You Look Like a Thing and I Love You

by Janelle Shane

  7. Lehman et al., “The Surprising Creativity of Digital Evolution.”

  8. David Clements (@davecl42), Twitter, March 18, 2018, https://twitter.com/davecl42/status/975406071182479361.

  9. Nick Cheney et al., “Unshackling Evolution: Evolving Soft Robots with Multiple Materials and a Powerful Generative Encoding,” ACM SIGEVOlution 7, no. 1 (August 2014): 11–23, https://doi.org/10.1145/2661735.2661737.

  10. John Timmer, “Meet Wolbachia: The Male-Killing, Gender-Bending, Gonad-Eating Bacteria,” Ars Technica, October 24, 2011, https://arstechnica.com/science/news/2011/10/meet-wolbachia-the-male-killing-gender-bending-gonad-chomping-bacteria.ars.

  11. @forgek_, Twitter, October 10, 2018, https://twitter.com/forgek_/status/1050045261563813888.

  12. R. Feldt, “Generating Diverse Software Versions with Genetic Programming: An Experimental Study,” IEE Proceedings—Software 145, no. 6 (December 1998): 228–36, https://doi.org/10.1049/ip-sen:19982444.

  13. George Johnson, “Eurisko, the Computer With a Mind of Its Own,” Alicia Patterson Foundation,” updated April 6, 2011, https://aliciapatterson.org/stories/eurisko-computer-mind-its-own.

  14. Eric Schulte, Stephanie Forrest, and Westley Weimer, “Automated Program Repair through the Evolution of Assembly Code,” Proceedings of the IEEE/ACM International Conference on Automated Software Engineering, ASE ’10 (New York, NY: ACM, 2010), 313–316, https://doi.org/10.1145/1858996.1859059.

  Chapter 7: Unfortunate shortcuts

  1. Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin, “‘Why Should I Trust You?’: Explaining the Predictions of Any Classifier,” ArXiv:1602.04938 [Cs, Stat], February 16, 2016, http://arxiv.org/abs/1602.04938.

  2. Luke Oakden-Rayner, “Exploring the ChestXray14 Dataset: Problems,” Luke Oakden-Rayner (blog), December 18, 2017, https://lukeoakdenrayner.wordpress.com/2017/12/18/the-chestxray14-dataset-problems/.

  3. David M. Lazer et al., “The Parable of Google Flu: Traps in Big Data Analysis,” Science 343, no. 6176 (March 14, 2014): 1203–5, https://doi.org/10.1126/science.1248506.

  4. Gidi Shperber, “What I’ve Learned from Kaggle’s Fisheries Competition,” Medium, May 1, 2017, https://medium.com/@gidishperber/what-ive-learned-from-kaggle-s-fisheries-competition-92342f9ca779.

  5. J. Bird and P. Layzell, “The Evolved Radio and Its Implications for Modelling the Evolution of Novel Sensors,” Proceedings of the 2002 Congress on Evolutionary Computation, CEC’02 (Cat. No.02TH8600) vol. 2 (2002 World Congress on Computational Intelligence—WCCI’02, Honolulu, HI, USA: IEEE, 2002): 1836–41, https://doi.org/10.1109/CEC.2002.1004522.

  6. Hannah Fry, Hello World: Being Human in the Age of Algorithms (New York: W. W. Norton & Company, 2018).

  7. Lo Bénichou, “The Web’s Most Toxic Trolls Live in… Vermont?,” Wired, August 22, 2017, https://www.wired.com/2017/08/internet-troll-map/.

  8. Violet Blue, “Google’s Comment-Ranking System Will Be a Hit with the Alt-Right,” Engadget, September 1, 2017, https://www.engadget.com/2017/09/01/google-perspective-comment-ranking-system/.

  9. Jessamyn West (@jessamyn), Twitter, August 24, 2017, https://twitter.com/jessamyn/status/900867154412699649.

  10. Robyn Speer, “ConceptNet Numberbatch 17.04: Better, Less-Stereotyped Word Vectors,” ConceptNet blog, April 24, 2017, http://blog.conceptnet.io/posts/2017/conceptnet-numberbatch-17-04-better-less-stereotyped-word-vectors/.

