Collaborative Discussion 2
Legal and Ethical Views on ANN Applications: Units 8 - 10
Overview
This collaborative discussion explored the benefits and risks of robo-writers across various business sectors. Participants in this discussion came from different professional backgrounds and, alongside discussion on general business tasks, there was focus on indiviual business areas. It was interesting to observe that, despite these varied business sectors, the potential benefits and risks, and the technical and ethical challenges consistently highlighted similar themes.
The potential for biased output and potential ethical issues were raised by several peers, and the effect on human creativity was a common theme. The use of robo-writers as an ‘assistant’ as opposed to a replacement for humans, was a common conclusion by authors.
On this page my summary post is shown, alongside relevent references.
Summary Post
This collaborative discussion covered the use of artificial intelligence (AI) writers for differing tasks, from routine administrative, through high-level business, to creative tasks. Common themes emerged from the discussions, the potential benefits from AI writers in terms of productivity, efficiency and accessibility were recognised across posts, as were potential risks, specifically in relation to introduction of bias, hallucinations, potential to produce harmful content, lack of transparency and reduction in diversity of outputs.
In my initial post I discussed the use of Retrieval-Augmented-Generation (RAG) which, by grounding responses to verified source documentation can help mitigate concerns linked to fabricated outputs. Richard added to this comment in the peer response by discussing the EU AI Act’s Article 50 transparency obligations, which become enforceable from August 2026 and are designed to increase transparency and increase accountability with respect to AI use (European Union, 2024).
Whilst mitigations such as RAG and Article 50 should improve transparency, accuracy and accountability, RAG is limited to niche tasks only, and the EU article whilst addressing some concerns, has a greater focus on disclosure and transparency as opposed to quality control.
A study on trust, attitudes and use of AI, reported that 66% of workers relied on AI output without evaluating accuracy (KPMG and The University of Melbourne, 2025). This together with the reported drive of many businesses to increase employees use of AI (Fay, 2026), is a worrying statistic. This links to an additional common theme within the discussion, that AI writers should be considered as assistants complementary to human knowledge and that appropriate human oversight remains critical.
References
European Union (2024) Regulation (EU) 2024/1689 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act), Article 50. Available at: https://artificialintelligenceact.eu/article/50/ (Accessed: 2 July 2026).
Fay, J (2026). How ‘confused’ AI rollout hurts firms and baffles staff. Available at: https://www.bbc.com/news/articles/c74d1ydv01eo (Accessed: 2 July 2026).
KPMG & The University of Melbourne (2025) Trust, attitudes and use of artificial intelligence: A global study 2025. Available at: https://assets.kpmg.com/content/dam/kpmgsites/xx/pdf/2025/05/trust-attitudes-and-use-of-ai-global-report.pdf (Accessed: 2 July 2026).