pragma.vision Technology observatory & strategic foresight

Verification register Security & Identity

Readiness verdict

Claude Text Watermarking

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

As of
2026-08-20
Revision
1
Method
v1.3.0

Current reading

Readiness band and full integer triple

AI-assisted assembly · derived results

Readiness band

Watch

Primary summary from verified readiness

Confidence
58% · fresh
Computed at
2026-08-20T10:56:06.658161+00:00
Claimed
60

Public ambition and stated capability

Reported
52

Observed practitioner reporting

Verified
46

Independently supported evidence

Gap
+14

Claimed leads verified

Evidence strength Critical

Decision

What the current evidence supports

Human editorial judgment · 2026-08-20

Track; not yet

Why
First hyperscale, default-on EU AI Act Article 50 text-provenance rollout and directly relevant to pv's AI-discoverability/DRCP posture, but the detection tool that would make it actionable isn't public yet, and both Anthropic's own disclosure and independent researchers already flag spoofing/removal via paraphrase or cross-model rewriting — too immature to build a workflow dependency on today.
Next
Re-evaluate once Anthropic ships the public watermark detection API and independent researchers publish robustness/false-positive testing against paraphrase and cross-model rewriting attacks.

Constraints

Blockers

No named blocker is present in the current public projection.

Evidence summary

Derived counts

AI-assisted assembly

Total
5
Tier 1
0
Tier 2
3
Tier 3
2
Supports
1
Contradicts
2
Context
2
Latest observed
2026-08-17

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

Publication record

Revisions

Initial public reading

  1. 2026-08-17 Reading moved from watch 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.