Lile Procedural People Playing politics with generators Kate Compton University of California, Santa Cruz Santa Cruz, CA 95064 kecompto@ucsc.edu ABSTRACT Do generators have politics? What about generators that generate around people, with people, or even create generative people. is paper proposes four initial sites of inquiry that deserve further aention from this community, or at least those members who nd themselves building a person-generator: characters who engage so- cially with people, generators which make use of data created by or about people, the use of cultural and social signiers in generators, and simulations or models which represent people. CCS CONCEPTS •Computing methodologies → Procedural animation; KEYWORDS Generativity, prototyping, interaction design, ethics ACM Reference format: Kate Compton. 2017. Lile Procedural People. In Proceedings of FDG’17, Hyannis, MA, USA, August 14-17, 2017, 2 pages. DOI: 10.1145/3102071.3110573 e procedural game-content generation community (and our neigh- bors in generative art, generative text, and computational creativity) are used to generating many kinds of things. We generate buildings and landscapes, trees and owers, creatures, animations, dances, game levels, music, and poetry. What happens when the things that we generate are people? ere are several ways that we use people in our generators: • We create characters that act or speak like people • We use real locations or real user content as an input • We use cultural and social signiers in our generators • We model a possibility space of “what people can be” is paper proposes these as four sites where generativity inter- sects with political and social responsibility. Building generators can be a fun and expressive practice, but as a community we should build our generators with consideration and awareness when we are generating around, with, about, or in communication with people. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for prot or commercial advantage and that copies bear this notice and the full citation on the rst page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permied. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specic permission and/or a fee. Request permissions from permissions@acm.org. FDG’17, Hyannis, MA, USA © 2017 Copyright held by the owner/author(s). Publication rights licensed to ACM. 978-1-4503-5319-9/17/08. . . $15.00 DOI: 10.1145/3102071.3110573 1 GENERATIVITY IN SOCIAL ENVIRONMENTS Of the many communities creating generative artifacts, the chatbot and twierbot-making community has established conversations about the political and social implications of generativity[3]. A popular expectation in this community (of practitioners, industry, and audience) is that we are aempting to build a human-like –or at least personable– character, so it is not a large jump to imagine that such a character could need all of social considerations a human would have when operating in the same space. Oen bots are coded as male or female, professional artist, young teen, or servant [5], priming the interactor to read their generativity through that lens of social expectations. On Twier, bots and humans can easily interact on equal terms, using identical social tools (text and image posts, likes, retweets, follows). is equality come with a set of expectations: that which it is unethical for humans to do is also unethical for bots to do, and unethical things that are possible for humans are also possible for bots. Not only must bots consider what they generate, but how they post it. A bot which broadcasts its productions into a social platform has a dierent set of ethical considerations than a generator living on a webpage or in a game. Each bot follows its own rules for what it says and when it re- sponds. e Twierbot @infinite scream will respond to users’ direct tweets at it, but can only respond with variations of ”aaaah”, making it a safe and reliable conversational partner. Other bots will interject into conversations with reinterpretations of users words (@godtributes) or repost tweets to ll a generative tem- plate (@pentametron), but only of users who have ”consented” by following them. From a combination of their rules of engagement and their rules of generativity, each of these bots has constructed a character which engages socially with the human users of Twier. 2 REAL LOCATIONS, REAL CONTENT, REAL ISSUES Many generative works scrape real-world data (as godtributes does with tweets). Others use real-world locations through Google maps APIs or augmented-reality overlays on physical space. When this works well, like Pokemon Go, there is a sense of a magical ”alternate reality” co-existing with our own. Normal spaces like bus-stops become uplied and turned into game spaces. But many spaces are owned, or meaningful to the people already in them. Constructing a virtual side to an existing place is not apolitical, as when Pokemon started appearing in the Holocaust Museum. One project already ran into this, using technology that I built: the ”Every Rat in NYC (at People Complained About in 2016)”