pragma.vision Your technology observatory

Verification register AI & Agents

Readiness verdict

DeepSeek V4

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

As of
2026-06-28
Revision
1
Method
v1.0.0

Current reading

The readiness gap, in one scan

AI-assisted assembly · derived results

Claimed
88

Public ambition and stated capability

Reported
76

Observed practitioner reporting

Verified
65

Independently supported evidence

Gap
+23

Claimed minus verified

Evidence strength Strong

Decision

What the current evidence supports

Human editorial judgment · 2026-06-28

Adopt with guardrails

Why
Strong open-weight coding capability at a fraction of frontier closed-model cost (AA: V4-Pro runs the Index for $1,071 vs Opus 4.7 $4,811) is compelling, but a 94% hallucination rate and a verified ~8-month frontier gap mean it cannot be trusted unsupervised on high-stakes or long-horizon work.
Next
Pilot V4-Pro (high/max tier) on a non-critical coding workflow with mandatory abstention prompting + human review on long-context tasks; benchmark internal hallucination rate before any production path.

Constraints

Blockers

No named blocker is present in the current public projection.

Evidence summary

Derived counts

AI-assisted assembly

Total
7
Tier 1
1
Tier 2
2
Tier 3
4
Supports
2
Contradicts
4
Context
1
Latest observed
2026-06-13

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

Publication record

Revisions

Initial public reading

This is the initial public reading. No earlier readiness change is recorded.

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.