Detecto
Sample report

Two pieces of writing on productivity. One human. One generated.

The same report shape Detecto returns to a logged-in reviewer. A human-written essay submitted to an opinion section, and an AI-generated post submitted to a business section, both about productivity. Read them with the signals open.

Side by side

The same Detecto report, on two passages on the same subject.

No vendor name appears in either report, by design and per privacy policy.

Human-writtenPublic share view
textrep_h_8b14f3c2
Detection report
First-person reflection on returning to the office after years of remote work.
6%Likely human
0%±4 pts100%
Top signals
  • Average perplexitySurprise per token relative to a base language model.
    74.2
  • BurstinessVariance in sentence-level perplexity across the passage.
    0.81
  • Idiom densityIdiomatic constructions and unconventional phrasing.
    6.4 / 1k
  • Lexical diversityType-token ratio across the passage.
    0.62 TTR
  • Sentence-length variance
    14.8
Sentence-level likelihood

I didn’t expect to miss the commute. For eleven years I complained about that train, and then one Tuesday it was gone, and so was the part of the day that was neither work nor home. The first month felt like a snow day that wouldn’t end. By the third, my kitchen table had a permanent laptop dent and I was taking calls in the closet, because it was the only quiet room in the house. Nobody warns you how loud the silence gets. No one drifts past your desk. No one overhears the half-formed idea you say out loud by accident and tells you it’s actually good. I’m more productive now, mostly, the afternoons are finally mine. But I shipped two things this quarter I’m proud of and can’t name a single hallway conversation that led to either one. Maybe that’s the trade. You get the deep hours back and hand over the accidental ones: the elevator, the burnt break-room coffee, the small talk that somehow added up to knowing people. I still don’t miss the train. I just didn’t realize it had been carrying something I’d eventually want back.

Reviewer notes

Scored Likely human. Burstiness and perplexity sit well above the AI range, idiom density is high, and sentence length swings hard from line to line, all consistent with unprompted human writing. A reviewer can clear this with confidence.

All signals
  • Average perplexitySurprise per token relative to a base language model.
    74.2
  • BurstinessVariance in sentence-level perplexity across the passage.
    0.81
  • Idiom densityIdiomatic constructions and unconventional phrasing.
    6.4 / 1k
  • Lexical diversityType-token ratio across the passage.
    0.62 TTR
  • Sentence-length variance
    14.8
  • Confidence varianceDetector’s own variance across resampled detection passes.
    0.04
Perplexity & burstiness
Perplexity
Burstiness
AI-generatedPublic share view
textrep_a_2f30d9a1
Detection report
Short five-paragraph essay weighing the pros and cons of remote work.
92%Likely AI-generated
0%±5 pts100%
Top signals
  • Average perplexitySurprise per token relative to a base language model.
    18.3
  • BurstinessVariance in sentence-level perplexity across the passage.
    0.21
  • Connector frequency"First", "Second", "Ultimately", "In today’s world" and similar.
    11.2 / 1k
  • Sentence-length variance
    3.4
  • Lexical diversity
    0.41 TTR
Sentence-level likelihood

In today’s rapidly evolving world, remote work has become an increasingly important topic for businesses and employees alike. First and foremost, remote work offers a wide range of benefits, including increased flexibility, reduced commuting time, and improved work-life balance. Moreover, numerous studies have shown that remote employees can be just as productive, if not more productive, than their in-office counterparts. However, it is important to consider the potential challenges, such as feelings of isolation, communication barriers, and the difficulty of separating work from personal life. Ultimately, the success of remote work depends on a variety of factors, including company culture, available technology, and individual preferences. In conclusion, while remote work presents both advantages and disadvantages, it is clear that it will continue to play a significant role in the future of work.

Reviewer notes

Scored Likely AI-generated. Low perplexity, low burstiness, and a high connector frequency (“First”, “Moreover”, “Ultimately”) match templated model output, and the “In today’s world” opener is a common tell. Strong, consistent signal, but treat it as a reason to talk to the author, not as proof.

All signals
  • Average perplexitySurprise per token relative to a base language model.
    18.3
  • BurstinessVariance in sentence-level perplexity across the passage.
    0.21
  • Connector frequency"First", "Second", "Ultimately", "In today’s world" and similar.
    11.2 / 1k
  • Sentence-length variance
    3.4
  • Lexical diversity
    0.41 TTR
  • Confidence varianceDetector’s own variance across resampled detection passes.
    0.05
Perplexity & burstiness
Perplexity
Burstiness
Reviewer view

What the reviewer sees in their dashboard.

Notes pane, share toggle, export to PDF. Reports save privately to your dashboard, shared only when you choose.

Reviewer notes

A pane attached to every scan. Author of the note and timestamp logged.

Sign up to use it
Public share toggle

Generates a detec.to share link when the reviewer opts in. Default is private.

Sign up to use it
PDF export

One-click export of the full report (score, signals, highlights) for the editorial folder.

Sign up to use it

Honest limitations

A detection score is probabilistic, not proof. Treat the report as evidence to discuss, not a verdict to publish.

  • These two passages are well-separated in score because they are designed examples. Real submissions land all over the band.
  • The score is a probability, not a verdict. Treat it as evidence to discuss.
  • A 78% AI score is the start of a conversation. Don’t auto-reject.
  • Confidence widens for short input.
  • See /accuracy-and-limitations and /false-positives for the full posture.
See the report on your own copy

Run a scan. Read the signals. Decide.

Sign-in optional for the demo. Demo scans aren't saved.