When I ask Grok for a timeline of an incident discussed on X, I need to know when things happened—not just when someone posted about them.

A post published on Tuesday might describe an outage from Monday night. Another could reshare an old screenshot, and a later update might correct the original time. Mixing those together could produce a plausible but misleading sequence.

Has anyone tested Grok on event timeline reconstruction with those distinctions made explicit? I’d like each entry to separate:

• The claimed event time, including its time zone if available.
• The publication time of the supporting post or report.
• Whether the event time is stated directly, inferred, or still unknown.

I’d rather see “time not established” than have a posting timestamp silently used as the event time. Is asking for those fields enough to expose the ambiguity, or do you verify the chronology independently even when Grok provides links?