Benchmarks

Academi.cx vs Winston AI

We benchmarked Academi.cx and Winston AI against Turnitin across 429 documents, measuring score differences, false flags, and missed AI.

Brayden · 2 min read

For this benchmark, we submitted the same 429 documents to Academi.cx and Winston AI through their APIs, then compared each tool’s scores with the documents’ actual Turnitin AI scores. Together those documents add up to close to 944,000 words, averaging about 2,200 words each, ranging from a 311 word short answer to a 25,000 word thesis.

How close were the predictions

Across all 429 documents, Academi.cx averaged 1.92 points off the real Turnitin score. Winston AI averaged 9.43 points off. Its median error is only 1.0, meaning the typical prediction was fairly close, but a share of much larger misses pulled its average error well above that.

Paired scatterplots show the AI score for all 429 matched documents against Turnitin. Academi.cx appears in blue and Winston AI in gray, with identical 0 to 100 percent axes. The dashed diagonal marks matching scores. Mean absolute difference from Turnitin is 1.92 percentage points for Academi.cx and 9.43 for Winston AI. No documents are omitted; some points overlap.

Breaking it down by document type

The 429 documents split into 220 that Turnitin scored as 0% AI, 99 it scored as 100% AI, and 110 that fell somewhere in between.

Grouped bars show average absolute deviation from Turnitin in percentage points. All documents (n = 429): Academi.cx 1.92 points and Winston AI 9.43 points. Turnitin 0% (n = 220): Academi.cx 0.18 points and Winston AI 9.22 points. Turnitin 20–99% (n = 110): Academi.cx 3.95 points and Winston AI 14.69 points. Turnitin 100% (n = 99): Academi.cx 3.53 points and Winston AI 4.04 points. Blue represents Academi.cx and gray Winston AI. Lower is better. All bars share a 0 to 40 point scale.

Winston AI's numbers flip depending on the bucket. On the 100% documents it averaged 4.04 points of error, close to Academi.cx's 3.53, its strongest showing of the three groups. But on the 0% documents it averaged 9.22 points off against Academi.cx's 0.18, and in the mixed bucket it climbed to 14.69 against Academi.cx's 3.95. Winston AI is better at catching fully AI written work than it is at leaving genuine human writing alone.

False flags and missed AI

Academi.cx had a 0% false flag rate and a 0% missed AI rate across all 429 documents, meaning it never scored a genuinely 0% document above 20%, and never scored a genuinely 100% document below 80%. Winston AI came in at a 14.5% false flag rate and a 7.1% missed AI rate, the highest false flag rate of any tool covered in this benchmark so far.

Sentence level accuracy

We also checked how often each tool correctly classified individual sentences as AI written or human written, matched against Turnitin's own sentence level calls. Across all 429 documents, Academi.cx correctly classified 96.64% of sentences. Winston AI came in at 76.59%.

The takeaway

Across all 429 documents, Academi.cx stayed close to the actual Turnitin percentage on every measure checked. Winston AI held up reasonably well on fully AI written content but its false flag rate stood out as its clearest weakness, wrongly flagging genuine human writing more often than any other tool in this benchmark.

See the full benchmark results.

Related reading