Line
Passages Flagged as Likely Heavily AI-Written: S&P 500 Calls and 10-Ks
Share of passages an open AI-text detector flags, in earnings-call prepared remarks (trailing 90 days) and 10-K risk factors (latest 10-K per company)
Plays the chart through its history. Pauses are where the data set a record or moved most.
Read both lines against the baseline, not against zero: the cutoff was set so that a fixed small share of human text from before ChatGPT crosses it, and that share is stated in the methodology. A line near the baseline means no detectable change; only a sustained rise well above it would point to more heavily AI-written text. The detector (TMR, an open model at a pinned revision) catches whole passages written by a model, not light AI editing, and Palanor never scores or names any company or document.
An open AI-text detector reads passages of what S&P 500 companies script for their earnings calls and write in their annual risk factors, and counts the share it flags as likely heavily AI-written. Set beside the AI in Text Index, which tracks the vocabulary language models overuse, it separates a change in word choice from text handed over to a model wholesale.
What this chart reads
- Passages Flagged as Likely Heavily AI-Written: S&P 500 Earnings-Call Prepared Remarksas “Earnings-call prepared remarks”latest 2026-10-23
- Passages Flagged as Likely Heavily AI-Written: S&P 500 10-K Risk Factorsas “10-K risk factors”latest 2026-10-23
Sources
Passages Flagged as Likely Heavily AI-Written: S&P 500 Earnings-Call Prepared Remarks. Source: Palanor calculation from earnings-call transcripts received through a licensed data provider. A value is published only when at least 150 companies have a scored call in the window.
Passages Flagged as Likely Heavily AI-Written: S&P 500 10-K Risk Factors. Source: Palanor calculation from Item 1A (Risk Factors) of annual reports on Form 10-K filed with the SEC (public domain). A value is published only when at least 150 companies have a scored 10-K in the window.
Charts are drawn from Palanor's signal library and recomputed from source data on every view. Transformations (changes, indexing, year-on-year rates) are Palanor's.