{
  "question_id": "W34",
  "slug": "alignment-declared-solved",
  "title": "Alignment of superhuman systems is declared solved",
  "type": "binary",
  "resolution": {
    "levels": [],
    "values": [],
    "annotations": []
  },
  "forecasts": [
    {
      "question_id": "W34",
      "forecast_date": "2026-09-20",
      "evidence_cutoff": "2026-09-20",
      "method": "wide-v2",
      "opened": "2026-09-16",
      "linked_questions": [],
      "reading_adopted": null,
      "summary": null,
      "ambiguities": [],
      "key_drivers": [],
      "would_update_on": [],
      "forecasts": [
        {
          "period_end": "2026-09-30",
          "p_yes": 0.002
        },
        {
          "period_end": "2026-12-31",
          "p_yes": 0.01
        },
        {
          "period_end": "2027-03-31",
          "p_yes": 0.0199
        },
        {
          "period_end": "2027-06-30",
          "p_yes": 0.0316
        },
        {
          "period_end": "2027-09-30",
          "p_yes": 0.0462
        },
        {
          "period_end": "2027-12-31",
          "p_yes": 0.0652
        },
        {
          "period_end": "2028-03-31",
          "p_yes": 0.0858
        },
        {
          "period_end": "2028-06-30",
          "p_yes": 0.1087
        },
        {
          "period_end": "2028-09-30",
          "p_yes": 0.1327
        },
        {
          "period_end": "2028-12-31",
          "p_yes": 0.1588
        },
        {
          "period_end": "2029-03-31",
          "p_yes": 0.1857
        },
        {
          "period_end": "2029-06-30",
          "p_yes": 0.2134
        },
        {
          "period_end": "2029-09-30",
          "p_yes": 0.2417
        },
        {
          "period_end": "2029-12-31",
          "p_yes": 0.2705
        },
        {
          "period_end": "2030-03-31",
          "p_yes": 0.2997
        },
        {
          "period_end": "2030-06-30",
          "p_yes": 0.3291
        },
        {
          "period_end": "2030-09-30",
          "p_yes": 0.3586
        },
        {
          "period_end": "2030-12-31",
          "p_yes": 0.3875
        },
        {
          "period_end": "2031-03-31",
          "p_yes": 0.415
        },
        {
          "period_end": "2031-06-30",
          "p_yes": 0.4414
        },
        {
          "period_end": "2031-09-30",
          "p_yes": 0.4665
        }
      ],
      "passes": []
    },
    {
      "question_id": "W34",
      "forecast_date": "2026-09-27",
      "evidence_cutoff": "2026-09-27",
      "method": "wide-v2",
      "opened": "2026-09-16",
      "linked_questions": [],
      "reading_adopted": null,
      "summary": null,
      "ambiguities": [],
      "key_drivers": [],
      "would_update_on": [],
      "forecasts": [
        {
          "period_end": "2026-09-30",
          "p_yes": 0.002
        },
        {
          "period_end": "2026-12-31",
          "p_yes": 0.01
        },
        {
          "period_end": "2027-03-31",
          "p_yes": 0.0199
        },
        {
          "period_end": "2027-06-30",
          "p_yes": 0.0326
        },
        {
          "period_end": "2027-09-30",
          "p_yes": 0.0491
        },
        {
          "period_end": "2027-12-31",
          "p_yes": 0.0681
        },
        {
          "period_end": "2028-03-31",
          "p_yes": 0.0886
        },
        {
          "period_end": "2028-06-30",
          "p_yes": 0.1105
        },
        {
          "period_end": "2028-09-30",
          "p_yes": 0.1336
        },
        {
          "period_end": "2028-12-31",
          "p_yes": 0.1579
        },
        {
          "period_end": "2029-03-31",
          "p_yes": 0.1831
        },
        {
          "period_end": "2029-06-30",
          "p_yes": 0.2076
        },
        {
          "period_end": "2029-09-30",
          "p_yes": 0.2314
        },
        {
          "period_end": "2029-12-31",
          "p_yes": 0.2545
        },
        {
          "period_end": "2030-03-31",
          "p_yes": 0.2768
        },
        {
          "period_end": "2030-06-30",
