As AI becomes a more integral part of our lives, so it raises many ethical questions around its application and uses.
In a paper by Napier’s Chief Data Scientist, Dr. Janet Bastiman, ‘Building trust and confidence in AI’ Dr. Janet discusses the issues surrounding traditional, quantitative approaches that have been used to measure AI and the question of consumer trust of AI systems. Janet argues that consumer trust - a qualitative measure affected by experience and context impacts traditional approaches to AI.
The paper also evaluates the factors that have contributed to a general lack of trust in AI and reticence to adopt AI systems, such as human nature and media hype around AI system failures. It also goes on to debunk myths surrounding explainability in AI, and best approaches to add explainability in AI models.
The full paper can be read at the link below, or through subscribing to the Journal of AI, Robotics & Workplace Automation.
Building trust and confidence in AI [FULL PAPER]
Abstract:
While some industries are rushing to adopt artificial intelligence (AI) technologies, there are those who are lagging behind, due either to their own lack of confidence or to perceived conflicts with regulation or their customer needs. This paper covers some of the myths perpetuated within the AI community regarding trust and confidence and how you can begin to build AI solutions with end user trust as a priority considering the latest regulatory proposals.
Contents:
DEFINING TRUST: WHAT DOES THE PUBLIC WANT?
EXPLAINABILITY: HISTORY AND MYTHS
ADDING EXPLAINABILITY TO MODELS
FEATURE ATTRIBUTIONS
- INSTANCE-BASED
- DIRECTLY INTERPRETABLE
SUPPORTING ACTIVITIES
CONCLUSION
Janet Bastiman is a committee member for the Royal Statistical Society Data Science Section and the IEEE UK STEM initiative. She regularly contributes evidence to the UK Parliament Science and Technology Committee. In addition to leading the artificial intelligence (AI) strategy at Napier, Janet is also an AI specialist for London-based venture capital company MMC Ventures.
Janet holds two undergraduate degrees and a Master’s in addition to her doctorate in computational neuroscience and has worked in industry for over 20 years. She regularly speaks at conferences including Minds Mastering Machines, InfoQ, London AI Summit, as well as giving her time to grassroots meet-ups.
Janet is passionate about ensuring that this pervasive technology is developed with best practices and that it is well understood by more than just its implementers.
Discover next-generation financial crime compliance technology
Book a demo of our AI-powered solutions or get in touch to find out how Napier can rapidly strengthen your AML defences and compliance capabilities.
