Bibliography

Every source cited in AI Safety Atlas: 963 references, by label.

1. 1985 — cited in Chapter 4: Governance
2. 1a3orn, 2023 — cited in Chapter 3: Strategies
3. Aarne et al., 2024 — cited in Chapter 4: Governance
4. Abid et al., 2021 — cited in Chapter 6: Specification Gaming
5. Abraham, 2024 — cited in Chapter 2: Risks
6. Adan et al., 2024 — cited in Chapter 4: Governance
7. Aguirre, 2025 — cited in Chapter 2: Risks
8. Aguirre, 2025 — cited in Chapter 3: Strategies
9. AI Digest, 2023 — cited in Chapter 1: Capabilities
10. AI Digest, 2024 — cited in Chapter 5: Evaluations
11. AI Impacts, 2022 — cited in Chapter 1: Capabilities
12. AI Impacts, 2022 — cited in Chapter 1: Capabilities
13. AI Incident Database, 2025 — cited in Chapter 2: Risks
14. AI Index Report, 2025 — cited in Chapter 1: Capabilities
15. AI Optimists, 2023 — cited in Chapter 3: Strategies
16. AI Safety in China, 2025 — cited in Chapter 4: Governance
17. AISI, 2025 — cited in Chapter 3: Strategies
18. Akash, 2023 — cited in Chapter 3: Strategies
19. Alexander, 2023 — cited in Chapter 3: Strategies
20. Althaus & Gloor, 2016 — cited in Chapter 2: Risks
21. Altman, 2023 — cited in Chapter 1: Capabilities
22. Amazon, 2023 — cited in Chapter 1: Capabilities
23. Amazon, 2025 — cited in Chapter 1: Capabilities
24. Ammann, 2025 — cited in Chapter 3: Strategies
25. Amodei, 2023 — cited in Chapter 1: Capabilities
26. Anderljung et al. 2024 — cited in Chapter 4: Governance
27. Anderljung et al., 2023 — cited in Chapter 3: Strategies, Chapter 4: Governance
28. Anderljung et al., 2023 — cited in Chapter 4: Governance
29. Anderljung, 2023 — cited in Chapter 4: Governance
30. Anderljung, 2023 — cited in Chapter 4: Governance
31. Anderljung, 2024 — cited in Chapter 4: Governance
32. Andreas 2022 — cited in Chapter 7: Goal Misgeneralization
33. Andriushchenko et al., 2024 — cited in Chapter 5: Evaluations
34. Andy Zou et al., 2024 — cited in Chapter 3: Strategies
35. Anna Desmarais, 2024 — cited in Chapter 3: Strategies
36. Anthropic, 2023 — cited in Chapter 1: Capabilities
37. Anthropic, 2023 — cited in Chapter 5: Evaluations
38. Anthropic, 2023 — cited in Chapter 5: Evaluations
39. Anthropic, 2023 — cited in Chapter 6: Specification Gaming
40. Anthropic, 2024 — cited in Chapter 1: Capabilities
41. Anthropic, 2024 — cited in Chapter 1: Capabilities
42. Anthropic, 2024 — cited in Chapter 5: Evaluations
43. Anthropic, 2024 — cited in Chapter 5: Evaluations
44. Anthropic, 2024 — cited in Chapter 5: Evaluations
45. Anthropic, 2024 — cited in Chapter 5: Evaluations
46. Anthropic, 2024 — cited in Chapter 5: Evaluations
47. Anthropic, 2024 — cited in Chapter 5: Evaluations
48. Anthropic, 2024 — cited in Chapter 5: Evaluations
49. Anthropic, 2025 — cited in Chapter 1: Capabilities
50. Anthropic, 2025 — cited in Chapter 1: Capabilities
51. Anthropic, 2025 — cited in Chapter 1: Capabilities
52. Anthropic, 2025 — cited in Chapter 3: Strategies, Chapter 7: Goal Misgeneralization
53. Anthropic, 2025 — cited in Chapter 3: Strategies
54. Anthropic, 2025 — cited in Chapter 5: Evaluations
55. Anthropic, 2025 — cited in Chapter 5: Evaluations, Chapter 7: Goal Misgeneralization
56. Anthropic, 2025 — cited in Chapter 5: Evaluations
57. AP News, 2017 — cited in Chapter 4: Governance
58. Apollo Research, 2024 — cited in Chapter 5: Evaluations
59. Apvrille & Nakov, 2025 — cited in Chapter 2: Risks
60. ARC-AGI, 2024 — cited in Chapter 1: Capabilities
61. ArtificialAnalysis, 2025 — cited in Chapter 1: Capabilities
62. Aschenbrenner, 2024 — cited in Chapter 1: Capabilities, Chapter 2: Risks
63. Aschenbrenner, 2024 — cited in Chapter 1: Capabilities
64. Aschenbrenner, 2024 — cited in Chapter 2: Risks, Chapter 3: Strategies
65. Askell et al., 2024 — cited in Chapter 2: Risks, Chapter 4: Governance
66. Badie et al., 2011 — cited in Chapter 4: Governance
67. Bai et al., 2022 — cited in Chapter 6: Specification Gaming
68. Baker et al., 2020 — cited in Chapter 2: Risks
69. Baker et al., 2025 — cited in Chapter 3: Strategies
70. Bakhtin et al., 2022 — cited in Chapter 1: Capabilities
71. Bansal et al., 2022 — cited in Chapter 2: Risks
72. Barnes et al., 2020 — cited in Chapter 8: Scalable Oversight
73. Barnes et al., 2020 — cited in Chapter 8: Scalable Oversight
74. Barnes, 2023 — cited in Chapter 7: Goal Misgeneralization
75. Barnett & Thiergart, 2024 — cited in Chapter 4: Governance
76. Barnett, 2025 — cited in Chapter 2: Risks
77. Barnett, 2025 — cited in Chapter 2: Risks
78. Barnett, 2025 — cited in Chapter 4: Governance
79. Barros, 2025 — cited in Chapter 2: Risks
80. Baumann, 2017 — cited in Chapter 2: Risks
81. BBC, 2023 — cited in Chapter 6: Specification Gaming
82. Belfield & Hua 2022 — cited in Chapter 4: Governance
83. Bengio et al., 2023 — cited in Chapter 1: Capabilities
84. Bengio et al., 2025 — cited in Chapter 4: Governance
85. Bengio et al., 2025 — cited in Chapter 4: Governance
86. Bengio, 2023 — cited in Chapter 3: Strategies
87. Beraja et al., 2023, AI-tocracy — cited in Chapter 2: Risks
88. Beren, 2023 — cited in Chapter 7: Goal Misgeneralization
89. Berglund et al., 2024 — cited in Chapter 5: Evaluations
90. Bernardi, 2024 — cited in Chapter 3: Strategies
91. Besiroglu et al., 2024 — cited in Chapter 4: Governance
92. Besiroglu et al., 2024 — cited in Chapter 5: Evaluations
93. Besta et al., 2023 — cited in Chapter 5: Evaluations
94. Betker et al., 2023 — cited in Chapter 1: Capabilities
95. Betley et al., 2025 — cited in Chapter 7: Goal Misgeneralization
96. Bhatt et al., 2023 — cited in Chapter 5: Evaluations
97. Bhatt et al., 2024 — cited in Chapter 5: Evaluations
98. Bhatt et al., 2024 — cited in Chapter 5: Evaluations
99. Binder et al., 2024 — cited in Chapter 2: Risks, Chapter 5: Evaluations
100. Blablová, 2025 — cited in Chapter 1: Capabilities
101. Black et al., 2025 — cited in Chapter 2: Risks
102. Bode & Watts, 2023 — cited in Chapter 2: Risks
103. Bogdan et al., 2025 — cited in Chapter 7: Goal Misgeneralization
104. Boiko et al., 2023 — cited in Chapter 4: Governance
105. Bommasani et al., 2022 — cited in Chapter 1: Capabilities
106. Bondarenko et al., 2025 — cited in Chapter 2: Risks, Chapter 3: Strategies
107. Boston Dynamics, 2024 — cited in Chapter 1: Capabilities
108. Boston Dynamics, 2024 — cited in Chapter 1: Capabilities
109. Bostrom, 2002 — cited in Chapter 2: Risks
110. Bostrom, 2012 — cited in Chapter 2: Risks
111. Bostrom, 2014 — cited in Chapter 1: Capabilities
112. Bostrom, 2014 — cited in Chapter 6: Specification Gaming
113. Bostrom, 2014 — cited in Chapter 7: Goal Misgeneralization
114. Bowman et al., 2022 — cited in Chapter 8: Scalable Oversight
115. Bowman, 2023 — cited in Chapter 1: Capabilities
116. Bowman, 2024 — cited in Chapter 3: Strategies
117. Bowman, 2024 — cited in Chapter 8: Scalable Oversight
118. Bowman, 2025 — cited in Chapter 3: Strategies
119. Bradford, 2020 — cited in Chapter 4: Governance
120. Brass & Aarne, 2024 — cited in Chapter 4: Governance
121. Brennan et al., 2025 — cited in Chapter 4: Governance
122. Brown et al., 2020 — cited in Chapter 1: Capabilities, Chapter 5: Evaluations
123. Brown, 2024 — cited in Chapter 5: Evaluations
