Record STMT-W88032AAdvisory

The Challenges and Opportunities of Human-Centered AI for Trustworthy Robots and Autonomous Systems

White paper Institutional policyGlobalin force

This research article systematically explores the key facets of human-centered artificial intelligence for trustworthy robots and autonomous systems (RAS). It identifies five key properties of trustworthy RAS—safety, security, system health, human-machine interaction, and legal/ethical compliance—and examines the challenges and AI opportunities in implementing them. Finally, it proposes a new acceptance model of RAS to ensure both worthiness and trustiness for human adoption and highlights the goal of augmenting human capabilities.

Summary and content fields are drawn from the document text and are pending editorial review.

Who issued it

De Montfort UniversityAcademic

Lead issuer

University of ManchesterAcademic · GB

Co-issuer

Loughborough UniversityAcademic · GB

Co-issuer

Nottingham Trent UniversityAcademic · GB

Co-issuer

Ulster UniversityAcademic

Co-issuer

University of StrathclydeAcademic

Co-issuer

Legal forceAdvisory

Under whose authorityInstitutional policy

Judged fromfull text

Who it governs

Researchers and academicsGeneral publicAI developers and providersAI deployers and users

Researchers, engineers, and developers seeking a comprehensive academic framework on human-centered AI and trustworthiness for autonomous systems.

Sectors

Research and science · Robotics and Autonomous SystemsCross-sector

What the document contains

What's covered

The document addresses the technological and ethical foundations required to build trustworthy robots and autonomous systems. It reviews the challenges and AI-driven solutions across safety, security, system health, human-machine interaction, and ethical considerations.

What's missing

The text lacks binding regulatory enforcement mechanisms, specific legislative text, or actionable compliance checklists for commercial developers.

Key takeaways

Trustworthy Robots and Autonomous Systems (RAS) require five core properties: safety, security, system health, effective human-machine interaction, and compliance with law and ethics.
Cyber threats to autonomous systems extend from IT infrastructure into physical actuators via connected technologies like the Internet of Robotic Things.
Internal system health and fault tolerance are critical for autonomous systems to prevent catastrophic failures without human intervention.
A new acceptance model of RAS combines functional worthiness with trustiness properties to promote human adoption.
Human-centered AI should focus on promoting human-level intelligence, augmenting human capabilities, and ensuring beneficial impacts on humanity.

Principles it states

Trust is built on predictability and understanding.Design for Error principles should assume faults will occur and provide appropriate scope for human interventions.Robots are manufactured artefacts and should not be designed in a deceptive way to exploit vulnerable users.RAS must obey Asimov's three laws of robotics as an essential condition for safety and ethical alignment.

Outline

· Abstract
· I. Introduction
· II. Key Factors That Affect TRAS
· III. The Challenges in Implementing TRAS
· IV. Opportunities of AI Technologies
· V. Human-Centered AI for TRAS
· VI. Conclusions

Issued2 December 2021

Length14,447 words · 15 pages

FormatNot recorded

Scopeglobal

How it scores

Pending review

Each score is backed by a quote verified against the document text. Editorial review has not yet confirmed them.

Fairness

2/5

Non-discrimination and equitable treatment

Technology is trusted if it benefits humans, is demonstrably fair, safe and reliable, is well regulated

Section V.D

screening score · 2026-10-23

Human Dignity

2/5

Respect for the inherent worth of persons

The moral standing and dignity of a human cannot be conveyed on any machine.

Section II

screening score · 2026-10-23

Human Oversight

4/5

Humans remain in control of consequential decisions

While designing RAS, we should have a solution that allows humans to interrupt robots’ work at any execution points and restrict the situation to one that would not harm humans.

Section V.D

screening score · 2026-10-23

Transparency

2/5

Openness about how AI systems work and are used

Logicality, dependability, correctness and transparency of decision-making, performance and overall quality are essential requirements of autonomous systems upon which trust is built.

Section II

screening score · 2026-10-23

Enforcement

0/5

Consequences and compliance mechanisms with real force

screening score · 2026-10-23

Topics

Physical AI Governancecontext

autonomous_systems trustworthy_ai cyber_security human_robot_interaction fault_diagnosis robotics_ethics