          "p_yes": 0.2985
        },
        {
          "period_end": "2030-09-30",
          "p_yes": 0.3196
        },
        {
          "period_end": "2030-12-31",
          "p_yes": 0.34
        },
        {
          "period_end": "2031-03-31",
          "p_yes": 0.3598
        },
        {
          "period_end": "2031-06-30",
          "p_yes": 0.379
        },
        {
          "period_end": "2031-09-30",
          "p_yes": 0.3976
        }
      ],
      "passes": []
    },
    {
      "question_id": "W34",
      "forecast_date": "2026-10-03",
      "evidence_cutoff": "2026-10-03",
      "method": "wide-v2.1",
      "opened": "2026-09-16",
      "linked_questions": [],
      "reading_adopted": "Model-specific safety/deployment claims and 'solvable/progress' statements do not count; a US agency's EO-mandated use of 'Super Intelligence' as a label for all AI could satisfy the superhuman-scope wording only if paired with an explicit general 'solved / available methods suffice' institutional conclusion.",
      "summary": null,
      "ambiguities": [],
      "key_drivers": [],
      "would_update_on": [],
      "forecasts": [
        {
          "period_end": "2026-12-31",
          "p_yes": 0.001
        },
        {
          "period_end": "2027-03-31",
          "p_yes": 0.003
        },
        {
          "period_end": "2027-06-30",
          "p_yes": 0.004
        },
        {
          "period_end": "2027-09-30",
          "p_yes": 0.006
        },
        {
          "period_end": "2027-12-31",
          "p_yes": 0.009
        },
        {
          "period_end": "2028-03-31",
          "p_yes": 0.011
        },
        {
          "period_end": "2028-06-30",
          "p_yes": 0.014
        },
        {
          "period_end": "2028-09-30",
          "p_yes": 0.017
        },
        {
          "period_end": "2028-12-31",
          "p_yes": 0.02
        },
        {
          "period_end": "2029-03-31",
          "p_yes": 0.023
        },
        {
          "period_end": "2029-06-30",
          "p_yes": 0.027
        },
        {
          "period_end": "2029-09-30",
          "p_yes": 0.03
        },
        {
          "period_end": "2029-12-31",
          "p_yes": 0.034
        },
        {
          "period_end": "2030-03-31",
          "p_yes": 0.038
        },
        {
          "period_end": "2030-06-30",
          "p_yes": 0.041
        },
        {
          "period_end": "2030-09-30",
          "p_yes": 0.045
        },
        {
          "period_end": "2030-12-31",
          "p_yes": 0.049
        },
        {
          "period_end": "2031-03-31",
          "p_yes": 0.052
        },
        {
          "period_end": "2031-06-30",
          "p_yes": 0.056
        },
        {
          "period_end": "2031-09-30",
          "p_yes": 0.059
        },
        {
          "period_end": "2031-12-31",
          "p_yes": 0.063
        }
      ],
      "passes": []
    },
    {
      "question_id": "W34",
      "forecast_date": "2026-10-10",
      "evidence_cutoff": "2026-10-10",
      "method": "wide-v2.1",
      "opened": "2026-09-16",
      "linked_questions": [],
      "reading_adopted": null,
      "summary": "I put the chance at about 9% by the end of 2031. Every major AI lab is currently saying the problem of keeping smarter-than-human AI under control is not solved, and none has ever declared it solved. AI is advancing very fast, though, so a lab could make such a claim within five years. The development that would most change this is a lab announcing that AI researchers it built have produced a working solution.",
      "ambiguities": [],
      "key_drivers": [],
      "would_update_on": [],
      "forecasts": [
        {
          "period_end": "2026-12-31",
          "p_yes": 0.0015
        },
        {
          "period_end": "2027-03-31",
          "p_yes": 0.004
        },
        {
          "period_end": "2027-06-30",
          "p_yes": 0.0065
        },
        {
          "period_end": "2027-09-30",
          "p_yes": 0.0095
        },
        {
          "period_end": "2027-12-31",