124. Brundage et al., 2018 — cited in Chapter 4: Governance
125. Brundage et al., 2018 — cited in Chapter 4: Governance
126. Brundage, 2025 — cited in Chapter 3: Strategies
127. Brunskill, 2022 — cited in Chapter 6: Specification Gaming
128. Bubeck et al., 2023 — cited in Chapter 1: Capabilities
129. Buchanan, 2020 — cited in Chapter 4: Governance
130. Buhl et al., 2024 — cited in Chapter 4: Governance
131. Burden, 2024 — cited in Chapter 5: Evaluations
132. Burns et. al. 2023 — cited in Chapter 8: Scalable Oversight
133. Buterin, 2023 — cited in Chapter 3: Strategies
134. Buterin, 2025 — cited in Chapter 3: Strategies
135. Caballero et al., 2023 — cited in Chapter 1: Capabilities
136. CAIS, 2023 — cited in Chapter 2: Risks
137. Campos et al., 2024 — cited in Chapter 3: Strategies
138. Canell, 2015 — cited in Chapter 1: Capabilities
139. Canell, 2022 — cited in Chapter 1: Capabilities
140. Canell, 2022 — cited in Chapter 1: Capabilities
141. Cao et al., 2023 — cited in Chapter 3: Strategies
142. Cao et al., 2024 — cited in Chapter 6: Specification Gaming
143. Carlini et al., 2021 — cited in Chapter 2: Risks, Chapter 6: Specification Gaming
144. Carlsmith, 2020 — cited in Chapter 1: Capabilities
145. Carlsmith, 2021 — cited in Chapter 2: Risks
146. Carlsmith, 2023 — cited in Chapter 2: Risks, Chapter 5: Evaluations, Chapter 7: Goal Misgeneralization
147. Carlson, 2009 — cited in Chapter 2: Risks
148. Carter et al., 2023 — cited in Chapter 2: Risks
149. Carvalho et al., 2025 — cited in Chapter 7: Goal Misgeneralization
150. Casper et al., 2023 — cited in Chapter 3: Strategies
151. Casper et al., 2023 — cited in Chapter 3: Strategies, Chapter 4: Governance
152. Casper et al., 2024 — cited in Chapter 5: Evaluations
153. Casper, 2023 — cited in Chapter 3: Strategies
154. Casper, 2024 — cited in Chapter 4: Governance
155. catubc, 2024 — cited in Chapter 3: Strategies
156. Caucheteux & King, 2022 — cited in Chapter 1: Capabilities
157. Cave & Ó hÉigeartaigh, 2018 — cited in Chapter 2: Risks
158. Cerutti et al., 2025 — cited in Chapter 4: Governance
159. Ceruzzi, 1989 — cited in Chapter 2: Risks
160. CFG, 2025 — cited in Chapter 3: Strategies
161. Cha, 2024 — cited in Chapter 4: Governance
162. Chan et al., 2024 — cited in Chapter 4: Governance
163. Chan et al., 2024 — cited in Chapter 4: Governance
164. Chandra et al., 2024 — cited in Chapter 2: Risks
165. Charvet, 2021 — cited in Chapter 1: Capabilities
166. Cheerla, 2018 — cited in Chapter 2: Risks
167. Chehoudi, 2025 — cited in Chapter 4: Governance
168. Chen et al, 2021 — cited in Chapter 5: Evaluations
169. Chen et al., 2017 — cited in Chapter 2: Risks
170. Chen et al., 2022 — cited in Chapter 2: Risks
171. Chen et al., 2024 — cited in Chapter 2: Risks
172. Chen et al., 2025 — cited in Chapter 7: Goal Misgeneralization
173. Chen et al., 2025 — cited in Chapter 7: Goal Misgeneralization
174. Cheng & McKernon, 2024 — cited in Chapter 4: Governance
175. Cheng, 2024 — cited in Chapter 4: Governance
176. Cheng, 2024 — cited in Chapter 4: Governance
177. Cherny, 2025, Twitter — cited in Chapter 1: Capabilities
178. Chess.com, 2014 — cited in Chapter 2: Risks
179. Chollet et al., 2024 — cited in Chapter 5: Evaluations
180. Chollet et. al, 2024 — cited in Chapter 1: Capabilities, Chapter 2: Risks
181. Chollet, 2019 — cited in Chapter 1: Capabilities, Chapter 5: Evaluations
182. Chollet, 2023 — cited in Chapter 1: Capabilities
183. Chollet, 2024 — cited in Chapter 5: Evaluations
184. Chollet, 2025 — cited in Chapter 1: Capabilities, Chapter 5: Evaluations
185. Christiano et al., 2017 — cited in Chapter 6: Specification Gaming
186. Christiano Paul, 2016 — cited in Chapter 8: Scalable Oversight
187. Christiano, 2016 — cited in Chapter 8: Scalable Oversight
188. Christiano, 2016 — cited in Chapter 8: Scalable Oversight
189. Christiano, 2017 — cited in Chapter 3: Strategies
190. Christiano, 2018 — cited in Chapter 1: Capabilities
191. Christiano, 2018 — cited in Chapter 3: Strategies
192. Christiano, 2018 — cited in Chapter 3: Strategies
193. Christiano, 2019 — cited in Chapter 2: Risks
194. Christiano, 2019 — cited in Chapter 5: Evaluations
195. Christiano, 2019 — cited in Chapter 6: Specification Gaming
196. Christiano, 2020 — cited in Chapter 8: Scalable Oversight
197. Christiano, 2022 — cited in Chapter 3: Strategies
198. Christiano, 2023 — cited in Chapter 2: Risks
199. Christiano, 2023 — cited in Chapter 3: Strategies
200. Christiano, 2023 — cited in Chapter 3: Strategies
201. Cihon et al., 2020 — cited in Chapter 3: Strategies
202. Cihon et al., 2021 — cited in Chapter 4: Governance
203. Cihon, 2019 — cited in Chapter 4: Governance
204. Cihon, 2019 — cited in Chapter 4: Governance
205. Cima et al, 2010 — cited in Chapter 3: Strategies
206. CISA, 2021 — cited in Chapter 2: Risks
207. CISA, 2024 — cited in Chapter 2: Risks
208. Clark & Amodei,2016 — cited in Chapter 2: Risks, Chapter 6: Specification Gaming
209. Clymer et al, 2024 — cited in Chapter 3: Strategies
210. Clymer et al., 2024 — cited in Chapter 5: Evaluations, Chapter 7: Goal Misgeneralization
211. Clymer, 2025 — cited in Chapter 3: Strategies
212. Cobbe et al., 2019 — cited in Chapter 7: Goal Misgeneralization
213. Cobbe et al., 2021 — cited in Chapter 5: Evaluations
214. Conn, 2015 — cited in Chapter 2: Risks
215. Control AI, 2024 — cited in Chapter 3: Strategies
216. Corwin, 2002 — cited in Chapter 3: Strategies
217. Cotra, 2020 — cited in Chapter 1: Capabilities
218. Cotra, 2021 — cited in Chapter 7: Goal Misgeneralization
219. Cotra, 2021 — cited in Chapter 8: Scalable Oversight
220. Cotra, 2022 — cited in Chapter 4: Governance
221. Cotra, 2023 — cited in Chapter 1: Capabilities, Chapter 2: Risks
222. Cotra, 2023 — cited in Chapter 1: Capabilities
223. Cotra, 2023 — cited in Chapter 2: Risks, Chapter 4: Governance
224. Cottier et al., 2025 — cited in Chapter 1: Capabilities
225. Creswell et al., 2022 — cited in Chapter 1: Capabilities
226. Critch and Russel, 2023 — cited in Chapter 2: Risks, Chapter 3: Strategies
227. Critch, 2021 — cited in Chapter 2: Risks, Chapter 3: Strategies, Chapter 7: Goal Misgeneralization
228. Critch, 2022 — cited in Chapter 3: Strategies
229. Critch, 2023 — cited in Chapter 5: Evaluations
230. CrowdStrike, 2024 — cited in Chapter 2: Risks
231. Cumulative number of large-scale AI models by domain since 2017 — cited in Chapter 1: Capabilities
232. Cumulative number of large-scale AI systems by country since 2019 — cited in Chapter 4: Governance
233. Cunha & Estima, 2023 — cited in Chapter 2: Risks
234. Cunningham et al., 2023 — cited in Chapter 5: Evaluations, Chapter 7: Goal Misgeneralization
235. Dafoe, 2018 — cited in Chapter 4: Governance
236. Dafoe, 2020 — cited in Chapter 4: Governance
237. Dafoe, 2022 — cited in Chapter 4: Governance
238. Dafoe, 2023 — cited in Chapter 4: Governance
239. Dafoe, 2023 — cited in Chapter 4: Governance
240. Dai, 2019 — cited in Chapter 2: Risks
241. Dalrymple, 2022 — cited in Chapter 3: Strategies
242. Davidson et al, 2025 — cited in Chapter 3: Strategies
243. Davidson et al., 2023 — cited in Chapter 4: Governance, Chapter 5: Evaluations
244. Davidson, 2023 — cited in Chapter 1: Capabilities
