What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Only a detector built to check OpenAI’s watermark can test for that specific signal—and its result is not definitive proof of who wrote a passage. OpenAI’s textGrain detector is initially available only to approved researchers and expert organizations. Its reported performance varies with passage length, subject, and editing. Generic AI-writing detectors estimate whether text resembles AI output; they do not verify a ChatGPT watermark.
What a ChatGPT text watermark detector checks
OpenAI calls its watermarking system textGrain. It adds an invisible statistical signal through a model’s word choices, and a corresponding detector looks for that OpenAI-specific signal in a passage. It is not the same as recognizing a general “AI writing style.” OpenAI’s October 2026 announcement says the company plans to publish more technical details and open-source the technology; those are plans, not features readers can assume are already available.
A generic AI text detector instead estimates authorship from statistical or stylistic patterns. It does not establish whether text contains textGrain. OpenAI’s API documentation says its provenance check does not currently detect content from other AI providers. The available information also does not establish that commercial tools such as Turnitin or GPTZero can read OpenAI’s watermark. OpenAI’s text provenance guide
Who can use OpenAI’s detector, and where watermarking is rolling out
As of OpenAI’s October 5, 2026 announcement, detector applications are initially limited to approved researchers and expert organizations—not generally available as a public checker. OpenAI says API customers globally can opt in to watermarking for select models; it is off by default in the API. For eligible ChatGPT and Codex text output, the company says it will introduce invisible watermarking in the EU over the coming weeks, across plans. It is not launching as a global ChatGPT default. OpenAI’s announcement
Eligibility for every model and the eventual public access terms are not established in that announcement. The rollout should therefore be understood as limited and staged, rather than a signal that all ChatGPT text is currently watermarked.
How reliable are textGrain’s reported results?
OpenAI reports detection results at a target false-positive rate of 1%. For psychology passages, it detected a watermark in about 80% of 200-token passages and about 95% of 400-token passages. It reports substantially lower detection for mathematics, where there is less flexibility in word choice. These are OpenAI’s stated evaluation results, not guaranteed accuracy across languages, subjects, models, or real-world editing conditions. OpenAI’s announcement
Editing can make detection less dependable. In OpenAI’s reported tests on 400-token passages, replacing 10% of words with synonyms reduced detection from about 92% to 66%; replacing 25% reduced it to 17%. These figures describe the company’s evaluation and should not be generalized to every kind of rewrite or text.
The results illustrate why reliability depends on more than the detector’s name. A useful assessment asks what signal is being checked, which models and languages are covered, how long a sample must be, the false-positive rate at the stated threshold, and how editing or translation affects detection. The European Commission’s technical report also frames evaluation in terms of effectiveness, robustness, reliability across scenarios, user interpretability, accessibility, and interoperability. European Commission technical report
Why a negative result does not rule out ChatGPT
OpenAI’s API guide lists several reasons text originating from OpenAI might not be detected: metadata may be stripped, the watermark may be tampered with or degraded, the text may come from a legacy model, or it may predate the availability of provenance signals. Editing can also weaken the statistical signal. A negative result therefore means the detector did not identify the signal under its conditions; it does not prove the passage was written by a person or never came from OpenAI. OpenAI’s text provenance guide
Likewise, a positive watermark result is evidence of a provider-specific signal, not a complete account of who wrote, edited, or submitted the text. It should be interpreted within the detector’s scope and limitations.
Rank #4
Why generic AI detector scores are not watermark evidence
OpenAI’s 2023 AI Text Classifier was an authorship classifier, not a watermark verifier. On its English challenge set, it identified 26% of AI-written text as “likely AI-written” and falsely labeled 9% of human-written text as AI-written. OpenAI said it was unreliable on short text, performed significantly worse outside English and on code, and should not be used as a primary decision-making tool. Those historical figures apply to that classifier and evaluation—not to textGrain. OpenAI’s 2023 classifier announcement
A 2023 academic study likewise found that recursive paraphrasing could significantly reduce detection rates for the detector types it evaluated. It was not a test of textGrain. The study
Free tools Windows power users keep installed
One-click scans. No signup required.
Google’s SynthID Text is a separate watermarking system, not a ChatGPT detector. Google describes its detection as probabilistic and notes that thorough rewriting or translation can reduce confidence; it also says watermarking is less effective when factual precision leaves little room to vary word choices. This comparison reinforces the importance of distinguishing watermark verification from generic authorship estimation, but it does not establish how textGrain performs. Google’s SynthID Text documentation
How to use detector results in a high-stakes decision
Do not treat a generic detector score—or a watermark check on its own—as proof of misconduct or authorship. OpenAI characterizes watermarking and detection as early technologies with significant limitations. For an academic or employment concern, treat a detector result as one limited signal and consider corroborating evidence such as version history, drafts, process documentation, and a fair conversation with the writer. OpenAI’s statement on text watermarking
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
