The accelerated development of Artificial Intelligence (AI) capabilities and systems is driving a paradigm shift in productivity, innovation and growth. Despite this generational opportunity, AI is fraught with significant challenges and risks. To address these challenges, responsible AI has emerged as a modus operandi that ensures protections while not stifling innovations. Responsible AI minimizes risks to people, society, and the environment. However, responsible AI principles and practice are impacted by ‘principle proliferation’ as they are diverse and distributed across the applications, stakeholders, risks, and downstream impact of AI systems. This article presents a systematic review of responsible AI principles and practice with the objectives of discovering the current state, the foundations and the need for responsible AI, followed by the principles of responsible AI, and translation of these principles into the responsible practice of AI. Starting with 22,711 relevant peer-reviewed articles from comprehensive bibliographic databases, the review filters through to 9700 at de-duplication, 5205 at abstract screening, 1230 at semantic screening and 553 at final full-text screening. The analysis of this final corpus is presented as six findings that contribute towards the increased understanding and informed implementation of responsible AI.

Generative Artificial Intelligence (AI) is heralding a new era in AI for performing a spectrum of complex tasks that are indistinguishable from humans. Alongside language and text, Generative AI models have been built for all other modalities of digital data, image, video, audio, and code. The full extent of Generative AI and its opportunities, challenges, contributions, and risks are still being explored by academic researchers, industry practitioners, and government policymakers. While this deep understanding of Generative AI continues to evolve, the lack of fluency, literacy, and effective interaction with Generative and conventional AI technologies are common challenges across all domains. Tertiary education institutions are uniquely positioned to address this void. In this article, we present the human-centred design of a universal AI literacy module, followed by its four primary constructs that provide core competence in AI to coursework and research students and academic and professional staff in a tertiary education setting. In comparison to related work in AI literacy, our design is inclusive due to the collaborative approach between multiple stakeholder groups and is comprehensive given the descriptive formulation of the primary constructs of this module with exemplars of how they activate core operational competence across the four groups.

Text classifiers are Artificial Intelligence (AI) models used to classify new documents or text vectors into predefined classes. They are typically built using supervised learning algorithms and labelled datasets. Text classifiers produce a predefined class as an output, which also makes them susceptible to adversarial attacks. Text classifiers with high accuracy that are trained using complex deep learning algorithms are equally susceptible to adversarial examples, due to subtle differences that are indiscernible to human experts. Recent work in this space is mostly focused on improving adversarial robustness and adversarial example detection, instead of detecting adversarial attacks. In this paper, we propose a novel approach, explainable AI with integrated gradients (IGs) for the detection of adversarial attacks on text classifiers. This approach uses IGs to unpack model behavior and identify terms that positively and negatively influence the target prediction. Instead of random substitution of words in the input, we select the top p% words with the greatest positive and negative influence as substitute candidates using attribution scores obtained from IGs to generate k samples of transformed inputs by replacing them with synonyms. This approach does not require changes to the model architecture or the training algorithm. The approach was empirically evaluated on three benchmark datasets, IMDB, SST-2, and AG News. Our approach outperforms baseline models on word substitution rate, detection accuracy, and F1 scores while maintaining equivalent detection performance against adversarial attacks.