245. Davidson, 2024 — cited in Chapter 1: Capabilities
246. Davidson, 2025 — cited in Chapter 3: Strategies
247. Davidson, 2025 — cited in Chapter 3: Strategies
248. DavidW, 2023 — cited in Chapter 7: Goal Misgeneralization
249. de Haan et al., 2019 — cited in Chapter 7: Goal Misgeneralization
250. Dean, 2025 — cited in Chapter 3: Strategies
251. DeepMind, 2016 — cited in Chapter 1: Capabilities, Chapter 6: Specification Gaming
252. DeepMind, 2018 — cited in Chapter 6: Specification Gaming
253. DeepMind, 2018 — cited in Chapter 6: Specification Gaming
254. DeepMind, 2019 — cited in Chapter 6: Specification Gaming
255. DeepMind, 2020 — cited in Chapter 1: Capabilities
256. DeepMind, 2022 — cited in Chapter 1: Capabilities
257. DeepMind, 2022 — cited in Chapter 7: Goal Misgeneralization
258. DeepMind, 2023 — cited in Chapter 1: Capabilities
259. DeepMind, 2024 — cited in Chapter 1: Capabilities
260. DeepMind, 2024 — cited in Chapter 2: Risks
261. DeepMind, 2024 — cited in Chapter 4: Governance
262. DeepMind, 2025 — cited in Chapter 1: Capabilities
263. DeepMind, 2025 — cited in Chapter 2: Risks
264. DeepSeek, 2025 — cited in Chapter 3: Strategies, Chapter 4: Governance
265. DeepSeek, 2025 — cited in Chapter 5: Evaluations
266. Defense Innovation Unit, 2023 — cited in Chapter 2: Risks
267. Delétang et al., 2023 — cited in Chapter 1: Capabilities
268. Deng, 2009 — cited in Chapter 5: Evaluations
269. Dev & Hobbhahn, 2024 — cited in Chapter 5: Evaluations
270. Dewey, 2011 — cited in Chapter 3: Strategies
271. DiGiovanni, 2023 — cited in Chapter 2: Risks
272. Ding & Dafoe, 2024 — cited in Chapter 4: Governance
273. Ding, 2018 — cited in Chapter 4: Governance
274. DnaScript, 2024 — cited in Chapter 2: Risks
275. Dong et al., 2024 — cited in Chapter 4: Governance
276. Douillard et al, 2023 — cited in Chapter 3: Strategies
277. Douillard et al, 2024 — cited in Chapter 3: Strategies
278. Douillard et al., 2024 — cited in Chapter 3: Strategies, Chapter 4: Governance
279. Dowie, 1977 — cited in Chapter 2: Risks
280. Dragan et al., 2024 — cited in Chapter 5: Evaluations
281. Edelman & Ho, 2025 — cited in Chapter 1: Capabilities
282. Egan & Heim, 2023 — cited in Chapter 3: Strategies, Chapter 4: Governance
283. Egan & Heim, 2023 — cited in Chapter 4: Governance
284. Eiras et al., 2024 — cited in Chapter 3: Strategies, Chapter 4: Governance
285. El-Mhamdi et al., 2023 — cited in Chapter 3: Strategies
286. Electricity generation by country and region in 2025 — cited in Chapter 1: Capabilities
287. Elmore, 2024 — cited in Chapter 1: Capabilities
288. Encyclopedia Britannica, 2025 — cited in Chapter 3: Strategies
289. Epicural, 2021 — cited in Chapter 6: Specification Gaming
290. Epoch AI, 2025 — cited in Chapter 1: Capabilities
291. Epoch AI, 2025 — cited in Chapter 1: Capabilities
292. Epoch AI, 2025 — cited in Chapter 1: Capabilities
293. Epoch AI, 2025 — cited in Chapter 1: Capabilities
294. Epoch AI, 2025 — cited in Chapter 1: Capabilities
295. Epoch AI, 2025 — cited in Chapter 1: Capabilities
296. Epoch AI, 2025 — cited in Chapter 1: Capabilities
297. Epoch AI, 2025 — cited in Chapter 1: Capabilities
298. EpochAI, 2025 — cited in Chapter 1: Capabilities
299. EpochAI, 2025 — cited in Chapter 1: Capabilities
300. EpochAI, 2025 — cited in Chapter 1: Capabilities
301. EpochAI, 2025 — cited in Chapter 3: Strategies
302. EpochAI, 2026 — cited in Chapter 1: Capabilities, Chapter 2: Risks, Chapter 5: Evaluations
303. Erdil & Besiroglu, 2024 — cited in Chapter 1: Capabilities
304. Erdil and Barnett, 2025 — cited in Chapter 2: Risks
305. Erdil et al., 2025 — cited in Chapter 1: Capabilities
306. Esvelt, 2022 — cited in Chapter 2: Risks
307. Ethayarajh et al., 2022 — cited in Chapter 6: Specification Gaming
308. Etienne & Flammarion, 2025 — cited in Chapter 7: Goal Misgeneralization
309. EU Commission, 2025 — cited in Chapter 1: Capabilities
310. European Commission, 2024 — cited in Chapter 5: Evaluations
311. Evans et al., 2023 — cited in Chapter 2: Risks, Chapter 5: Evaluations
312. Evans, 2024 — cited in Chapter 5: Evaluations
313. Everitt et al., 2025 — cited in Chapter 7: Goal Misgeneralization
314. Ewing, 2017 — cited in Chapter 5: Evaluations
315. Exfilbench, 2025 — cited in Chapter 3: Strategies
316. Eykholt et al., 2018 — cited in Chapter 2: Risks
317. Faggella, 2025 — cited in Chapter 3: Strategies
318. Fang et al., 2024 — cited in Chapter 2: Risks, Chapter 4: Governance
319. Farrell, 2024 — cited in Chapter 4: Governance
320. Federspiel et al., 2023 — cited in Chapter 2: Risks
321. Feldstein, 2019 — cited in Chapter 2: Risks
322. Feldstein, 2021 — cited in Chapter 2: Risks
323. Feng & Florian Tramèr, 2024 — cited in Chapter 2: Risks
324. Fenwick, 2023 — cited in Chapter 3: Strategies
325. Feuillade-Montixi & Peigné, 2023 — cited in Chapter 1: Capabilities
326. Field, 2025 — cited in Chapter 2: Risks
327. FLI, 2025 — cited in Chapter 2: Risks, Chapter 3: Strategies
328. Flint, 2022 — cited in Chapter 3: Strategies
329. Fluri et al., 2023 — cited in Chapter 1: Capabilities
330. Forethought, 2025 — cited in Chapter 3: Strategies
331. Friedman et al., 2007 — cited in Chapter 4: Governance
332. Friston et al., 2022 — cited in Chapter 2: Risks
333. Fu et al., 2024 — cited in Chapter 1: Capabilities
334. Future of Life Institute, 2024 — cited in Chapter 2: Risks
335. Future of Life Institute, 2024 — cited in Chapter 2: Risks
336. Gabriel et al., 2024 — cited in Chapter 2: Risks
337. Ganguli et al., 2022 — cited in Chapter 4: Governance
338. Gao, 2022 — cited in Chapter 6: Specification Gaming
339. Geirhos et al., 2019 — cited in Chapter 7: Goal Misgeneralization
340. Gennari et al., 2024 — cited in Chapter 5: Evaluations
341. Giattino et al., 2023 — cited in Chapter 1: Capabilities
342. Giattino et al., 2023 — cited in Chapter 1: Capabilities, Chapter 2: Risks, Chapter 4: Governance, Chapter 5: Evaluations
343. Giattino et al., 2023 — cited in Chapter 4: Governance
344. Gil, 2023 — cited in Chapter 3: Strategies
345. Glaese et al, 2022 — cited in Chapter 6: Specification Gaming
346. Glazer et al., 2024 — cited in Chapter 5: Evaluations
347. Gleave, 2025 — cited in Chapter 3: Strategies
348. Gnanasambandam et al, 2021 — cited in Chapter 2: Risks
349. Goemans et al., 2024 — cited in Chapter 3: Strategies
350. Goertzel & Pitt, 2012 — cited in Chapter 3: Strategies
351. Goertzel et. al., 2023 — cited in Chapter 1: Capabilities
352. Goertzel, 2010 — cited in Chapter 3: Strategies
353. Goldie et al., 2024 — cited in Chapter 2: Risks
354. Goldowsky-Dill et al., 2025 — cited in Chapter 7: Goal Misgeneralization
355. Goldwasser et al., 2024 — cited in Chapter 2: Risks
356. Golovneva et al., 2024 — cited in Chapter 5: Evaluations
357. Goodfellow et al., 2014 — cited in Chapter 1: Capabilities
358. Goodfellow et al., 2014 — cited in Chapter 2: Risks
359. Goodhart, 1975 — cited in Chapter 2: Risks
360. Google DeepMind, 2019 — cited in Chapter 1: Capabilities
361. Google DeepMind, 2020 — cited in Chapter 1: Capabilities, Chapter 6: Specification Gaming
362. Google DeepMind, 2022 — cited in Chapter 1: Capabilities
363. Google DeepMind, 2024 — cited in Chapter 1: Capabilities