  11. Aylin Caliskan, Joanna J. Bryson, and Arvind Narayanan, “Semantics Derived Automatically from Language Corpora Contain Human-like Biases,” Science 356, no. 6334 (April 14, 2017): 183–86, https://doi.org/10.1126/science.aal4230.

  12. Anthony G. Greenwald, Debbie E. McGhee, and Jordan L. K. Schwartz, “Measuring Individual Differences in Implicit Cognition: The Implicit Association Test,” Journal of Personality and Social Psychology 74 (June 1998): 1464–80.

  13. Brian A. Nosek, Mahzarin R. Banaji, and Anthony G. Greenwald, “Math = Male, Me = Female, Therefore Math Not = Me,” Journal of Personality and Social Psychology 83, no. 1 (July 2002): 44–59.

  14. Speer, “ConceptNet Numberbatch 17.04.”

  15. Larson et al., “How We Analyzed the COMPAS.”

  16. Jeff Larson and Julia Angwin, “Bias in Criminal Risk Scores Is Mathematically Inevitable, Researchers Say,” ProPublica, December 30, 2016, https://www.propublica.org/article/bias-in-criminal-risk-scores-is-mathematically-inevitable-researchers-say.

  17. James Regalbuto, “Insurance Circular Letter No. 1 (2019),” New York State Department of Financial Services, January 18, 2019, https://www.dfs.ny.gov/industry_guidance/circular_letters/cl2019_01.

  18. Jeffrey Dastin, “Amazon Scraps Secret AI Recruiting Tool That Showed Bias against Women,” Reuters, October 10, 2018, https://www.reuters.com/article/us-amazon-com-jobs-automation-insight-idUSKCN1MK08G.

  19. James Vincent, “Amazon Reportedly Scraps Internal AI Recruiting Tool That Was Biased against Women,” The Verge, October 10, 2018, https://www.theverge.com/2018/10/10/17958784/ai-recruiting-tool-bias-amazon-report.

  20. Paola Cecchi-Dimeglio, “How Gender Bias Corrupts Performance Reviews, and What to Do About It,” Harvard Business Review, April 12, 2017, https://hbr.org/2017/04/how-gender-bias-corrupts-performance-reviews-and-what-to-do-about-it.

  21. Dave Gershgorn, “Companies Are on the Hook If Their Hiring Algorithms Are Biased,” Quartz, October 22, 2018, https://qz.com/1427621/companies-are-on-the-hook-if-their-hiring-algorithms-are-biased/.

  22. Karen Hao, “Police across the US Are Training Crime-Predicting AIs on Falsified Data,” MIT Technology Review, February 13, 2019, https://www.technologyreview.com/s/612957/predictive-policing-algorithms-ai-crime-dirty-data/.

  23. Steve Lohr, “Facial Recognition Is Accurate, If You’re a White Guy,” New York Times, February 9, 2018, https://www.nytimes.com/2018/02/09/technology/facial-recognition-race-artificial-intelligence.html.

  24. Julia Carpenter, “Google’s Algorithm Shows Prestigious Job Ads to Men, but Not to Women. Here’s Why That Should Worry You,” Washington Post, July 6, 2015, https://www.washingtonpost.com/news/the-intersect/wp/2015/07/06/googles-algorithm-shows-prestigious-job-ads-to-men-but-not-to-women-heres-why-that-should-worry-you/.

  25. Mark Wilson, “This Breakthrough Tool Detects Racism and Sexism in Software,” Fast Company, August 22, 2017, https://www.fastcompany.com/90137322/is-your-software-secretly-racist-this-new-tool-can-tell.

  26. ORCAA, accessed August 3, 2019, http://www.oneilrisk.com.

  27. Faisal Kamiran and Toon Calders, “Data Preprocessing Techniques for Classification without Discrimination,” Knowledge and Information Systems 33, no. 1 (October 1, 2012): 1–33, https://doi.org/10.1007/s10115-011-0463-8.

  Chapter 8: Is an AI brain like a human brain?

