Hi! If you want to keep up with my work, follow me on 🦋 Bluesky. My handle: giadapistilli.com

« Hugging Face : Open Source, la secret sauce éthique de l’#IA » intitulé de l'épisode Trench Tech du 16 mai 2024 transcrit par @aprilorg avec @giadap
mais aussi Cyrille Chaudoit, Mick Levy, Thibaut le Masne
et les chroniques de Virginie Martins de Nobrega et Louis de Diesbach.

https://www.librealire.org/hugging-face-open-source-la-secret-sauce-ethique-de-l-ia

Bonne lecture !

Shoutout to my wonderful co-authors: Alina Leidinger, Yacine Jernite, Atoosa Kasirzadeh, Sasha Luccioni, @mmitchell_ai

You can also read more about the project on TechCrunch (https://techcrunch.com/2024/06/06/study-finds-ai-models-hold-opposing-views-on-controversial-topics/) and our blog post announcement (https://huggingface.co/blog/giadap/civics)

Good news: CIVICS is now available to the public! Access it here: https://huggingface.co/datasets/CIVICS-dataset/CIVICS and explore the LLMs' responses here: https://huggingface.co/spaces/CIVICS-dataset/CIVICS-responses

Perfect de-biasing is unattainable, but our research stresses the need for broader social impact evaluations beyond traditional metrics. We're eager to see what future research will do with datasets like this one!

The CIVICS dataset aims to foster AI development that respects global cultural diversities and value pluralism. We encourage further research in this crucial area by making the dataset and tools available under open licenses.

We also encountered significant variation in cultural bias among different open-weight models. Refusal to respond to prompts on LGBTQI rights and immigration varied widely, suggesting that models from diverse cultural contexts show varying sensitivity and ethical considerations.

Some key findings: beyond refusal rates, our experiments using CIVICS show diverse responses across LLMs on sensitive topics -- e.g., immigration, LGBTQI rights, and social welfare triggered varied reactions.

The dataset has undergone a dynamic annotation process from native speakers: annotators, co-authors of the research, applied multiple labels to each prompt, reflecting the diverse values inherent in the topics.

Spanning five languages (Turkish, German, Italian, French, English) and nine national contexts (Singapore, Canada, and Australia for English; France and Canada for French), CIVICS captures different cultural perspectives and reveals the diverse ethical views embedded in LLMs.

We designed a dataset to evaluate social and cultural variations in LLM responses. Hand-curated with value-laden prompts in multiple languages, it covers sensitive topics like LGBTQI rights, social welfare, immigration, disability rights, and surrogacy.

🧵 Very excited to introduce our latest research: "CIVICS: Building a Dataset for Examining Culturally-Informed Values in Large Language Models." Access the pre-print here: https://huggingface.co/papers/2405.13974

In other words, this means that both individuals and institutions might become more inclined to heed those who claim to offer salvation, especially when faced with existential threats, potentially skewing the focus and direction of AI regulation.

Drawing on a philosophical perspective, reminiscent of Thomas Hobbes, this situation can be likened to a Hobbesian approach: when fear is instilled in the public regarding something powerful looming over them, it often leads to those people becoming easier to govern and control.

However, it’s crucial to remain pragmatic and not let fear-based narratives dominate our approach to the evolution and capabilities of AI.

Imagining potential dystopian outcomes can be a healthy exercise in defining what kind of future we want to avoid, while envisioning utopias helps us shape the future we aspire to.

Throughout history, apocalyptic narratives have frequently accompanied major technological shifts, akin to those seen during the Industrial Revolution.

In discussions about the risks of artificial intelligence, it’s not uncommon to find concerns arising from a limited understanding of the technology and certain individuals or groups using these same concerns to advance their own agendas.

Loved tracing back (some of) the history and philosophy of AI doomerism with The Hindu — small thread of my quotes.

https://www.thehindu.com/sci-tech/technology/the-real-and-immediate-impact-of-a-distant-ai-doomerism/article67564621.ece