364. Google DeepMind, 2024 — cited in Chapter 1: Capabilities
365. Google DeepMind, 2024 — cited in Chapter 1: Capabilities
366. Google DeepMind, 2025 — cited in Chapter 1: Capabilities
367. Google DeepMind, 2025 — cited in Chapter 1: Capabilities
368. Google DeepMind, 2025 — cited in Chapter 1: Capabilities
369. Google DeepMind, 2025 — cited in Chapter 1: Capabilities
370. Google DeepMind, 2025 — cited in Chapter 1: Capabilities
371. Google DeepMind, 2025 — cited in Chapter 1: Capabilities
372. Google, 2023 — cited in Chapter 1: Capabilities
373. Google, 2025 — cited in Chapter 1: Capabilities
374. Google, 2025 — cited in Chapter 1: Capabilities
375. Grace et al., 2017 — cited in Chapter 4: Governance
376. Grace et al., 2024 — cited in Chapter 4: Governance
377. Greenblatt & Roger, 2024 — cited in Chapter 5: Evaluations
378. Greenblatt & Shlegeris, 2024 — cited in Chapter 3: Strategies
379. Greenblatt & Shlegeris, 2024 — cited in Chapter 5: Evaluations
380. Greenblatt & Shlegreis, 2024 — cited in Chapter 5: Evaluations
381. Greenblatt et al., 2023 — cited in Chapter 3: Strategies, Chapter 5: Evaluations
382. Greenblatt et al., 2023 — cited in Chapter 5: Evaluations
383. Greenblatt et al., 2024 — cited in Chapter 2: Risks, Chapter 3: Strategies, Chapter 5: Evaluations, Chapter 7: Goal Misgeneralization
384. Greenblatt et al., 2024 — cited in Chapter 3: Strategies
385. Greenblatt et al., 2024 — cited in Chapter 5: Evaluations
386. Greenblatt et al., 2024 — cited in Chapter 5: Evaluations
387. Greenblatt, 2025 — cited in Chapter 3: Strategies
388. Greenblatt, 2025 — cited in Chapter 3: Strategies
389. Greenblatt, 2025 — cited in Chapter 5: Evaluations
390. Greenblatt, 2025 — cited in Chapter 5: Evaluations
391. Greenwalt, 2023 — cited in Chapter 2: Risks
392. Griffin et al., 2024 — cited in Chapter 5: Evaluations
393. Gruetzemacher et al., 2021 — cited in Chapter 4: Governance
394. Gruetzemacher et al., 2024 — cited in Chapter 2: Risks, Chapter 4: Governance
395. Grunewald, 2023 — cited in Chapter 4: Governance
396. Gurney & Tegmark, 2024 — cited in Chapter 1: Capabilities
397. Gwern, 2016 — cited in Chapter 2: Risks
398. Gwern, 2020 — cited in Chapter 1: Capabilities, Chapter 5: Evaluations
399. Habli et al., 2025 — cited in Chapter 3: Strategies
400. Hadley et al., 2024 — cited in Chapter 4: Governance
401. Hai et al, 2024 — cited in Chapter 3: Strategies
402. Haldane and May, 2011 — cited in Chapter 2: Risks
403. Hammond et al., 2025 — cited in Chapter 2: Risks, Chapter 3: Strategies
404. Hanson, 2023 — cited in Chapter 3: Strategies
405. Hao et al, 2024 — cited in Chapter 3: Strategies, Chapter 7: Goal Misgeneralization
406. Harvard, 2025 — cited in Chapter 3: Strategies
407. Haugen, 2021 — cited in Chapter 2: Risks
408. Hausenloy et al., 2023 — cited in Chapter 4: Governance
409. Hausenloy et al., 2024 — cited in Chapter 4: Governance
410. Heelan, 2025 — cited in Chapter 2: Risks
411. Heim & Koessler, 2024 — cited in Chapter 4: Governance
412. Heim et al., 2024 — cited in Chapter 4: Governance
413. Heim et al., 2024 — cited in Chapter 4: Governance
414. Heim et al., 2024 — cited in Chapter 4: Governance
415. Heim, 2021 — cited in Chapter 4: Governance
416. Hendrycks & Wang, 2024 — cited in Chapter 5: Evaluations
417. Hendrycks et al., 2020 — cited in Chapter 5: Evaluations, Chapter 8: Scalable Oversight
418. Hendrycks et al., 2021 — cited in Chapter 5: Evaluations
419. Hendrycks et al., 2021 — cited in Chapter 5: Evaluations
420. Hendrycks et al., 2022 — cited in Chapter 1: Capabilities
421. Hendrycks et al., 2022 — cited in Chapter 2: Risks, Chapter 3: Strategies, Chapter 4: Governance
422. Hendrycks et al., 2023 — cited in Chapter 5: Evaluations
423. Hendrycks et al., 2024 — cited in Chapter 4: Governance
424. Hendrycks et al., 2025 — cited in Chapter 1: Capabilities
425. Hendrycks et al., 2025 — cited in Chapter 3: Strategies
426. Hendrycks et al., 2025 — cited in Chapter 3: Strategies
427. Hendrycks, 2024 — cited in Chapter 2: Risks
428. Hendrycks, 2024 — cited in Chapter 2: Risks
429. Hendrycks, 2024 — cited in Chapter 4: Governance, Chapter 5: Evaluations
430. Hendrycks, 2024 — cited in Chapter 6: Specification Gaming
431. Hendrycks, 2025 — cited in Chapter 2: Risks
432. Hendryks, 2024 — cited in Chapter 4: Governance
433. Hernandez, 2025 — cited in Chapter 3: Strategies
434. Herrera-Poyatos et al., 2025 — cited in Chapter 4: Governance
435. Hill, 2024 — cited in Chapter 3: Strategies
436. Hinton, 2024 — cited in Chapter 3: Strategies
437. Ho & Berg, 2025 — cited in Chapter 2: Risks
438. Ho et al., 2023 — cited in Chapter 1: Capabilities
439. Ho et al., 2023 — cited in Chapter 4: Governance
440. Ho et al., 2023 — cited in Chapter 4: Governance
441. Ho et al., 2024 — cited in Chapter 1: Capabilities
442. Ho et al., 2025 — cited in Chapter 1: Capabilities
443. Ho, 2022 — cited in Chapter 1: Capabilities
444. Ho, 2025 — cited in Chapter 5: Evaluations
445. Hobbahn et al., 2023 — cited in Chapter 1: Capabilities
446. Hobbahn, 2024 — cited in Chapter 5: Evaluations
447. Hobbhahn et al., 2024 — cited in Chapter 3: Strategies
448. Hoffmann et al., 2022 — cited in Chapter 1: Capabilities
449. Hofstadter, 2023 — cited in Chapter 2: Risks
450. Hooker, 2024 — cited in Chapter 4: Governance
451. Houzel, 2012 — cited in Chapter 1: Capabilities
452. Hsu et al., 2023 — cited in Chapter 4: Governance
453. Huang et al., 2022 — cited in Chapter 3: Strategies
454. Hubinger et al., 2019 — cited in Chapter 5: Evaluations, Chapter 6: Specification Gaming, Chapter 7: Goal Misgeneralization
455. Hubinger et al., 2019 — cited in Chapter 7: Goal Misgeneralization
456. Hubinger et al., 2023 — cited in Chapter 5: Evaluations
457. Hubinger et al., 2024 — cited in Chapter 3: Strategies, Chapter 5: Evaluations
458. Hubinger Evan, 2020 — cited in Chapter 8: Scalable Oversight
459. Hubinger, 2020 — cited in Chapter 1: Capabilities
460. Hubinger, 2022 — cited in Chapter 7: Goal Misgeneralization
461. Hubinger, 2023 — cited in Chapter 3: Strategies, Chapter 5: Evaluations
462. Hubinger, 2023 — cited in Chapter 5: Evaluations
463. Human Rights Watch, 2024 — cited in Chapter 2: Risks
464. Hung, 2025 — cited in Chapter 4: Governance
465. HYAS, 2023 — cited in Chapter 2: Risks
466. IBM, 2023 — cited in Chapter 1: Capabilities
467. IBM, 2026 — cited in Chapter 1: Capabilities
468. IBM, 2026 — cited in Chapter 1: Capabilities
469. Inan et al., 2021 — cited in Chapter 6: Specification Gaming
470. Intel, 2024 — cited in Chapter 4: Governance
471. Ionut-Cirstea, 2023 — cited in Chapter 3: Strategies
472. Irving & Askell, 2019 — cited in Chapter 4: Governance
473. Irving et al., 2017 — cited in Chapter 8: Scalable Oversight
474. Jaghouar et al., 2024 — cited in Chapter 4: Governance
475. Jagielski, 2024 — cited in Chapter 4: Governance
476. Janssen et al., 2025 — cited in Chapter 4: Governance
477. Janus, 2022 — cited in Chapter 7: Goal Misgeneralization
478. Ji et al., 2024 — cited in Chapter 6: Specification Gaming
479. Jiang et al., 2019 — cited in Chapter 2: Risks
480. Jonker et al., 2024 — cited in Chapter 3: Strategies
481. Jumper et al., 2021 — cited in Chapter 1: Capabilities
482. Juneja et al., 2023 — cited in Chapter 7: Goal Misgeneralization
