← Predictions register under audit
Constitutional AI Adoption
By 2028, at least 3 of the top 5 AI labs will adopt constitutional/principle-based alignment approaches as their primary safety methodology.
The previous Predictions Observatory has been withdrawn while its dates and scoring rules are audited. Historical entries mixed source-visible dates, retrospective similarities and heterogeneous outcome classes. All twelve entries asserted a public date of 1 October 2024 for which no immutable source was located.
An entry will return only when it has frozen wording, an independently verifiable public date, a defined time horizon, a base-rate assessment, a prospective resolution rule fixed before the outcome, a named adjudicator and an append-only correction history. No withdrawn entry is counted as confirmed, and none carries evidential weight anywhere on this site.
Entries remain visible for correction history only. See corrections.
Falsification criteria
- If by 2028, fewer than 3 top labs use constitutional approaches, this is falsified
- If a fundamentally different paradigm dominates (e.g., pure interpretability, formal verification), this is partially falsified
- Measurement: Examine published safety methodologies of OpenAI, Anthropic, Google DeepMind, Meta AI, xAI
Supporting evidence
- Anthropic's Constitutional AI methodology 2023-05-09
- Google DeepMind's principles-based Gemini safety 2024-02-08
Timeline
- Made public
- Expected resolution
Checkpoints
- 2026-01-01 Mid-term assessment