BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//AI Forecast Calendar//Forecast Milestones//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Peter Wildeford’s AGI Timelines Model
X-WR-CALDESC:Probability-threshold dates from a frozen run of Peter Wildefo
 rd’s METR-based Monte Carlo model of AGI timelines.
BEGIN:VEVENT
UID:wildeford-agi-timelines.agi-p10@ai-forecast-calendar
DTSTAMP:20260209T000000Z
DTSTART;VALUE=DATE:20281006
DTEND;VALUE=DATE:20281007
SUMMARY:Peter Wildeford’s AGI Timelines Model: Model distribution reaches
  10% cumulative AGI probability
DESCRIPTION:Forecast: Ten percent of a frozen 100\,000-run snapshot of Wild
 eford’s default METR-based model reaches its AGI criterion by this date.
 \n\nSource timing: 10th percentile: October 6\, 2028\n\nCalendar anchor: 2
 028-10-06 (This is the exact extracted percentile date from the frozen mod
 el extraction.)\n\nSource context: AGI Timelines Model default Monte Carlo
  distribution\, upstream commit 664e1cec (2026-02-09)\n\nForecast uncertai
 nty: The model result is highly sensitive to its benchmark extrapolation\,
  AGI task threshold\, reliability\, domain\, and acceleration assumptions.
 \n\nSource: https://github.com/peterhurford/agi_timelines
URL:https://github.com/peterhurford/agi_timelines
TRANSP:TRANSPARENT
END:VEVENT
BEGIN:VEVENT
UID:wildeford-agi-timelines.agi-p25@ai-forecast-calendar
DTSTAMP:20260209T000000Z
DTSTART;VALUE=DATE:20300128
DTEND;VALUE=DATE:20300129
SUMMARY:Peter Wildeford’s AGI Timelines Model: Model distribution reaches
  25% cumulative AGI probability
DESCRIPTION:Forecast: Twenty-five percent of the model snapshot reaches its
  AGI criterion by this date.\n\nSource timing: 25th percentile: January 28
 \, 2030\n\nCalendar anchor: 2030-01-28 (This is the exact extracted percen
 tile date from the frozen model extraction.)\n\nSource context: AGI Timeli
 nes Model default Monte Carlo distribution\, upstream commit 664e1cec (202
 6-02-09)\n\nForecast uncertainty: A model percentile is not a guarantee an
 d should not be read as independent evidence.\n\nSource: https://github.co
 m/peterhurford/agi_timelines
URL:https://github.com/peterhurford/agi_timelines
TRANSP:TRANSPARENT
END:VEVENT
BEGIN:VEVENT
UID:wildeford-agi-timelines.agi-p50@ai-forecast-calendar
DTSTAMP:20260209T000000Z
DTSTART;VALUE=DATE:20320312
DTEND;VALUE=DATE:20320313
SUMMARY:Peter Wildeford’s AGI Timelines Model: Model median AGI date
DESCRIPTION:Forecast: Half of the model snapshot reaches its AGI criterion 
 by this date.\n\nSource timing: 50th percentile: March 12\, 2032\n\nCalend
 ar anchor: 2032-03-12 (This is the exact extracted median date from the fr
 ozen model extraction.)\n\nSource context: AGI Timelines Model default Mon
 te Carlo distribution\, upstream commit 664e1cec (2026-02-09)\n\nForecast 
 uncertainty: The repository is an evolving model\; this calendar preserves
  one dated snapshot rather than updating silently.\n\nSource: https://gith
 ub.com/peterhurford/agi_timelines
URL:https://github.com/peterhurford/agi_timelines
TRANSP:TRANSPARENT
END:VEVENT
BEGIN:VEVENT
UID:wildeford-agi-timelines.agi-p75@ai-forecast-calendar
DTSTAMP:20260209T000000Z
DTSTART;VALUE=DATE:20340819
DTEND;VALUE=DATE:20340820
SUMMARY:Peter Wildeford’s AGI Timelines Model: Model distribution reaches
  75% cumulative AGI probability
DESCRIPTION:Forecast: Seventy-five percent of the model snapshot reaches it
 s AGI criterion by this date.\n\nSource timing: 75th percentile: August 19
 \, 2034\n\nCalendar anchor: 2034-08-19 (This is the exact extracted percen
 tile date from the frozen model extraction.)\n\nSource context: AGI Timeli
 nes Model default Monte Carlo distribution\, upstream commit 664e1cec (202
 6-02-09)\n\nForecast uncertainty: The percentile summarizes sampled assump
 tions and is not a discrete predicted milestone.\n\nSource: https://github
 .com/peterhurford/agi_timelines
URL:https://github.com/peterhurford/agi_timelines
TRANSP:TRANSPARENT
END:VEVENT
BEGIN:VEVENT
UID:wildeford-agi-timelines.agi-p90@ai-forecast-calendar
DTSTAMP:20260209T000000Z
DTSTART;VALUE=DATE:20370307
DTEND;VALUE=DATE:20370308
SUMMARY:Peter Wildeford’s AGI Timelines Model: Model distribution reaches
  90% cumulative AGI probability
DESCRIPTION:Forecast: Ninety percent of the model snapshot reaches its AGI 
 criterion by this date.\n\nSource timing: 90th percentile: March 7\, 2037\
 n\nCalendar anchor: 2037-03-07 (This is the exact extracted percentile dat
 e from the frozen model extraction.)\n\nSource context: AGI Timelines Mode
 l default Monte Carlo distribution\, upstream commit 664e1cec (2026-02-09)
 \n\nForecast uncertainty: The model caps and parameter distributions affec
 t its tail behavior.\n\nSource: https://github.com/peterhurford/agi_timeli
 nes
URL:https://github.com/peterhurford/agi_timelines
TRANSP:TRANSPARENT
END:VEVENT
END:VCALENDAR