483. Kadavath, 2022 — cited in Chapter 1: Capabilities
484. Kaplan et al., 2020 — cited in Chapter 1: Capabilities
485. Kapoor et al., 2024 — cited in Chapter 3: Strategies
486. Karimi, 2023 — cited in Chapter 2: Risks
487. Karnofsky, 2016 — cited in Chapter 1: Capabilities
488. Karnofsky, 2022 — cited in Chapter 2: Risks
489. Karnofsky, 2024 — cited in Chapter 4: Governance
490. Kasirzadeh, 2024 — cited in Chapter 2: Risks
491. Keefe, 2017 — cited in Chapter 2: Risks
492. Kees-Dupuis & Janus, 2023 — cited in Chapter 3: Strategies
493. Kenton et al., 2024 — cited in Chapter 8: Scalable Oversight
494. Keskar et al.; 2017 — cited in Chapter 7: Goal Misgeneralization
495. Khan et al., 2022 — cited in Chapter 3: Strategies
496. Khan et al., 2024 — cited in Chapter 8: Scalable Oversight
497. King & Meinhardt, 2024 — cited in Chapter 2: Risks
498. Kinniment et al., 2024 — cited in Chapter 3: Strategies
499. Kinniment, 2025 — cited in Chapter 7: Goal Misgeneralization
500. Kirilenko et al., 2017 — cited in Chapter 2: Risks
501. Kirillov et al., 2023 — cited in Chapter 1: Capabilities
502. Kleinberg & Raghavan, 2021 — cited in Chapter 3: Strategies
503. Koessler & Schuett, 2023 — cited in Chapter 4: Governance
504. Koessler, 2023 — cited in Chapter 4: Governance
505. Kokotajlo et al., 2025 — cited in Chapter 2: Risks, Chapter 3: Strategies, Chapter 4: Governance
506. Kokotajlo, 2021 — cited in Chapter 2: Risks
507. Kolt et al., 2024 — cited in Chapter 4: Governance
508. Korbak et al., 2023 — cited in Chapter 6: Specification Gaming
509. Korbak et al., 2025 — cited in Chapter 3: Strategies, Chapter 5: Evaluations
510. Korbak et al., 2025 — cited in Chapter 3: Strategies
511. Korbak et al., 2025 — cited in Chapter 7: Goal Misgeneralization
512. Korzekwa, 2022 — cited in Chapter 1: Capabilities
513. Kosinski 2023 — cited in Chapter 1: Capabilities
514. Kosoy, 2023 — cited in Chapter 3: Strategies
515. Kovařík et al., 2019 — cited in Chapter 8: Scalable Oversight
516. Krakovna et al., 2020 — cited in Chapter 1: Capabilities, Chapter 2: Risks
517. Krakovna et al., 2020 — cited in Chapter 6: Specification Gaming
518. Kreps & Kriner, 2023 — cited in Chapter 4: Governance
519. Krizhevsky, 2009 — cited in Chapter 5: Evaluations
520. Krueger et al., 2020 — cited in Chapter 7: Goal Misgeneralization
521. Kulveit, 2025 — cited in Chapter 3: Strategies
522. Kumar, 2022 — cited in Chapter 7: Goal Misgeneralization
523. Kwa et al., 2025 — cited in Chapter 1: Capabilities, Chapter 5: Evaluations
524. LaCroix & Luccioni, 2022 — cited in Chapter 3: Strategies
525. Ladish & Heim, 2022 — cited in Chapter 3: Strategies
526. Ladish, 2023 — cited in Chapter 3: Strategies
527. Laine et al., 2024 — cited in Chapter 2: Risks, Chapter 5: Evaluations, Chapter 7: Goal Misgeneralization
528. Lam et al., 2023 — cited in Chapter 3: Strategies
529. Lancieri et al., 2024 — cited in Chapter 4: Governance
530. Langosco et al., 2022 — cited in Chapter 7: Goal Misgeneralization
531. Lanham et al., 2023 — cited in Chapter 3: Strategies, Chapter 8: Scalable Oversight
532. Lanham, 2022 — cited in Chapter 3: Strategies, Chapter 8: Scalable Oversight
533. Lazar, 2024 — cited in Chapter 4: Governance
534. Leahy & Alfour, 2023 — cited in Chapter 3: Strategies
535. Leahy et al., 2024 — cited in Chapter 2: Risks
536. Leake, 2023 — cited in Chapter 3: Strategies
537. LeCun, 1998 — cited in Chapter 5: Evaluations
538. LeCun, 2022 — cited in Chapter 1: Capabilities
539. LeCun, 2025 — cited in Chapter 1: Capabilities
540. Legg & Hutter, 2007 — cited in Chapter 1: Capabilities
541. Lehalleur et al., 2025 — cited in Chapter 7: Goal Misgeneralization
542. Leike, 2022 — cited in Chapter 3: Strategies, Chapter 8: Scalable Oversight
543. Leike, 2023 — cited in Chapter 3: Strategies
544. Leike, 2023 — cited in Chapter 8: Scalable Oversight
545. Lermen et al., 2024 — cited in Chapter 3: Strategies
546. Li et al., 2017 — cited in Chapter 7: Goal Misgeneralization
547. Li et al., 2022 — cited in Chapter 2: Risks
548. Li et al., 2023 — cited in Chapter 6: Specification Gaming
549. Li et al., 2024 — cited in Chapter 2: Risks, Chapter 5: Evaluations
550. Li, 2025 — cited in Chapter 4: Governance
551. Liang et al., 2022 — cited in Chapter 3: Strategies
552. Lightman et al., 2023 — cited in Chapter 1: Capabilities, Chapter 3: Strategies, Chapter 8: Scalable Oversight
553. Lin et al., 2022 — cited in Chapter 1: Capabilities, Chapter 5: Evaluations
554. Liu et al., 2023 — cited in Chapter 2: Risks
555. Liu et al., 2024 — cited in Chapter 2: Risks
556. Liu, 2023 — cited in Chapter 5: Evaluations
557. Liu, 2024 — cited in Chapter 3: Strategies
558. Lizka, 2023 — cited in Chapter 3: Strategies
559. Longpre et al., 2023 — cited in Chapter 2: Risks, Chapter 4: Governance
560. Longpre et al., 2024 — cited in Chapter 4: Governance
561. Lu et al., 2024 — cited in Chapter 7: Goal Misgeneralization
562. Luo et al., 2019 — cited in Chapter 7: Goal Misgeneralization
563. Maas & Villalobos, 2024 — cited in Chapter 4: Governance
564. Maas, 2019 — cited in Chapter 4: Governance
565. MacAskill at al, 2025 — cited in Chapter 3: Strategies
566. MacDermott et al., 2024 — cited in Chapter 7: Goal Misgeneralization
567. MacDiarmid et al., 2024 — cited in Chapter 7: Goal Misgeneralization
568. Manheim & Garrabrant, 2018 — cited in Chapter 2: Risks
569. Mannheim, 2023 — cited in Chapter 3: Strategies
570. Marchal et al., 2024 — cited in Chapter 4: Governance
571. Marcucci et al., 2023 — cited in Chapter 4: Governance
572. Marcus, 2025 — cited in Chapter 1: Capabilities
573. Marks et al., 2023 — cited in Chapter 5: Evaluations
574. Martin & Eth, 2021 — cited in Chapter 1: Capabilities
575. Masi, 2024 — cited in Chapter 4: Governance
576. Maynez et al., 2020 — cited in Chapter 1: Capabilities
577. Mazeika et al., 2024 — cited in Chapter 5: Evaluations
578. McAleese et al., 2024 — cited in Chapter 8: Scalable Oversight
579. McCoy et al., 2019 — cited in Chapter 7: Goal Misgeneralization
580. McGrath et al., 2021 — cited in Chapter 1: Capabilities
581. McGregor, 2020 — cited in Chapter 2: Risks
582. McKernon et al., 2024 — cited in Chapter 4: Governance
583. McKernon et al., 2024 — cited in Chapter 4: Governance
584. Meinke et al., 2024 — cited in Chapter 3: Strategies, Chapter 5: Evaluations, Chapter 7: Goal Misgeneralization
585. Meng et al., 2023 — cited in Chapter 1: Capabilities
586. Merchant et al., 2023 — cited in Chapter 3: Strategies
587. META, 2022 — cited in Chapter 2: Risks, Chapter 5: Evaluations
588. Meta, 2022 — cited in Chapter 5: Evaluations
589. Metaculus, 2025 — cited in Chapter 2: Risks
590. Metaculus, 2025 — cited in Chapter 2: Risks
591. METR, 2023 — cited in Chapter 2: Risks
592. METR, 2023 — cited in Chapter 2: Risks
593. METR, 2023 — cited in Chapter 3: Strategies, Chapter 5: Evaluations
594. METR, 2023 — cited in Chapter 5: Evaluations
595. METR, 2024 — cited in Chapter 2: Risks
596. METR, 2024 — cited in Chapter 2: Risks, Chapter 5: Evaluations
597. METR, 2024 — cited in Chapter 5: Evaluations
598. METR, 2025 — cited in Chapter 2: Risks
599. METR, 2025 — cited in Chapter 3: Strategies
