My research

AI Research

Evaluating the Social Impact of Generative AI Systems in Systems and Society

A guide and categorization of social impacts that are able to be evaluated across generative AI modalities. We describe social impact categories and how to conduct evaluations in the base technical system, then in people and society. Suggested methods for evaluation apply to all modalities and analyses of the limitations of existing evaluations serve as a starting point for necessary investment in future evaluations.

Solaiman and Talat et al, Evaluating the Social Impact of Generative AI Systems in Systems and Society, preprint, 2023.

The Gradient of Generative AI Release: Methods and Considerations

A framework to assess six levels of access to generative AI systems, analysis of past release methods and tensions, and enumerated safety controls and guardrails for generative systems and necessary investments to improve future releases. Accepted to FAccT 2023.

Solaiman, The Gradient of Generative AI Release: Methods and Considerations, FAccT, 2023.

Process for Adapting Language Models to Society (PALMS) with Values-Targeted Datasets

A proposed process for crafting a small, curated dataset on to adapt language model behavior to cultural context via fine-tuning. Accepted to NeurIPS 2021 as a spotlight presentation (less than 3% of papers).

Research: Solaiman and Dennison, Process for Adapting Language Models to Society (PALMS) with Values-Targeted Datasets, NeurIPS, 2021.

Release Strategies and the Social Impacts of Language Models

OpenAI's policy research paper on language model GPT-2; insights on misuse, detection, and biases; and its staged release and recommendations for responsible publicaiton.

Research: Solaiman et al, Release Strategies and the Social Impacts of Language Models, OpenAI, 2019.

Press Publications

Five Myths About How AI Will Affect 2024 Elections

Insights on generative AI and elections, debunking popular myths.

Solaiman,Five Myths About How AI Will Affect 2024 Elections, Tech Policy Press, 2024.

Generative AI Systems Aren't Just Open or Closed

Op-Ed on release methods of generative AI systems.

Solaiman, Generative AI Systems Aren't Just Open or Closed, Wired Ideas, 2023.

Press Highlights

AI Detection, EuroNews: Video on ChatGPT: Is it possible to detect AI-generated text? , January 2023.

AI Detection and Watermarks, MIT Tech Review: Melissa Heikkilä on A watermark for chatbots can expose text written by an AI, January 2023.

AI Generated Scientific Papers, Nature: Holly Else on Abstracts written by ChatGPT fool scientists, January 2023.

AI Detection Methods, Wall Street Journal: Ann-Marie Alcántara on Is It Human or AI? New Tools Help You Spot the Bots, January 2023.

AI Detection, Consent, and Norms, Libération: Florian Gouthière on ChatGPT : comment détecter qu’un texte a été écrit par l’intelligence artificielle ?, January 2023.

PALMS in NLP, New York Times: Steven Johnson on A.I. Is Mastering Language. Should We Trust What It Says?, April 2022.

AI Detection for Language Models, MIT Tech Review: Melissa Heikkilä on How to spot AI-generated text, December 2022.

EU AI Act and General Purpose Systems, Venture Beat: Taryn Plumb on AI Act: What does general purpose AI (GPAI) even mean?, September 2022.

Mozilla Panel, The EU’s struggle with general-purpose AI, December 2022.

Microsoft Data Science & Law Forum 4.0: Governing AI, Deploying Large Language Models Safely and Lawfully, May 2022.

All Tech is Human Panel, Algorithmic Transparency in Digital Spaces, May 2022.

Center for Data Innovation, Should the EU Regulate General-Purpose AI Systems?, September 2022.

100 Brilliant Women in AI Ethics™ – 2023

AI Developers to Watch 2023