  1. Ha and Schmidhuber, “World Models.”

  2. Anthony J. Bell and Terrence J. Sejnowski, “The ‘Independent Components’ of Natural Scenes Are Edge Filters,” Vision Research 37, no. 23 (December 1, 1997): 3327–38, https://doi.org/10.1016/S0042-6989(97)00121-1.

  3. Andrea Banino et al., “Vector-Based Navigation Using Grid-Like Representations in Artificial Agents,” Nature 557, no. 7705 (May 2018): 429–33, https://doi.org/10.1038/s41586-018-0102-6.

  4. Bau et al., “GAN Dissection.”

  5. Larry S. Yaeger, “Computational Genetics, Physiology, Metabolism, Neural Systems, Learning, Vision, and Behavior or PolyWorld: Life in a New Context,” Santa Fe Institute Studies in the Sciences of Complexity, vol. 17 (Los Alamos, NM: Addison-Wesley Publishing Company, 1994), 262–63.

  6. Baba Narumi et al., “Trophic Eggs Compensate for Poor Offspring Feeding Capacity in a Subsocial Burrower Bug,” Biology Letters 7, no. 2 (April 23, 2011): 194–96, https://doi.org/10.1098/rsbl.2010.0707.

  7. Robert M. French, “Catastrophic Forgetting in Connectionist Networks,” Trends in Cognitive Sciences 3, no. 4 (April 1999): 128–35.

  8. Jieyu Zhao et al., “Men Also Like Shopping: Reducing Gender Bias Amplification Using Cor
pus-Level Constraints,” ArXiv:1707.09457 [Cs, Stat], July 28, 2017, http://arxiv.org/abs/1707.09457.

  9. Danny Karmon, Daniel Zoran, and Yoav Goldberg, “LaVAN: Localized and Visible Adversarial Noise,” ArXiv:1801.02608 [Cs], January 8, 2018, http://arxiv.org/abs/1801.02608.

  10. Andrew Ilyas et al., “Black-Box Adversarial Attacks with Limited Queries and Information,” ArXiv:1804.08598 [Cs, Stat], April 23, 2018, http://arxiv.org/abs/1804.08598.

  11. Battista Biggio et al., “Poisoning Behavioral Malware Clustering,” ArXiv:1811.09985 [Cs, Stat], November 25, 2018, http://arxiv.org/abs/1811.09985.

  12. Tom White, “Synthetic Abstractions,” Medium, August 23, 2018, https://medium.com/@tom_25234/synthetic-abstractions-8f0e8f69f390.

  13. Samuel G. Finlayson et al., “Adversarial Attacks Against Medical Deep Learning Systems,” ArXiv:1804.05296 [Cs, Stat], April 14, 2018, http://arxiv.org/abs/1804.05296.

  14. Philip Bontrager et al., “DeepMasterPrints: Generating MasterPrints for Dictionary Attacks via Latent Variable Evolution,” ArXiv:1705.07386 [Cs], May 20, 2017, http://arxiv.org/abs/1705.07386.

  15. Stephen Buranyi, “How to Persuade a Robot That You Should Get the Job,” The Observer, March 4, 2018, https://www.theguardian.com/technology/2018/mar/04/robots-screen-candidates-for-jobs-artificial-intelligence.

  16. Lauren Johnson, “4 Deceptive Mobile Ad Tricks and What Marketers Can Learn From Them,” Adweek, February 16, 2018, https://www.adweek.com/digital/4-deceptive-mobile-ad-tricks-and-what-marketers-can-learn-from-them/.

  17. Wieland Brendel and Matthias Bethge, “Approximating CNNs with Bag-of-Local-Features Models Works Surprisingly Well on ImageNet,” ArXiv:1904.00760 [Cs, Stat], March 20, 2019, http://arxiv.org/abs/1904.00760.

  Chapter 9: Human bots (where can you not expect to see AI?)

  1. @yoco68, Twitter, July 12, 2018, https://twitter.com/yoco68/status/1017404857190268928.

