Prof. Por Lip Yee, Professor at CSNET, Universiti Malaya.
Prof. Por Lip Yee was invited as a Keynote Speaker at the International Conference on Big Data, Communication Technology and Computer Applications (BDCTA 2025), where he delivered research on artificial intelligence in journalism titled "The Impact of Artificial Intelligence on Contemporary Journalism". The keynote draws on the rapid integration of AI-driven technologies in newsrooms to examine both the structural changes they enable and the social risks they introduce.
Reshaping News Production
Across major international newsrooms, artificial intelligence is actively changing how journalism is practised. Natural language processing enables automated reporting, allowing data-rich events to be converted into readable articles within seconds of the underlying data becoming available. Machine learning powers content recommendation, deciding which stories reach which readers and through which channels. Big data analytics guides editorial strategy, surfacing trending topics and helping teams allocate coverage resources more effectively. These shifts represent a structural change in how journalism operates, not a set of incremental tool improvements that leave existing workflows intact.
Personalisation and Its Limits
Among the most consequential AI applications in contemporary journalism is personalised news delivery. Recommendation systems analyse user behaviour to surface content most likely to be read, shared, or engaged with, which increases individual relevance and platform engagement metrics. However, this approach also creates systemic risks. Communities and viewpoints outside the mainstream may receive lower algorithmic amplification, effectively narrowing the range of perspectives that any given user encounters. This filter bubble effect poses a significant concern for public discourse and democratic information environments, where broad exposure to diverse viewpoints is a foundational requirement.
Digital Divide and Marginalisation
AI-powered journalism also risks exacerbating the digital divide. Sophisticated content generation and curation tools are predominantly developed by and for well-resourced organisations operating in dominant languages and markets. Smaller outlets, community-based journalism, and publications serving minority language communities may find themselves further disadvantaged as AI raises the technological baseline of the industry. This asymmetry can compound existing marginalisation of communities that already have limited media representation, leaving them with fewer resources to tell their own stories effectively.
A Path Forward
The paper concludes with a policy-oriented argument: that AI-powered journalism can serve all segments of society, but only through intentional design choices. Inclusive AI algorithms must be evaluated not only for accuracy but for distributional fairness across demographics and communities. Investments in digital literacy are needed so that citizens can critically engage with algorithmically mediated news and understand how curation decisions are shaping what they see. Equitable access policies should ensure that the analytical and production capabilities enabled by AI are available to journalism organisations beyond the major platforms and well-funded newsrooms.
The paper was presented at BDCTA 2025, the International Conference on Big Data, Communication Technology and Computer Applications.