pragma.vision Technology observatory & strategic foresight

Verification register AI & Agents

Readiness verdict

OpenAI Codex (GPT-5.5)

A dated reading of what is claimed, reported, and independently verified in the current evidence.

As of
2026-10-01
Revision
3
Method
v1.4.0

Current reading

Readiness band and full integer triple

AI-assisted assembly · derived results

Readiness band

Watch

Primary summary from verified readiness

Confidence
84% · fresh
Computed at
2026-10-01T01:22:22.621563+00:00
Claimed
65

Public ambition and stated capability

Reported
55

Observed practitioner reporting

Verified
45

Independently supported evidence

Gap
+20

Claimed leads verified

Evidence strength Strong

Decision

What the current evidence supports

Agent-assisted assessment · independently reviewed · 2026-10-01

Use for supervised coding while preparing a supported-model migration.

Why
Independent repository tests show useful performance with substantial failures, and GPT-5.5 retires from Codex on October 14, 2026.
Next
Evaluate a supported replacement on your repositories and retain executable tests and human review before merging.

Constraints

Blockers

  • Codex retirement on October 14, 2026
  • Generated patches require executable tests and review
  • Performance varies by repository and reasoning setting
  • Limited independent long-horizon production validation

Evidence summary

Derived counts

AI-assisted assembly

Total
11
Tier 1
5
Tier 2
6
Tier 3
0
Supports
6
Contradicts
3
Context
2
Latest observed
2026-09-14

Counts and dates only. Raw signals, private excerpts, trust records, and internal corpus material are not published here.

Publication record

Revisions

Revision 3

  1. 2026-07-19 Reading moved from mature to mature.
  2. 2026-09-22 Reading moved from mature to mature.
  3. 2026-10-01 Reading moved from mature to watch.

Your opinion

Tell us anything.

What works, what doesn't, what's missing — especially about our watches, lenses, and the register itself. Anonymous is fine; leave an email if you'd like a reply.