600. METR, 2025 — cited in Chapter 4: Governance
601. Michael et al., 2023 — cited in Chapter 8: Scalable Oversight
602. Mikulik, 2019 — cited in Chapter 7: Goal Misgeneralization
603. Miller, 2022 — cited in Chapter 3: Strategies
604. Millidge, 2023 — cited in Chapter 5: Evaluations
605. Millidge, 2025 — cited in Chapter 3: Strategies
606. Mingard et al., 2019 — cited in Chapter 7: Goal Misgeneralization
607. Miotti et al, 2024 — cited in Chapter 3: Strategies
608. Mirhoseini et al., 2020 — cited in Chapter 2: Risks
609. Mishra, 2024 — cited in Chapter 4: Governance
610. Mnih et al., 2013 — cited in Chapter 1: Capabilities
611. Morris et al., 2024 — cited in Chapter 1: Capabilities
612. Moskvichev et al. 2023 — cited in Chapter 5: Evaluations
613. Mouton et al., 2023 — cited in Chapter 2: Risks, Chapter 4: Governance
614. Mowshowitz, 2023 — cited in Chapter 3: Strategies
615. Mowshowitz, 2023 — cited in Chapter 3: Strategies
616. Mowshowitz, 2025 — cited in Chapter 3: Strategies
617. Mowshowitz, 2025 — cited in Chapter 3: Strategies
618. Muehlhauser, 2012 — cited in Chapter 1: Capabilities
619. Mukobi, 2024 — cited in Chapter 4: Governance
620. Mukobi, 2024 — cited in Chapter 5: Evaluations
621. Munk Debate, 2024 — cited in Chapter 2: Risks
622. Murphy, 2013 — cited in Chapter 2: Risks
623. Nakano et al, 2021 — cited in Chapter 5: Evaluations
624. Nanda et al., 2023 — cited in Chapter 1: Capabilities
625. Nanda, 2025 — cited in Chapter 5: Evaluations
626. Narayan & Kapoor, 2024 — cited in Chapter 4: Governance
627. Narayanan & Kapoor, 2024 — cited in Chapter 4: Governance
628. Nardo, 2023 — cited in Chapter 6: Specification Gaming, Chapter 7: Goal Misgeneralization
629. NASA, 2004 — cited in Chapter 2: Risks
630. Nasr et al., 2023 — cited in Chapter 2: Risks
631. Neuralt, 2024 — cited in Chapter 3: Strategies
632. Nevo et al., 2024 — cited in Chapter 3: Strategies, Chapter 4: Governance
633. Nevo et al., 2024 — cited in Chapter 4: Governance
634. Newman, 2024 — cited in Chapter 2: Risks
635. Ng, 2025 — cited in Chapter 3: Strategies
636. Ngo, 2023 — cited in Chapter 1: Capabilities
637. Ngo, 2024 — cited in Chapter 3: Strategies
638. Nguyen et al., 2023 — cited in Chapter 2: Risks
639. NIST, 2023 — cited in Chapter 3: Strategies
640. Novikov et al., 2025 — cited in Chapter 2: Risks
641. NSCEB, 2024 — cited in Chapter 2: Risks
642. NVIDIA, 2025 — cited in Chapter 4: Governance
643. O'Brien et al., 2023 — cited in Chapter 4: Governance
644. OECD, 2025 — cited in Chapter 1: Capabilities
645. OECD, 2025 — cited in Chapter 4: Governance
646. Office of the Attorney General, 2023 — cited in Chapter 2: Risks
647. Olah, 2023 — cited in Chapter 5: Evaluations
648. Olds & Milner, 1954 — cited in Chapter 2: Risks
649. Olds, 1956 — cited in Chapter 2: Risks
650. Omohundro, 2008 — cited in Chapter 6: Specification Gaming
651. OpenAI, 2017 — cited in Chapter 2: Risks
652. OpenAI, 2017 — cited in Chapter 6: Specification Gaming
653. OpenAI, 2018 — cited in Chapter 6: Specification Gaming
654. OpenAI, 2019 — cited in Chapter 1: Capabilities, Chapter 6: Specification Gaming
655. OpenAI, 2022 — cited in Chapter 3: Strategies
656. OpenAI, 2022 — cited in Chapter 3: Strategies
657. OpenAI, 2022 — cited in Chapter 6: Specification Gaming
658. OpenAI, 2022 — cited in Chapter 6: Specification Gaming
659. OpenAI, 2023 — cited in Chapter 1: Capabilities
660. OpenAI, 2023 — cited in Chapter 1: Capabilities
661. OpenAI, 2023 — cited in Chapter 2: Risks, Chapter 3: Strategies, Chapter 5: Evaluations, Chapter 6: Specification Gaming
662. OpenAI, 2023 — cited in Chapter 3: Strategies, Chapter 5: Evaluations
663. OpenAI, 2023 — cited in Chapter 3: Strategies
664. OpenAI, 2023 — cited in Chapter 5: Evaluations
665. OpenAI, 2023 — cited in Chapter 6: Specification Gaming
666. OpenAI, 2024 — cited in Chapter 1: Capabilities
667. OpenAI, 2024 — cited in Chapter 3: Strategies
668. OpenAI, 2024 — cited in Chapter 5: Evaluations
669. OpenAI, 2024 — cited in Chapter 5: Evaluations
670. OpenAI, 2024 — cited in Chapter 5: Evaluations
671. OpenAI, 2025 — cited in Chapter 1: Capabilities
672. OpenAI, 2025 — cited in Chapter 1: Capabilities
673. OpenAI, 2025 — cited in Chapter 1: Capabilities
674. OpenAI, 2025 — cited in Chapter 3: Strategies
675. OpenAI, 2025 — cited in Chapter 4: Governance
676. OpenAI, 2025 — cited in Chapter 5: Evaluations
677. Ord, 2020 — cited in Chapter 2: Risks
678. Ought, 2018 — cited in Chapter 8: Scalable Oversight
679. Ought, 2022 — cited in Chapter 8: Scalable Oversight
680. Our World in Data, 2024 — cited in Chapter 1: Capabilities
681. Our World in Data, 2026 — cited in Chapter 1: Capabilities
682. Owen, 2025 — cited in Chapter 1: Capabilities
683. OWID, 2025 — cited in Chapter 2: Risks
684. OWID, 2025 — cited in Chapter 2: Risks
685. Oxford Reference, 2016 — cited in Chapter 5: Evaluations
686. Oxford Union Debate, 2024 — cited in Chapter 2: Risks
687. O’Brien et al., 2024 — cited in Chapter 4: Governance
688. O’Keefe, 2020 — cited in Chapter 4: Governance
689. Pacchiardi et al., 2023 — cited in Chapter 5: Evaluations
690. Pacchiardi et al., 2025 — cited in Chapter 1: Capabilities
691. Pace, 2020 — cited in Chapter 3: Strategies
692. Pan et al., 2020 — cited in Chapter 6: Specification Gaming
693. Pan et al., 2023 — cited in Chapter 5: Evaluations
694. Panel of Experts on Libya, 2021 — cited in Chapter 2: Risks
695. Pang et al., 2021 — cited in Chapter 8: Scalable Oversight
696. Pannu et al., 2024 — cited in Chapter 2: Risks
697. Papagiannidis, 2025 — cited in Chapter 4: Governance
698. Papyshev et al., 2023 — cited in Chapter 4: Governance
699. Park et al., 2023 — cited in Chapter 2: Risks, Chapter 3: Strategies
700. Parrish et al., 2022 — cited in Chapter 8: Scalable Oversight
701. Parrish et al., 2022 — cited in Chapter 8: Scalable Oversight
702. Patel, 2023 — cited in Chapter 4: Governance
703. PauseAI, 2023 — cited in Chapter 3: Strategies
704. PauseAI, 2025 — cited in Chapter 2: Risks, Chapter 3: Strategies
705. Pearson et al., 2023 — cited in Chapter 1: Capabilities
706. Peng & Shang, 2024 — cited in Chapter 7: Goal Misgeneralization
707. Peng et al.,2024 — cited in Chapter 5: Evaluations
708. Peppin et al., 2024 — cited in Chapter 2: Risks
709. Perez et al., 2022 — cited in Chapter 2: Risks
710. Perez et al., 2022 — cited in Chapter 5: Evaluations, Chapter 6: Specification Gaming
711. Perez et al., 2022 — cited in Chapter 5: Evaluations
712. Perlman, 2024 — cited in Chapter 5: Evaluations
713. Petillo et al., 2025 — cited in Chapter 7: Goal Misgeneralization
714. Petrie et al., 2024 — cited in Chapter 4: Governance
715. Petropoulos et al., 2025 — cited in Chapter 4: Governance
716. Phan et al., 2025 — cited in Chapter 5: Evaluations
717. Phuong et al., 2024 — cited in Chapter 4: Governance, Chapter 5: Evaluations
718. Pilz & Heim, 2023 — cited in Chapter 4: Governance
719. Pilz et al., 2025 — cited in Chapter 4: Governance
720. Piper, 2023 — cited in Chapter 5: Evaluations
721. Pistillo, 2025 — cited in Chapter 4: Governance
722. Pistillo, 2025 — cited in Chapter 4: Governance
723. Pope, 2023 — cited in Chapter 3: Strategies
724. Potlogea and Ho, 2025 — cited in Chapter 2: Risks
