AI-assisted tools are starting to show up in programme analysis — flagging logic errors, float erosion, and pattern changes across programme revisions far faster than a manual review ever could.
What they're not doing, at least not yet, is replacing the judgement calls that actually decide a delay dispute: whether a particular activity was genuinely on the critical path, whether a resequencing was a contractor choice or a forced response, whether two causes were truly concurrent or just close in time. The tools are getting better at showing you where to look. They're not yet good at telling you what it means.
