Samotsvety - Footnote 16

NoteThe percentages for transformative AI in 2030, 2050, and 2100 are different in the source. The source only provides a percentage for 2100 (81%), while the claim provides percentages for 2030 (28%), 2050 (60%), and 2100 (89%). The source states 25% chance of misaligned AI takeover by 2100, while the claim states 25% by 2100. The source includes the caveat "barring pre- APS-AI catastrophe" which is not included in the claim.

NoteWRONG NUMBERS: The claim includes timelines for transformative AI (28% by 2030, 60% by 2050, 89% by 2100), but the source only mentions 81% chance of Transformative AI (TAI) by 2100. WRONG NUMBERS: The claim states 89% by 2100, but the source states 81% by 2100. UNSUPPORTED: The claim mentions that the group released influential AI risk forecasts throughout 2023-2024, but the source is dated September 9, 2022, and does not mention any activities in 2023-2024. UNSUPPORTED: The claim mentions that their work has been incorporated into literature reviews by Epoch AI and cited in discussions about AI safety policy, but the source does not mention this.