LIT Logo

Legal Innovation & Technology Lab
@ Suffolk Law School

This demonstration page is built on the Spot API, a project bringing free AI-powered issue spotting to legal aids and non-profits. You can improve it by playing Learned Hands or sharing relevant data. This page is presented for demonstration purposes only. The results presented here should NOT be relied on in isolation. To learn more about Spot's coverage, read up on its current performance, or take it for a spin below.

Managing Uncertainty: When Spot returns a list of issues, it provides a prediction accompanied by a lower and upper bound. See Documentation. The bounds are intended to provide users with information about the uncertainty for a given prediction, allowing them to tailor their results to their use case and tolerance for error. The larger the distance between the lower bound and the prediction, the more likely it is that the prediction is overly optimistic about finding the issue. Conversely, the larger the difference between the prediction and the upper bound, the more likely it is that the prediction is under-estimating the chance that the issue is present. Users are encouraged to consider this when setting the cutoffs for triggering the return of an issue. For example, a user interested in avoiding false negatives may set Spot to return results based on the upper bound being greater than 50%, while one interested in avoiding false positives might choose to base their cutoff on the lower bound. The distance between the lower and upper bounds communicates something about the overall uncertainty around a prediction, with the most certain predictions being those with the smallest distances between the two.

Normally a developer making use of Spot would preset cutoffs for the prediction and its upper and lower bounds, but here you can set them yourself to get a sense for how they work to filter results. Note: filtering is applied while traversing the LIST hierarchy. Consequently, Spot will not get to a child issue if its parent returns predictions outside of the cutoffs, even if the child's predictions would have fallen within them.

Lower Bound Cutoff: ?
Prediction Cutoff: ?
Upper Bound Cutoff: ?

There are several live implementations of Spot which either focus on issues well covered by Spot or supplement Spot with additional context. See e.g., Court Forms Online (MA); Massachusetts Legal Resource Finder (MA); CT Law Help (CT); Illinois Legal Aid Online (IL); Pine Tree Legal Assistance Triage (ME); and Reddit Eviction help bot (national).

Remember, live integrations are, and should, supplement Spot's results with additional information, like traditional content search, or focus only on areas of high performance. Seed reply below with sample text.

What seems to be the problem?

Longer replies, a sentence or two, tend to get the best results.   0 words

Where did this happen/is this taking place?