New research argues cultural complexity must be built into AI design

In an era where Artificial Intelligence (AI) rapidly pervades various sectors, researchers from The Alan Turing Institute, the University of Edinburgh, and the UK's Arts & Humanities Research Council (AHRC-UKRI) have unveiled a new vision for AI design. This vision uniquely integrates the humanities, recognising the profound cultural implications and opportunities these fields offer for enhancing AI technologies.

The white paper titled "Doing AI Differently," released in collaboration with Lloyd’s Register Foundation, advocates for a shift in AI design approach. It suggests that the text and images produced by AI systems resemble cultural artefacts more than data sets, necessitating an understanding rooted in humanities for interpreting these systems effectively.

Highlighting parallels to the rapid rise of social media platforms, which were launched without adequate contextual safeguards, the paper warns against repeating similar missteps with AI's current fast-paced deployment.

Central to this new AI design paradigm is the need for systems capable of understanding cultural nuance, context, and multiple perspectives. This is crucial as AI's remit extends to complex modern-day challenges like climate change and pandemics, areas where interpretive depth is indispensable.

Professor Drew Hemment, Theme Lead for Interpretive Technologies for Sustainability at The Alan Turing Institute, emphasises the urgency: "AI systems are increasingly operating in sensitive domains - such as healthcare, climate action and democratic discourse - where cultural context and interpretive judgement are essential." Signifying a fundamental gap in current systems, he calls for AI that can engage deeply with human meanings. He says "The last few years have made it clear that AI can be a formidable tool to solve modern day challenges. But these systems often fail when nuance and context matter most."

The white paper proposes three core innovations to bridge this gap:

1. Interpretive Technologies: Systems designed to accommodate ambiguity, context, and plurality from the outset.
2. New AI Architectures: These expand beyond the narrow design parameters of existing systems, envisioning a broader and more inclusive architecture.
3. Human-AI Ensembles: Rather than replacing human capabilities, these systems aim to enhance them, fostering collaboration and synergy between human insight and AI prowess.

Professor Hemment articulates the pivotal nature of this moment in AI history: "Decisions... about AI architecture today will shape the systems of tomorrow... We have a narrowing window to build in interpretive capabilities from the ground up. This is our opportunity to shape a new generation of AI - one that amplifies rather than erods human potential."

An international funding initiative accompanies the white paper, spearheaded by AHRC and the Social Sciences and Humanities Research Council of Canada (SSHRC), initiating collaborative projects across continents. This aims to build a cohesive research community dedicated to responsible AI deployment.

Jan Przydatek, Director of Technologies at Lloyd’s Register Foundation, underscores the safety imperative: "As a global safety charity, our priority is to ensure future AI systems, whatever shape they take, are deployed in a safe and reliable manner. This research proposes a fundamental shift: bringing humanities expertise into the core of AI design, to unlock AI's potential to solve humanity's most complex challenges while ensuring these technologies amplify rather than diminish human agency and cultural diversity. We hope the paper will inform industry leaders and policymakers around the world as we take a crucial step towards the safe, scaled up deployment of AI in safety-critical areas.”

In conclusion, this research champions a transformative approach, integrating humanities expertise at the heart of AI development. By fostering AI systems responsive to cultural complexity and human diversity, it charts a path towards responsible AI that aligns with humanity's values and needs.