SIMA 2: A Gemini-Powered AI Agent for 3D Digital Worlds

Acknowledgements

This analysis was developed by the SIMA 2 group: Maria Abi Raad, John Agapiou, Frederic Besse, Andrew Bolt, Sarah Chakera, Harris Chan, Jeff Clune, Alexandra Cordell, Martin Engelcke, Ryan Faulkner, Maxime Gazeau, Arne Olav Hallingstad, Tim Harley, Ed Hirst, Drew Hudson, Laura Kampis, Sheleem Kashem, Thomas Keck, Matija Kecman, Oscar Knagg, Alexander Lerchner, Bonnie Li, Yulan Liu, Cong Lu, Maria Loks-Thompson, Joseph Marino, Kay McKinney, Piermaria Mendolicchio, Anna Mitenkova, Alexandre Moufarek, Fabio Pardo, Ollie Purkiss, David Reichert, John Reid, Tyson Roberts, Daniel P. Sawyer, Tim Scholtes, Daniel Slater, Hubert Soyer, Kaustubh Sridhar, Peter Stys, Tayfun Terzi, Davide Vercelli, Bojan Vujatovic, Jane X. Wang, Luyu Wang, Duncan Williams, and Lei M. Zhang.

For his or her management, steering, and assist, we thank: Satinder Singh Baveja, Adrian Bolton, Zoubin Ghahramani, Raia Hadsell, Demis Hassabis, Shane Legg, Volodymyr Mnih, and Daan Wierstra.

With a lot gratitude to partial contributors and previous members: Alex Cullum, Karol Gregor, Rosemary Ke, Junkyung Kim, Matthew Jackson, Andrew Lampinen, Loic Matthey, Hannah Openshaw, and Zhengdong Wang.

Particular due to all the sport builders who partnered with us: Espresso Stain (Valheim, Passable, Goat Simulator 3), Foulball Hangover (Hydroneer), Hi there Video games (No Man’s Sky), Eager Software program Home (Area Engineers), RubberbandGames (Wobbly Life), Unusual Loop Video games (Eco), Thunderful Video games (ASKA, The Gunk, Steamworld Construct), Digixart (Street 96), and Tuxedo Labs & Saber Interactive (Teardown).

We thank Vika Koriakin, Duncan Smith, Nilesh Ray, Matt Miller, Leen Verburgh, Ashyana Kachra, Phil Esposito, Dimple Vijaykumar, Piers Wingfield, Lucie Kerley for his or her invaluable partnership in creating and refining key parts of this mission.

We additionally thank Jack Parker-Holder, Shlomi Fruchter, and the remainder of the Genie group for entry to the Genie 3 mannequin.

We’d like to acknowledge the numerous groups throughout Google and Google DeepMind which have contributed to this effort together with Authorized, Advertising and marketing, Communications, Accountability and Security Council, Accountable Improvement and Innovation, Coverage, Technique and Operations, and our Enterprise and Company Improvement groups. We might additionally prefer to thank all GDM groups that aren’t explicitly talked about right here for his or her continued assist.

Lastly, we dedicate this work to the reminiscence of our colleagues Felix Hill and Fabio Pardo, whose contributions to our area proceed to encourage us.