AI is part of a broader movement known as Human-centered AI (HAI) which refers to approaching AI from a human perspective by considering human conditions and contexts (Yang et al. 2021; Shneiderman 2020).”
Prompt Engineering for Text-Based Generative Art (2022, background): “The latter can be viewed from many different perspectives, such as human-centered AI [32], human-AI partnerships [29], and human-AI cooperation [5].”
The General Attitudes towards Artificial Intelligence Scale (GAAIS): Confirmatory Validation and Associations with Personality, Corporate Distrust, and General Trust (2022, background): “To combat this, balancing computer automation against human control may optimize perceived AI trustworthiness (Shneiderman, 2020).”
Human- versus Artificial Intelligence (2021, background): “In general, digital reasoning-and problem-solving agents only compare very super ﬁ cially to their biological counterparts, (e.g. Boden, 2017; Shneiderman, 2020b).”
A Systematic Literature Review of User Trust in AI-Enabled Systems: An HCI Perspective (2022, methodology): “In this case, a collective, multidisciplinary effort to close this gap is urgently needed (Hagendorff, 2020; Shneiderman, 2020c).”
Human-centered artificial intelligence in education: Seeing the invisible through the visible (2021, background): “HAI can be interpreted from two perspectives; one is AI under human control (Shneiderman, 2020), and the other is AI on the human condition (Stanford HAI, 2020).”
Position: Levels of AGI for Operationalizing Progress on the Path to AGI (2023, neutral): “Whether an AGI benchmark should include tests for potentially dangerous capabilities (e.g., the ability to deceive, to persuade (Veerabadran et al., 2023), or to perform advanced biochemistry (Morris, 2023)) is controversial.”
Expanding Explainability: Towards Social Transparency in AI systems (2021, background): “Since an explanation is only explanatory if it can be consumed by the recipient, many recognize the importance of taking user-centered approaches to XAI [70, 89, 100], and the indispensable role that the HCI community should play in advancing the field.”
Artificial intelligence in intelligent tutoring systems toward sustainable education: a systematic review (2023, neutral): “Shneiderman (2020) defined a two-dimensional approach for categorizing and providing guidance on how to produce system designs that are Reliable, Safe, and Trustworthy (RST).”
Bridging the Gap Between Ethics and Practice (2020, background): “In contrast, HCAI focuses on amplifying, augmenting, and enhancing human performance [103] in ways that make systems reliable, safe, and trustworthy [102].”
Design Principles for Generative AI Applications (2024, background): “Some of these guidelines include claims of universal applicability 8 or being of a general nature to AI-infused systems (e.g., [7, 153]).”