  2. Parmy Olson, “Nearly Half of All ‘AI Startups’ Are Cashing in on Hype,” Forbes, March 4, 2019, https://www.forbes.com/sites/parmyolson/2019/03/04/nearly-half-of-all-ai-startups-are-cashing-in-on-hype/#5b1c4a66d022.

  3. Carolyn Said, “Kiwibots Win Fans at UC Berkeley as They Deliver Fast Food at Slow Speeds,” San Francisco Chronicle, May 26, 2019, https://www.sfchronicle.com/business/article/Kiwibots-win-fans-at-UC-Berkeley-as-they-deliver-13895867.php.

  4. Olivia Solon, “The Rise of ‘Pseudo-AI’: How Tech Firms Quietly Use Humans to Do Bots’ Work,” The Guardian, July 6, 2018, https://www.theguardian.com/technology/2018/jul/06/artificial-intelligence-ai-humans-bots-tech-companies.

  5. Ellen Huet, “The Humans Hiding Behind the Chatbots,” Bloomberg.com, April 18, 2016, https://www.bloomberg.com/news/articles/2016-04-18/the-humans-hiding-behind-the-chatbots.

  6. Richard Wray, “SpinVox Answers BBC Allegations over Use of Humans Rather than Machines,” The Guardian, July 23, 2009, https://www.theguardian.com/business/2009/jul/23/spinvox-answer-back.

  7. Becky Lehr (@Breakaribecca), Twitter, July 7, 2018, https://twitter.com/Breakaribecca/status/1015787072102289408.

  8. Paul Mozur, “Inside China’s Dystopian Dreams: A.I., Shame and Lots of Cameras,” New York Times, July 8, 2018, https://www.nytimes.com/2018/07/08/business/china-surveillance-technology.html.

  9. Aaron Mamiit, “Facebook AI Invents Language That Humans Can’t Understand: System Shut Down Before It Evolves Into Skynet,” Tech Times, July 30, 2017, http://www.techtimes.com/articles/212124/20170730/facebook-ai-invents-language-that-humans-cant-understand-system-shut-down-before-it-evolves-into-skynet.htm.

  10. Kyle Wiggers, “Babysitter Screening App Predictim Uses AI to Sniff out Bullies,” VentureBeat (blog), October 4, 2018, https://venturebeat.com/2018/10/04/babysitter-screening-app-predictim-uses-ai-to-sniff-out-bullies/.

  11. Chelsea Gohd, “Here’s What Sophia, the First Robot Citizen, Thinks About Gender and Consciousness,” Live Science, July 11, 2018, https://www.livescience.com/63023-sophia-robot-citizen-talks-gender.html.

  12. C. D. Martin, “ENIAC: Press Conference That Shook the World,” IEEE Technology and Society Magazine 14, no. 4 (Winter 1995): 3–10, https://doi.org/10.1109/44.476631.

  13. Alexandra Petri, “A Bot Named ‘Eugene Goostman’ Passes the Turing Test… Kind Of,” Washington Post, June 9, 2014, https://www.washingtonpost.com/blogs/compost/wp/2014/06/09/a-bot-named-eugene-goostman-passes-the-turing-test-kind-of/.

  14. Brian Merchant, “Predictim Claims Its AI Can Flag ‘Risky’ Babysitters. So I Tried It on the People Who Watch My Kids,” Gizmodo, December 6, 2018, https://gizmodo.com/predictim-claims-its-ai-can-flag-risky-babysitters-so-1830913997.

  15. Drew Harwell, “AI Start-up That Scanned Babysitters Halts Launch Following Post Report,” Washington Post, December 14, 2018, https://www.washingtonpost.com/technology/2018/12/14/ai-start-up-that-scanned-babysitters-halts-launch-following-post-report/.

  16. Tonya Riley, “Get Ready, This Year Your Next Job Interview May Be with an A.I. Robot,” CNBC, March 13, 2018, https://www.cnbc.com/2018/03/13/ai-job-recruiting-tools-offered-by-hirevue-mya-other-start-ups.html.

