Multi-Agent Risks from Advanced AI

no code implementations • 19 Feb 2025 • Lewis Hammond, Alan Chan, Jesse Clifton, Jason Hoelscher-Obermaier, Akbir Khan, Euan McLean, Chandler Smith, Wolfram Barfuss, Jakob Foerster, Tomáš Gavenčiak, The Anh Han, Edward Hughes, Vojtěch Kovařík, Jan Kulveit, Joel Z. Leibo, Caspar Oesterheld, Christian Schroeder de Witt, Nisarg Shah, Michael Wellman, Paolo Bova, Theodor Cimpeanu, Carson Ezell, Quentin Feuillade-Montixi, Matija Franklin, Esben Kran, Igor Krawczuk, Max Lamparth, Niklas Lauffer, Alexander Meinke, Sumeet Motwani, Anka Reuel, Vincent Conitzer, Michael Dennis, Iason Gabriel, Adam Gleave, Gillian Hadfield, Nika Haghtalab, Atoosa Kasirzadeh, Sébastien Krier, Kate Larson, Joel Lehman, David C. Parkes, Georgios Piliouras, Iyad Rahwan

The rapid development of advanced AI agents and the imminent deployment of many instances of these agents will give rise to multi-agent systems of unprecedented complexity.

Ethics

Can Go AIs be adversarially robust?

2 code implementations • 18 Jun 2024 • Tom Tseng, Euan McLean, Kellin Pelrine, Tony T. Wang, Adam Gleave

Prior work found that superhuman Go AIs can be defeated by simple adversarial strategies, especially "cyclic" attacks.

Diversity

Exploiting Novel GPT-4 APIs

1 code implementation • 21 Dec 2023 • Kellin Pelrine, Mohammad Taufeeque, Michał Zając, Euan McLean, Adam Gleave

Language model attacks typically assume one of two extreme threat models: full white-box access to model weights, or black-box access limited to a text generation API.

Language Modeling Language Modelling +2

Inverse Scaling: When Bigger Isn't Better

no code implementations • 15 Jun 2023 • Ian R. McKenzie, Alexander Lyzhov, Michael Pieler, Alicia Parrish, Aaron Mueller, Ameya Prabhu, Euan McLean, Aaron Kirtland, Alexis Ross, Alisa Liu, Andrew Gritsevskiy, Daniel Wurgaft, Derik Kauffman, Gabriel Recchia, Jiacheng Liu, Joe Cavanagh, Max Weiss, Sicong Huang, The Floating Droid, Tom Tseng, Tomasz Korbak, Xudong Shen, Yuhui Zhang, Zhengping Zhou, Najoung Kim, Samuel R. Bowman, Ethan Perez

Here, we present evidence for the claim that LMs may show inverse scaling, or worse task performance with increased scale, e. g., due to flaws in the training objective and data.

Language models are better than humans at next-token prediction

1 code implementation • 21 Dec 2022 • Buck Shlegeris, Fabien Roger, Lawrence Chan, Euan McLean

Current language models are considered to have sub-human capabilities at natural language tasks like question-answering or writing code.

Question Answering