Judicial Law Clerks
The core of this work stays human, for now.
Most of this occupation's activities don't appear in measured AI usage. The paperwork fringe can be accelerated, but the core of the work stays human for now.
Where this occupation sits among all 785 measured. Bands are quintiles of the real distribution. Applicability measures task overlap with observed AI usage. It is not a probability of job loss.
What this job involves
Assist judges in court or by conducting research or preparing legal documents.
Where AI overlaps this work, activity by activity
The occupation's most important work activities (O*NET weights, normalized to this set), each with its measured AI performance across 200,000 real conversations: how often AI is used for it, how well it completes it, and how much of the activity it covers.
| Work activity | Share of role | AI performance |
|---|---|---|
| Record information about legal matters. | 25.5% | Not observed 0.00 |
| Prepare legal or regulatory documents. | 20.9% | Not observed 0.00 |
| Discuss legal matters with clients, disputants, or legal professionals or staff. | 19.5% | Not observed 0.00 |
| Consult legal materials or public records. | 18.2% | Emerging 0.38 |
| Research laws, precedents, or other legal data. | 8.6% | Moderate 0.54 |
| Direct legal activities. | 5.7% | Not observed 0.00 |
| Supervise personnel activities. | 0.8% | Not observed 0.00 |
| Perform court-related or other legal administrative activities. | 0.7% | Not observed 0.00 |
"Not observed" is a measurement, not missing data: that activity does not meaningfully appear in the usage sample. AI performance = completion × scope × coverage (user-goal view), 0–1.
What drives this score: the biggest AI-overlapped activity in this role is research laws, precedents, or other legal data; the most important activity AI is not observed doing is record information about legal matters. The plan below is built around exactly that split.
Protect the core, automate the fringe.
Next 30 days
- Automate the fringe. The small text-and-paperwork layer of this job, Research laws, precedents, or other legal data, is worth handing to AI now; it's the least valuable part of your week anyway.
- Confirm your reality. If your actual role involves far more screen work than the occupation average, the personalization above will say so.
Next 90 days
- Convert resilience into rate. Work like Record information about legal matters is measured as beyond current AI: supply of it doesn't scale with software. Skill up and price accordingly.
- Let AI carry your admin. Quotes, scheduling, follow-ups, records: the overhead around the core work is where the tools pay off for you.
The longer game
- Watch robotics, not chatbots. For low-exposure occupations the relevant automation frontier is usually physical or regulatory, which this dataset deliberately doesn't measure. Track your industry, not AI headlines.
- Build the customer asset. Reputation and repeat relationships are the moat generative AI cannot cross. Invest there.
- AI applicability score and per-activity metrics: Tomlinson, Jaffe, Wang, Counts & Suri, Working with AI: Measuring the Applicability of Generative AI to Occupations, Microsoft Research 2025 (arXiv:2507.07935), data CC BY 4.0.
- Occupation description and activity structure: O*NET 30.0 Database, USDOL/ETA, CC BY 4.0.
- Bands, personalization weights and action plans are JobRiskAI editorial layers; method and limitations on the methodology page.