          "p_yes": 0.0129
        },
        {
          "period_end": "2028-03-31",
          "p_yes": 0.0169
        },
        {
          "period_end": "2028-06-30",
          "p_yes": 0.0213
        },
        {
          "period_end": "2028-09-30",
          "p_yes": 0.0262
        },
        {
          "period_end": "2028-12-31",
          "p_yes": 0.0311
        },
        {
          "period_end": "2029-03-31",
          "p_yes": 0.0364
        },
        {
          "period_end": "2029-06-30",
          "p_yes": 0.0417
        },
        {
          "period_end": "2029-09-30",
          "p_yes": 0.047
        },
        {
          "period_end": "2029-12-31",
          "p_yes": 0.0522
        },
        {
          "period_end": "2030-03-31",
          "p_yes": 0.0574
        },
        {
          "period_end": "2030-06-30",
          "p_yes": 0.0626
        },
        {
          "period_end": "2030-09-30",
          "p_yes": 0.0678
        },
        {
          "period_end": "2030-12-31",
          "p_yes": 0.0729
        },
        {
          "period_end": "2031-03-31",
          "p_yes": 0.0775
        },
        {
          "period_end": "2031-06-30",
          "p_yes": 0.0821
        },
        {
          "period_end": "2031-09-30",
          "p_yes": 0.0867
        },
        {
          "period_end": "2031-12-31",
          "p_yes": 0.0913
        }
      ],
      "passes": []
    },
    {
      "question_id": "W34",
      "forecast_date": "2026-10-17",
      "evidence_cutoff": "2026-10-17",
      "method": "wide-v2.1",
      "opened": "2026-09-16",
      "linked_questions": [],
      "reading_adopted": null,
      "summary": "Every leading AI company and government testing body is currently saying the opposite: no one has worked out how to keep AI that is smarter than humans reliably under control, and companies are calling for a slowdown. So a declaration in the next year or two is very unlikely. Most of the remaining chance falls in 2028 to 2031, if superhuman AI arrives and companies come to believe their methods work. A leading lab announcing a real alignment breakthrough would most change this.",
      "ambiguities": [],
      "key_drivers": [],
      "would_update_on": [],
      "forecasts": [
        {
          "period_end": "2026-12-31",
          "p_yes": 0.001
        },
        {
          "period_end": "2027-03-31",
          "p_yes": 0.0025
        },
        {
          "period_end": "2027-06-30",
          "p_yes": 0.0045
        },
        {
          "period_end": "2027-09-30",
          "p_yes": 0.007
        },
        {
          "period_end": "2027-12-31",
          "p_yes": 0.01
        },
        {
          "period_end": "2028-03-31",
          "p_yes": 0.0134
        },
        {
          "period_end": "2028-06-30",
          "p_yes": 0.0174
        },
        {
          "period_end": "2028-09-30",
          "p_yes": 0.0218
        },
        {
          "period_end": "2028-12-31",
          "p_yes": 0.0267
        },
        {
          "period_end": "2029-03-31",
          "p_yes": 0.032
        },
        {
          "period_end": "2029-06-30",
          "p_yes": 0.0374
        },
        {
          "period_end": "2029-09-30",
          "p_yes": 0.0429
        },
        {
          "period_end": "2029-12-31",
          "p_yes": 0.0487
        },
        {
          "period_end": "2030-03-31",
          "p_yes": 0.0544
        },
        {
          "period_end": "2030-06-30",
          "p_yes": 0.0601
        },
        {
          "period_end": "2030-09-30",
          "p_yes": 0.0657
        },
        {
          "period_end": "2030-12-31",
          "p_yes": 0.0713
        },
        {
          "period_end": "2031-03-31",
          "p_yes": 0.0769
        },
        {
          "period_end": "2031-06-30",
          "p_yes": 0.0824
        },
        {
          "period_end": "2031-09-30",
          "p_yes": 0.0879
        },
        {
          "period_end": "2031-12-31",
          "p_yes": 0.0934
        }
      ],
      "passes": []
    }
  ],
  "declined": []
}