Towards generalisable hate speech detection: a review on obstacles and solutions

W Yin, A Zubiaga - PeerJ Computer Science, 2021‏ - peerj.com
Hate speech is one type of harmful online content which directly attacks or promotes hate
towards a group or an individual member based on their actual or perceived aspects of …

Social data: Biases, methodological pitfalls, and ethical boundaries

A Olteanu, C Castillo, F Diaz, E Kıcıman - Frontiers in big data, 2019‏ - frontiersin.org
Social data in digital form—including user-generated content, expressed or implicit relations
between people, and behavioral traces—are at the core of popular applications and …

Sparks of artificial general intelligence: Early experiments with gpt-4

S Bubeck, V Chandrasekaran, R Eldan… - ar** and refining large language
models (LLMs) that exhibit remarkable capabilities across a variety of domains and tasks …

Should chatgpt be biased? challenges and risks of bias in large language models

E Ferrara - arxiv preprint arxiv:2304.03738, 2023‏ - arxiv.org
As the capabilities of generative language models continue to advance, the implications of
biases ingrained within these models have garnered increasing attention from researchers …

Taxonomy of risks posed by language models

L Weidinger, J Uesato, M Rauh, C Griffin… - Proceedings of the …, 2022‏ - dl.acm.org
Responsible innovation on large-scale Language Models (LMs) requires foresight into and
in-depth understanding of the risks these models may pose. This paper develops a …

Teaching English in the age of AI: Embracing ChatGPT to optimize EFL materials and assessment

O Koraishi - Language Education and Technology, 2023‏ - langedutech.com
The sudden spike of artificial intelligence (AI) technologies has had and continues to have a
transformative impact on various domains, including education. The advent of AI-powered …

Ethical and social risks of harm from language models

L Weidinger, J Mellor, M Rauh, C Griffin… - arxiv preprint arxiv …, 2021‏ - arxiv.org
This paper aims to help structure the risk landscape associated with large-scale Language
Models (LMs). In order to foster advances in responsible innovation, an in-depth …

The capacity for moral self-correction in large language models

D Ganguli, A Askell, N Schiefer, TI Liao… - arxiv preprint arxiv …, 2023‏ - arxiv.org
We test the hypothesis that language models trained with reinforcement learning from
human feedback (RLHF) have the capability to" morally self-correct"--to avoid producing …

The gradient of generative AI release: Methods and considerations

I Solaiman - Proceedings of the 2023 ACM conference on fairness …, 2023‏ - dl.acm.org
As increasingly powerful generative AI systems are developed, the release method greatly
varies. We propose a framework to assess six levels of access to generative AI systems: fully …

Evaluating the social impact of generative ai systems in systems and society

I Solaiman, Z Talat, W Agnew, L Ahmad… - arxiv preprint arxiv …, 2023‏ - arxiv.org
Generative AI systems across modalities, ranging from text (including code), image, audio,
and video, have broad social impacts, but there is no official standard for means of …