725. Power et al., 2022 — cited in Chapter 3: Strategies
726. Prime Intellect, 2024 — cited in Chapter 3: Strategies
727. Prime Intellect, 2025 — cited in Chapter 3: Strategies, Chapter 4: Governance
728. Purtova et al., 2022 — cited in Chapter 4: Governance
729. Qin et al., 2023 — cited in Chapter 1: Capabilities
730. Qin et al., 2024 — cited in Chapter 2: Risks
731. Radhakrishnan et al., 2023 — cited in Chapter 3: Strategies
732. Rae et al., 2022 — cited in Chapter 1: Capabilities
733. Rafailov et al., 2023 — cited in Chapter 6: Specification Gaming
734. Rahaman et al., 2019 — cited in Chapter 7: Goal Misgeneralization
735. Raji et al., 2020 — cited in Chapter 4: Governance
736. Rapaport, 2020 — cited in Chapter 1: Capabilities
737. Rational Animations, 2025 — cited in Chapter 8: Scalable Oversight
738. Rational Animations, 2025 — cited in Chapter 8: Scalable Oversight
739. Reed et al., 2022 — cited in Chapter 1: Capabilities
740. Rein et al., 2023 — cited in Chapter 5: Evaluations
741. Ren & Sutherland, 2024 — cited in Chapter 7: Goal Misgeneralization
742. Ren et al., 2024 — cited in Chapter 2: Risks, Chapter 3: Strategies, Chapter 4: Governance, Chapter 5: Evaluations
743. Reuel & Bucknall, 2024 — cited in Chapter 4: Governance
744. Reuel & Bucknall, 2024 — cited in Chapter 4: Governance
745. Reuel et al., 2024 — cited in Chapter 4: Governance
746. Riché et al., 2024 — cited in Chapter 5: Evaluations
747. Rivera et al., 2024 — cited in Chapter 2: Risks
748. Roberts et al., 2024 — cited in Chapter 4: Governance
749. Rodriguez, 2020 — cited in Chapter 2: Risks
750. Roger & Greenblatt, 2023 — cited in Chapter 3: Strategies
751. Roger & Shlegeris, 2024 — cited in Chapter 5: Evaluations
752. Roger et al., 2023 — cited in Chapter 5: Evaluations
753. Roger, 2023 — cited in Chapter 3: Strategies
754. Roger, 2023 — cited in Chapter 3: Strategies
755. Rogers Commission, 1986 — cited in Chapter 2: Risks
756. Rolls et al., 1980 — cited in Chapter 2: Risks
757. Rudner et al., 2021 — cited in Chapter 3: Strategies
758. Rumbelow & Watkins, 2023 — cited in Chapter 3: Strategies
759. Russel & Norvig, 1994 — cited in Chapter 2: Risks
760. SaferAI, 2025 — cited in Chapter 3: Strategies
761. SaferAI, 2025 — cited in Chapter 3: Strategies
762. SakanaAI, 2024 — cited in Chapter 1: Capabilities, Chapter 2: Risks
763. SakanaAI, 2025 — cited in Chapter 3: Strategies
764. Sandbrink, 2023 — cited in Chapter 2: Risks
765. Sandoval-Segura et al., 2022 — cited in Chapter 2: Risks
766. Santurkar et al., 2023 — cited in Chapter 6: Specification Gaming
767. Sastry et al., 2024 — cited in Chapter 4: Governance
768. Saunders et al., 2022 — cited in Chapter 8: Scalable Oversight
769. Scherlis et. al. 2024 — cited in Chapter 8: Scalable Oversight
770. Scherlis, 2023 — cited in Chapter 7: Goal Misgeneralization
771. Scheurer et al., 2022 — cited in Chapter 6: Specification Gaming
772. Scheurer et al., 2024 — cited in Chapter 5: Evaluations
773. Schick et al., 2023 — cited in Chapter 1: Capabilities, Chapter 5: Evaluations
774. Schneider & McGrew, 2018 — cited in Chapter 1: Capabilities
775. Schrimpf et al., 2021 — cited in Chapter 1: Capabilities
776. Schuett et al., 2024 — cited in Chapter 4: Governance
777. Schuett, 2022 — cited in Chapter 3: Strategies, Chapter 4: Governance
778. Schuett, 2023 — cited in Chapter 4: Governance
779. Schwarzschild et al., 2021 — cited in Chapter 2: Risks
780. Schäfer et al., 2022 — cited in Chapter 4: Governance
781. Sclar et al., 2023 — cited in Chapter 5: Evaluations
782. Searle, 1980 — cited in Chapter 1: Capabilities
783. Seger et al., 2023 — cited in Chapter 3: Strategies, Chapter 4: Governance
784. Seger et al., 2023 — cited in Chapter 4: Governance
785. Segerie & Gédéon, 2024 — cited in Chapter 2: Risks
786. Segerie & Kolly, 2023 — cited in Chapter 3: Strategies
787. Segerie, 2024 — cited in Chapter 3: Strategies
788. Segerie, 2025 — cited in Chapter 3: Strategies
789. Sener & Koltun, 2019 — cited in Chapter 7: Goal Misgeneralization
790. Sevilla et. al., 2024 — cited in Chapter 1: Capabilities, Chapter 4: Governance
791. Sevilla, 2025 — cited in Chapter 3: Strategies
792. Sevilla, 2025 — cited in Chapter 3: Strategies
793. Sevilla, AXRP Podcast, 2024 — cited in Chapter 1: Capabilities
794. Shah et al., 2020 — cited in Chapter 7: Goal Misgeneralization
795. Shah et al., 2025 — cited in Chapter 2: Risks
796. Shao et al., 2024 — cited in Chapter 5: Evaluations
797. Shapira, 2025 — cited in Chapter 3: Strategies
798. Sharkey, 2023 — cited in Chapter 7: Goal Misgeneralization
799. Sharkey, 2024 — cited in Chapter 5: Evaluations
800. Shavit, 2023 — cited in Chapter 4: Governance
801. Shayegani et al., 2023 — cited in Chapter 2: Risks
802. Shen et al., 2023 — cited in Chapter 5: Evaluations
803. Shevlane & Dafoe, 2020 — cited in Chapter 3: Strategies
804. Shevlane et al., 2023 — cited in Chapter 3: Strategies, Chapter 5: Evaluations, Chapter 6: Specification Gaming
805. Shevlane et al., 2023 — cited in Chapter 4: Governance
806. Shevlane, 2024 — cited in Chapter 4: Governance
807. Shi et al., 2022 — cited in Chapter 5: Evaluations
808. Shimi, 2020 — cited in Chapter 7: Goal Misgeneralization
809. Shinn et al., 2023 — cited in Chapter 1: Capabilities
810. Shlegeris, 2022 — cited in Chapter 3: Strategies
811. Shlegeris, 2024 — cited in Chapter 3: Strategies
812. Shokri et al., 2017 — cited in Chapter 2: Risks
813. Siegmann & Anderljung 2022 — cited in Chapter 4: Governance
814. Simmons-Edler et al., 2024 — cited in Chapter 2: Risks
815. Skare et al., 2024 — cited in Chapter 4: Governance
816. Slattery et al., 2024 — cited in Chapter 2: Risks, Chapter 3: Strategies
817. Snyder et al., 2020 — cited in Chapter 3: Strategies
818. Soares, 2022 — cited in Chapter 3: Strategies, Chapter 7: Goal Misgeneralization
819. Soares, 2023 — cited in Chapter 2: Risks
820. Soares, 2023 — cited in Chapter 2: Risks
821. Soice et al., 2023 — cited in Chapter 2: Risks
822. Solaiman et al., 2019 — cited in Chapter 3: Strategies
823. Solaiman et al., 2023 — cited in Chapter 3: Strategies, Chapter 4: Governance
824. Solaiman et al., 2024 — cited in Chapter 4: Governance
825. Srivastava et al., 2022 — cited in Chapter 5: Evaluations
826. Stacey, 2025 — cited in Chapter 4: Governance
827. Stanford HAI, 2024 — cited in Chapter 4: Governance
828. Stanford HAI, 2025 — cited in Chapter 2: Risks, Chapter 4: Governance
829. Stanford HAI, 2025 — cited in Chapter 4: Governance
830. Stanford, 2024 — cited in Chapter 4: Governance
831. State of AI Report, 2023 — cited in Chapter 4: Governance
832. Steiner, 2022 — cited in Chapter 3: Strategies
833. Steinhardt, 2022 — cited in Chapter 2: Risks
834. Steinhardt, 2022 — cited in Chapter 2: Risks
835. Steinhardt, 2023 — cited in Chapter 2: Risks
836. Steinhardt, 2023 — cited in Chapter 3: Strategies
837. Steinhardt, 2024 — cited in Chapter 7: Goal Misgeneralization
838. Stewart & Plotkin, 2012 — cited in Chapter 4: Governance
839. Stix et al., 2025 — cited in Chapter 1: Capabilities, Chapter 4: Governance
840. Sutton, 2019 — cited in Chapter 1: Capabilities
841. Sutton, 2023 — cited in Chapter 3: Strategies
842. Sutton, 2023 — cited in Chapter 3: Strategies
