""The dirty secret," a veteran eDiscovery consultant told me, "is that AI still can't actually handle these massive document sets." While the industry rushes toward AI adoption, embracing its demonstrated strengths and - for some insane and persistent reason - its well-documented weaknesses, one hurdle it can't quite surmount is the ever-expanding size of discovery. AI is, rightly, lauded for its capacity to analyze and summarize large amounts of data, but crunching 100 deposition transcripts is one thing and getting on top of terabytes worth of discovery is quite another. Context windows just aren't that big."
"CoCounsel Legal's new bulk document review feature allows analysis of 10,000 documents at a clip, returning sortable, user-friendly results. The secret sauce is in taking these big batches and returning structured analysis that users can then - for lack of a better term - daisy chain to get on top of "hundreds or thousands of documents far more efficiently than traditional manual methods," as the press release puts it."
AI excels at analyzing and summarizing data but cannot handle terabytes of discovery due to limited context windows. Legal technology vendors are developing workarounds that process large batches and return structured outputs for further chaining. Thomson Reuters introduced a CoCounsel Legal beta feature that enables bulk review of 10,000 documents at once, producing sortable, user-friendly results. Human reviewers remain necessary to run batch searches and manage outputs, though fewer people are required compared with traditional manual review. Competitive advantage in legal tech will depend on designing practical solutions that mitigate AI's inherent context-window limitations.
Read at Above the Law
Unable to calculate read time
Collection
[
|
...
]