  17. Ibid.

  Chapter 10: A human-AI partnership

  1. Thu Nguyen-Phuoc et al., “HoloGAN: Unsupervised Learning of 3D Representations from Natural Images,” ArXiv:1904.01326 [Cs], April 2, 2019, http://arxiv.org/abs/1904.01326.

  2. Drew Linsley et al., “Learning What and Where to Attend,” ArXiv:1805.08819 [Cs], May 22, 2018, http://arxiv.org/abs/1805.08819.

  3. Hector Yee (@eigenhector), Twitter, September 14, 2018, https://twitter.com/eigenhector/status/1040501195989831680.

  4. Will Knight, “A Tougher Turing Test Shows That Computers Still Have Virtually No Common Sense,” MIT Technology Review, July 14, 2016, https://www.technologyreview.com/s/601897/tougher-turing-test-exposes-chatbots-stupidity/.

  5. James Regalbuto, “Insurance Circular Letter.”

  6. Abby Ohlheiser, “Trolls Turned Tay, Microsoft’s Fun Millennial AI Bot, into a Genocidal Maniac,” Chicago Tribune, March 26, 2016, https://www.chicagotribune.com/business/ct-internet-breaks-microsoft-ai-bot-tay-20160326-story.html.

  7. Glen Levy, “Google’s Bizarre Search Helper Assumes We Have Parakeets, Diarrhea,” Time, November 4, 2010, http://newsfeed.time.com/2010/11/04/why-why-wont-my-parakeet-eat-my-diarrhea-is-on-google-trends/.

  8. Michael Eisen, “Amazon’s $23,698,655.93 Book about Flies,” It Is NOT Junk (blog), April 22, 2011, http://www.michaeleisen.org/blog/?p=358.

  9. Emilio Calvano et al., “Artificial Intelligence, Algorithmic Pricing, and Collusion,” VoxEU (blog), February 3, 2019, https://voxeu.org/article/artificial-intelligence-algorithmic-pricing-and-collusion.

  10. Solon, “The Rise of ‘Pseudo-AI.’”

  11. Gale M. Lucas et al., “It’s Only a Computer: Virtual Humans Increase Willingness to Disclose,” Computers in Human Behavior 37 (August 1, 2014): 94–100, https://doi.org/10.1016/j.chb.2014.04.043.

  12. Liliana Laranjo et al., “Conversational Agents in Healthcare: A Systematic Review,” Journal of the American Medical Informatics Association 25, no. 9 (September 1, 2018): 1248–58, https://doi.org/10.1093/jamia/ocy072.

  13. Margi Murphy, “Artificial Intelligence Will Detect Child Abuse Images to Save Police from Trauma,” The Telegraph, December 18, 2017, https://www.telegraph.co.uk/technology/2017/12/18/artificial-intelligence-will-detect-child-abuse-images-save/.

  14. Adam Zewe, “In Automaton We Trust,” Harvard School of Engineering and Applied Science, May 25, 2016, https://www.seas.harvard.edu/news/2016/05/in-automaton-we-trust.

  15. David Streitfeld, “Computer Stories: A.I. Is Beginning to Assist Novelists,” New York Times, October 18, 2018, https://www.nytimes.com/2018/10/18/technology/ai-is-beginning-to-assist-novelists.html.

  * The old adage that a monkey writing randomly on a typewriter for an infinite amount of time will eventually produce the entire works of Shakespeare actually pretty accurately describes the “brute force” method of searching for a s
olution to a problem by systematically trying everything. Ideally, using AI to solve the problem is an improvement over this. Ideally.

  * The fact that the score is 27–0 at this point rather than 28–0 means that the Cougars might have missed one of their conversion points—a fact that Heliograf fails to mention.

  * It also had a tiny bit of long-term memory where it could keep track of information for longer than that sixty five-character window, but that amount of memory was too tiny to store an entire ingredients list. In machine learning terms, that makes this algorithm a Long Short-Term Memory (LSTM) neural network rather than a plain RNN.

  * The category was spelled “deserts” rather than “desserts” in the dataset, so this is how the neural net thinks it’s spelled.

  * The Google Translate algorithm is constantly being updated, so these results will change significantly over time.