Tracker›Metrics›domain-recurrence
domain-recurrenceThe share of a question's filter domains today that already appeared as filter domains for the same question in the same mode on an earlier measurement day. 1 = ChatGPT hands the search exactly the known domains, 0 = all new. From the second measurement day onward; runs without a domain filter produce no line. The motive is Konitzny's finding (24 Aug 2026) that the domain choice is a learned, stable signal (18.1% hard core across 500 prompts) - with our frozen questions the domain choice repeats even more strongly: 15 of 28 next-day pairs with an identical set, 9 of 25 domains on 4-5 of 5 days. If a question's value drops for good, ChatGPT has switched its source preset for that topic.
| Prompt | 25 Sept 2026 | 26 Sept 2026 | 27 Sept 2026 | 28 Sept 2026 | 29 Sept 2026 | 30 Sept 2026 | 1 Oct 2026 | 2 Oct 2026 | 3 Oct 2026 | 4 Oct 2026 | 5 Oct 2026 | 6 Oct 2026 | 7 Oct 2026 | 8 Oct 2026 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| K10-01 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K11-01 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K11-02 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K2-01 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K2-02 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K3-01 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K3-02 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K5-01 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K5-01 thinking | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K5-02 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K6-01 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K6-01 thinking | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K6-02 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K7-01 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K7-02 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| K9-02 instant | – | – | – | – | – | – | – | – | – | – | – | – | – | – |