843. SWE bench, 2025 — cited in Chapter 1: Capabilities
844. Swenson & Chan, 2024 — cited in Chapter 4: Governance
845. Takemoto, 2024 — cited in Chapter 6: Specification Gaming
846. Tallberg et al., 2023 — cited in Chapter 4: Governance
847. Tamirisa et al., 2024 — cited in Chapter 3: Strategies, Chapter 4: Governance
848. Tanzer et al., 2024 — cited in Chapter 5: Evaluations
849. Tegmark & Omohundro, 2023 — cited in Chapter 3: Strategies
850. Tegmark, 2017 — cited in Chapter 3: Strategies
851. Tegmark, 2023 — cited in Chapter 2: Risks
852. Tegmark, 2024 — cited in Chapter 2: Risks
853. The Bulletin, 2024 — cited in Chapter 2: Risks
854. Tian et al., 2023 — cited in Chapter 1: Capabilities
855. Tihanyi et al., 2024 — cited in Chapter 5: Evaluations
856. TIME, 2024 — cited in Chapter 4: Governance
857. Tong, 2023 — cited in Chapter 2: Risks
858. Trajano & Ang, 2023 — cited in Chapter 4: Governance
859. Trusilo, 2024 — cited in Chapter 4: Governance
860. Truth Initiative, 2017 — cited in Chapter 2: Risks
861. Tsoy & Konstantinov, 2024 — cited in Chapter 7: Goal Misgeneralization
862. Turing, 1950 — cited in Chapter 1: Capabilities
863. Turner & Tadepalli, 2022 — cited in Chapter 2: Risks
864. Turner et al., 2019 — cited in Chapter 2: Risks, Chapter 3: Strategies
865. Turner, 2022 — cited in Chapter 3: Strategies
866. Turner, 2022 — cited in Chapter 7: Goal Misgeneralization
867. Turner, 2023 — cited in Chapter 3: Strategies
868. Turpin et al., 2023 — cited in Chapter 3: Strategies, Chapter 4: Governance, Chapter 8: Scalable Oversight
869. U.K. government, 2023 — cited in Chapter 4: Governance
870. Uesato et. al, 2022 — cited in Chapter 8: Scalable Oversight
871. UK AISI, 2024 — cited in Chapter 5: Evaluations
872. Urbina et al., 2022 — cited in Chapter 2: Risks
873. US & UK AISI 2024 — cited in Chapter 5: Evaluations
874. US & UK AISI, 2024 — cited in Chapter 5: Evaluations
875. Vaintrob, 2023 — cited in Chapter 3: Strategies
876. Valle-Pérez et al., 2019 — cited in Chapter 7: Goal Misgeneralization
877. van der Weij et al., 2024 — cited in Chapter 5: Evaluations
878. van der Weij, 2024 — cited in Chapter 5: Evaluations
879. Variengien & Martinet, 2024 — cited in Chapter 4: Governance
880. Villalobos et al., 2024 — cited in Chapter 1: Capabilities
881. Villalobos et al., 2024 — cited in Chapter 1: Capabilities
882. vTaiwan, 2023 — cited in Chapter 3: Strategies
883. Wache, 2023 — cited in Chapter 3: Strategies
884. Wan et al., 2024 — cited in Chapter 5: Evaluations
885. Wang and Gleave et al., 2022 — cited in Chapter 1: Capabilities
886. Wang et al., 2018 — cited in Chapter 5: Evaluations
887. Wang et al., 2019 — cited in Chapter 5: Evaluations
888. Wang et al., 2023 — cited in Chapter 1: Capabilities, Chapter 5: Evaluations
889. Wang et al., 2023 — cited in Chapter 5: Evaluations
890. Wang et al., 2024 — cited in Chapter 1: Capabilities
891. Wang et al., 2024 — cited in Chapter 7: Goal Misgeneralization
892. Wang et al., 2025 — cited in Chapter 5: Evaluations
893. Wasil et al., 2024 — cited in Chapter 3: Strategies
894. Wasil et al., 2024 — cited in Chapter 4: Governance
895. Wasil, 2022 — cited in Chapter 3: Strategies
896. Wei et al., 2022 — cited in Chapter 1: Capabilities, Chapter 4: Governance
897. Wei et al., 2022 — cited in Chapter 5: Evaluations, Chapter 8: Scalable Oversight
898. Weidinger et al., 2021 — cited in Chapter 2: Risks
899. Weidinger et al., 2023 — cited in Chapter 3: Strategies, Chapter 5: Evaluations
900. Wentworth, 2022 — cited in Chapter 3: Strategies
901. Wentworth, 2022 — cited in Chapter 8: Scalable Oversight
902. Wentworth, 2025 — cited in Chapter 3: Strategies
903. Wentworth, 2025 — cited in Chapter 3: Strategies
904. Wentworth, 2025 — cited in Chapter 3: Strategies
905. Whitehouse, 2025 — cited in Chapter 2: Risks
906. Whitehouse, 2025 — cited in Chapter 2: Risks
907. Wijk, 2023 — cited in Chapter 3: Strategies
908. Wikipedia, 2022 — cited in Chapter 1: Capabilities
909. Williams et al., 2025 — cited in Chapter 2: Risks
910. Willis et al., 2024 — cited in Chapter 2: Risks
911. Wiseman & McClements, 2025 — cited in Chapter 8: Scalable Oversight
912. Wojton et al., 2020 — cited in Chapter 4: Governance
913. Wolf et al., 2024 — cited in Chapter 6: Specification Gaming
914. Wongkamjan et al., 2024 — cited in Chapter 5: Evaluations
915. Wooldridge, 2021 — cited in Chapter 1: Capabilities
916. Wooldridge, 2024 — cited in Chapter 2: Risks
917. WorldCoin, 2024 — cited in Chapter 2: Risks
918. Wu et al., 2021 — cited in Chapter 8: Scalable Oversight
919. Xiang, 2023 — cited in Chapter 2: Risks
920. Xiao et al., 2021 — cited in Chapter 6: Specification Gaming
921. Xu et al., 2022 — cited in Chapter 2: Risks
922. Xu et al., 2024 — cited in Chapter 1: Capabilities
923. Xu et al., 2025 — cited in Chapter 5: Evaluations
924. Yampolskiy, 2024 — cited in Chapter 2: Risks, Chapter 5: Evaluations
925. Yampolsky; 2024 — cited in Chapter 2: Risks, Chapter 5: Evaluations
926. Yang et al., 2022 — cited in Chapter 6: Specification Gaming, Chapter 8: Scalable Oversight
927. Yang et al., 2023 — cited in Chapter 5: Evaluations
928. Yao et al., 2023 — cited in Chapter 5: Evaluations
929. Yap, 2024 — cited in Chapter 1: Capabilities
930. Ye et al., 2021 — cited in Chapter 1: Capabilities, Chapter 6: Specification Gaming
931. You et al., 2025 — cited in Chapter 1: Capabilities
932. Yu et al., 2022 — cited in Chapter 1: Capabilities
933. Yudkowsky, 2002 — cited in Chapter 3: Strategies
934. Yudkowsky, 2002 — cited in Chapter 3: Strategies
935. Yudkowsky, 2004 — cited in Chapter 3: Strategies
936. Yudkowsky, 2007 — cited in Chapter 3: Strategies
937. Yudkowsky, 2008 — cited in Chapter 3: Strategies
938. Yudkowsky, 2013 — cited in Chapter 1: Capabilities
939. Yudkowsky, 2015 — cited in Chapter 2: Risks
940. Yudkowsky, 2022 — cited in Chapter 3: Strategies
941. Yudkowsky, 2022 — cited in Chapter 3: Strategies
942. Yudkowsky, 2022 — cited in Chapter 3: Strategies
943. Yudkowsky, 2023 — cited in Chapter 3: Strategies
944. Zaidan & Ibrahim, 2024 — cited in Chapter 4: Governance
945. Zellers et al., 2018 — cited in Chapter 5: Evaluations
946. Zellers et al., 2019 — cited in Chapter 5: Evaluations
947. Zhang et al., 2017 — cited in Chapter 2: Risks
948. Zhang et al., 2021 — cited in Chapter 4: Governance
949. Zhang et al., 2021 — cited in Chapter 4: Governance
950. Zhang et al., 2023 — cited in Chapter 8: Scalable Oversight
951. Zhang et al., 2024 — cited in Chapter 3: Strategies, Chapter 5: Evaluations
952. Zhang et al., 2024 — cited in Chapter 5: Evaluations
953. Zhang et al., 2025 — cited in Chapter 4: Governance
954. Zhao et al., 2024 — cited in Chapter 2: Risks
955. Zhou et al., 2023 — cited in Chapter 5: Evaluations
956. Zhou et al., 2024 — cited in Chapter 4: Governance
957. Zhu et al., 2021 — cited in Chapter 2: Risks
958. Zhu et al., 2024 — cited in Chapter 5: Evaluations
959. Ziegler et al., 2020 — cited in Chapter 2: Risks
960. Zou et al., 2023 — cited in Chapter 5: Evaluations
961. Zuho et al., 2024 — cited in Chapter 5: Evaluations
962. Zvi, 2025 — cited in Chapter 3: Strategies
963. Özcan, 2024 — cited in Chapter 4: Governance