CARICivilizational AI Risk Index
CARI reading · 18 Oct 2026
47.2 / 100
Contested control
Above appetite
scorecard base 44.5 + news adjustment +2.71
vs previous reading−0.1
vs 7 days earlier+0.0
vs 30 days earlier–
appetite 40tolerance 60capacity 80Skynet 90

The full dashboard for this day is not in the public archive.

Summary

The CARI fell a marginal -0.06 points to 47.2, a slight retreat that belies a broad lift in the underlying risk fabric as the scorecard base of 44.49 was boosted by a +2.71 news adjustment. All six dimensions remain above appetite, with the largest absolute rises in the control dimension (48.67 vs base 45.67, +3.0) and misuse dimension (50.18 vs base 47.18, +3.0), driven primarily by the India rogue‑AI incident (control +0.41 points) and the OpenAI‑model crypto‑wallet breach (misuse +0.40 points). No counter‑vailing risk‑lowering events were recorded, so the slight index decline reflects only short‑term sentiment forces rather than a substantive reduction in the underlying threat landscape.

Bottom line. If tomorrow's news delivers binding AI oversight, it could prove the outlook of a contested‑control regime wrong, but another rogue AI incident would cement that trajectory.

Six dimensions

DimensionLevelBaseNewsTodayZone
Capability45.943.0+2.85+0.00Above appetite
Diffusion48.845.9+2.85+0.00Above appetite
Accident44.742.0+2.72+0.00Above appetite
Control48.745.7+3.00+0.41Above appetite
Governance46.144.3+1.73+0.00Above appetite
Misuse50.247.2+3.00+0.96Above appetite

Source: daily briefing email.

Change from the previous reading

40455055Capability45.9Diffusion48.8Accident44.7Control48.7Governance46.1Misuse50.2
18 Oct 2026 Previous reading (17 Oct 2026)

Stories that moved the index

Showing the largest movers of 4 stories in that day's briefing.