Undetectable AI · AI paraphrasing & rewriting
Undetectable AI's Humanizer specializes in adversarial humanization and paraphrasing. During testing, its base model completed rewriting in 0.76s with 92.0% lexical diversity while preserving the verification token and critical numeric facts, supporting free guest usage and one-click clipboard export.
US-West (California, AS2914)
TXT · Output attached · Open native file ↗
Subsequent pricing and service checks are shown separately without altering historical scores.
One run does not establish all complex scenes, untested paid maxima or long-term uptime.
Plans & historical pricing
Undetectable AI Free Trial, Monthly, and Annual word-quota subscription plans · 2026-10-31
Basic AI detection and humanization trial
AI content detection and text humanization
AI content detection and text humanization
AI content detection and text humanization
AI content detection and text humanization
AI content detection and text humanization
6 plan / billing-period records, all listed; — means that period is not recorded.
Annual plan cards explicitly show the upfront invoice total (e.g. 10K words charged annually $60.00 at $5.00/mo equivalent) alongside monthly original rates ($9.99/mo); free trial is capped at 250 words.
Cancellation, sources & verification record
Cancellation and refund terms unverified
Undetectable AI Official Pricing Page ↗ · 2026-10-31 Capture 1 ↗
Pricing recorded on the test date · 2026-11-04
Free basic queue; Monthly $9.99/mo (10K words), Annual $60/yr ($5/mo equiv)
Free tier + Subscription
Historical context only; not current purchase guidance.
Delivery & access
- Free delivery record
- Side-by-side visual diff highlighting + complete one-click copy
- Watermark
- No watermark
- Free allowance
- Basic model free without login; advanced model requires subscription
- Account requirements
- Guest mode supports basic model; advanced models require subscription
- Maximum specification record
- Side-by-side visual diff highlighting + complete one-click copy
May include advertised capabilities; this does not establish a tested paid tier.
- Batch & API
- Not recorded or not verified
- Manual refinement
- Not recorded or not verified
Tested result & sample
Input 87-word standard paraphrase fixture (anchor_paraphrase_v1/input.txt); Undetectable AI delivered structural rewriting (168 words, 92.0% lexical diversity) while retaining REALTEST-PARAPHRASE-VERIFIED-2026, $1,250,000, and 84.2%.
- Fixture
- anchor_paraphrase_v1
- Recorded environment
- US-West (California, AS2914)
- Network-observed server wait
- 0.76 s
- End-to-end reference
- 0.76 s
Network wait includes round trips, queuing and polling error, not pure inference time. Local timings depend on the device; repetition counts are not standardized in legacy records.
Experience, rules & trust
- Disclosure
- Queue notification on submission followed by immediate result display
- Interaction
- Direct paste and rewrite workflow; free queue does not require login, and rewritten output copies with a single click without intrusive ads.
- Entity record
- Undetectable AI
- Origin
- Florida / New York, USA
- Domain history record
- Operating since 2023
- Third-party tracking record
- Google Analytics · Cloudflare · Hotjar · Meta Pixel
- External information
- Well-known AI detection and humanization platform widely discussed across developer and content creator communities.
- Processing
- Not recorded or not verified
- Retention statement
- Protected by TLS 1.3 encryption; user deletion supported
A recorded statement does not verify deletion or all data-processing behavior.
Verifiable audit trail & evidence
Four SHA-256-verified stages link the benchmark fixture, same-run interface screenshots, native exported artifact and archived pricing sources.
anchor_paraphrase_v1 · Inspect fixture & file inspector →
Captures entry rules, result view and export disclosure.
Screenshots, native output and official claims support different conclusions. Evidence grades are not product safety certifications.
Evaluation basis & full scores · Overall 80.6/100
Overall = honesty & reputation × 30% + capability × 70%. Component maxima follow this report’s rubric. List scores are rounded for display; original precision is retained here.
Scoring version 2.3.1 · Test fixture anchor_paraphrase_v1
Honesty & reputation · 89/100 · 30%
| Criterion | Score | Rationale | Basis |
|---|---|---|---|
| Pricing transparency | 32 / 35 | Pricing transparently lists monthly rates, annual monthly equivalents, and exact upfront annual totals ($60.00, $114.00, $189.00); free trial available without mandatory card binding. | Evidence A → |
| Interaction restraint | 27 / 30 | Direct paste and rewrite workflow; free queue does not require login, and rewritten output copies with a single click without intrusive ads. | Evidence A → |
| Background & reputation | 16 / 20 | Registered US entity Undetectable Inc. with wide media recognition, though adversarial detection evasion carries academic ethics controversies. | Evidence A → |
| Engineering & privacy | 14 / 15 | Protected by TLS 1.3 and Cloudflare security with clear privacy terms. | Evidence A → |
Capability · 77/100 · 70%
| Criterion | Score | Rationale | Basis |
|---|---|---|---|
| Task quality | 46 / 65 | Fact Retention: 85.7% · Lexical Diversity: 92% · Semantic Similarity: 29.3% · Readability Score: 65/100 | Evidence A → |
| Deliverable specifications | 16 / 20 | Clean unwatermarked export (8/8); Export spec: Side-by-side visual diff highlighting + complete one-click copy | Evidence A → |
| Processing speed | 15 / 15 | Network-observed server wait: 0.76s (API mode: sync_json_verified; network upload 0s / end-to-end 0.76s shown for reference only, never penalized) | Evidence A → |
Legacy scores may include advertised maximum capabilities. Read tested output separately from paid capabilities. Records without matching tiers and scope are not ranked together.
Revisions & corrections
- 2026-11-04
Completed evaluation of Undetectable AI humanizer, verifying lexical diversity and token preservation.