Triple
T1923010
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Shane Legg |
E40165
|
entity |
| Predicate | spokeAbout |
P3281
|
FINISHED |
| Object | existential risks from advanced AI |
—
|
LITERAL FINISHED |
How this triple was built (2 steps)
Every LLM step that produced this triple, in pipeline order — named-entity classification, the disambiguation choices (the exact options shown, with the pick highlighted), and the generated description. The batch + timestamp of each is in the Provenance table below.
NER
Named-entity recognition
gpt-5-mini
Instruction
Given a phrase, classify it is english named entity (e.g., persons, organizations, works of art) in Latin script, or not (e.g., literals, dates, URLs, verbose phrases). For disambiguation, the statement where the phrase occurs as object is also given. Please return a JSON object with `phrase` (string, the phrase being analyzed) and `is_ne` (boolean, indicating whether the phrase is a Named Entity).
Input
Phrase: existential risks from advanced AI | Statement: [Shane Legg, spokeAbout, existential risks from advanced AI]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: spokeAbout Context triple: [Shane Legg, spokeAbout, existential risks from advanced AI]
-
A.
spokeAt
Indicates that a person delivered a talk, speech, or presentation at a particular event or location.
-
B.
spokenOn
Indicates that an utterance or speech act occurred at or during a specific time or date.
-
C.
hasSpokenAbout
chosen
Indicates that one entity has verbally expressed, discussed, or mentioned another entity or topic.
-
D.
spokenBefore
Indicates that one entity has spoken or produced speech earlier in time than another entity.
-
E.
alsoSpeak
Indicates that an entity, in addition to another language or mode of communication already mentioned, speaks this additional language or communicates in this additional way.
- F. None of above.
Provenance (3 batches)
The batch behind each pipeline step, in order, with when it ran. Timestamps are batch-level — stages were processed in waves, so the object chain (NER → NED1 → NEDg → NED2) reads in order, but predicate / elicitation batches can sit in a different wave.
| Step | Stage | Batch ID | Status | When |
|---|---|---|---|---|
| creating | Elicitation | batch_69a8864298748190a2f2fd34f7ef8d77 |
completed | March 4, 2026, 7:21 p.m. |
| NER | Named-entity recognition | batch_69abb23459ac819088ded5bfac9d4aad |
completed | March 7, 2026, 5:05 a.m. |
| PD | Predicate disambiguation | batch_69abafed2ab481908920334e77b1021b |
completed | March 7, 2026, 4:56 a.m. |
Created at: March 4, 2026, 7:35 p.m.