# The five Is: Key principles for interpretable and safe conversational AI

arXiv: https://arxiv.org/abs/2108.13766 · HTML: https://mlsanity.com/paper/2108.13766/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 33 references that are in the map.
- First posted on arXiv 31 Aug 2021.

## Builds on (most cited first, 8 of 8)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Extracting Training Data from Large Language Models](https://mlsanity.com/paper/2012.07805/) (2020)
- [Large image datasets: A pyrrhic win for computer vision?](https://mlsanity.com/paper/2006.16923/) (2020)
- [The Hardware Lottery](https://mlsanity.com/paper/2009.06489/) (2020)
- [Artificial Intelligence: the global landscape of ethics guidelines](https://mlsanity.com/paper/1906.11668/) (2019)
- [Foundations for Near-Term Quantum Natural Language Processing](https://mlsanity.com/paper/2012.03755/) (2020)
- [Logical Semantics and Commonsense Knowledge: Where Did we Go Wrong, and How to Go Forward, Again](https://mlsanity.com/paper/1808.01741/) (2018)
- [Interactive Teaching for Conversational AI](https://mlsanity.com/paper/2012.00958/) (2020)

## Built on by (most cited first, 2 of 2)

- [Foundation Metrics for Evaluating Effectiveness of Healthcare Conversations Powered by Generative AI](https://mlsanity.com/paper/2309.12444/) (2023)
- [A Conceptual Framework for AI-based Decision Systems in Critical Infrastructures](https://mlsanity.com/paper/2504.16133/) (2025)

## Abstract (arXiv)

In this position paper, we present five key principles, namely interpretability, inherent capability to explain, independent data, interactive learning, and inquisitiveness, for the development of conversational AI that, unlike the currently popular black box approaches, is transparent and accountable. At present, there is a growing concern with the use of black box statistical language models: While displaying impressive average performance, such systems are also prone to occasional spectacular failures, for which there is no clear remedy. In an effort to initiate a discussion on possible alternatives, we outline and exemplify how our five principles enable the development of conversational AI systems that are transparent and thus safer for use. We also present some of the challenges inherent in the implementation of those principles.