Terms of use

What CommunitySafe is

CommunitySafe is a neighborhood-level safety-awareness tool that surfaces publicly published police-incident data for 30 supported US cities, compared to the FBI Crime in the Nation national average. It is informational and educational.

What CommunitySafe is NOT

• Not a substitute for 911 or any emergency service. In an emergency, call 911. CommunitySafe does not contact emergency services on your behalf.
• Not a substitute for professional safety, legal, medical, or financial advice.
• Not predictive of future crime. All scores reflect historical reporting only — what has already been published in the city's own open-data feed.
• Not a surveillance product. Individual people are never identified, tracked, or geolocated.

Permitted uses

Personal use to inform travel or routine planning, awareness of recent activity in your area, and orientation when moving between cities.

Prohibited uses

You agree not to use CommunitySafe — or any data exported from it — as the sole or primary basis for:

• Housing decisions (renting to, leasing to, denying housing to a person).
• Lending or insurance underwriting decisions.
• Hiring or employment decisions.
• Any decision regulated by the Fair Housing Act, Equal Credit Opportunity Act, or similar laws that prohibit discrimination based on protected characteristics or geographic proxies for them.
• Confronting, following, filming, or approaching any individual.
• Vigilante action of any kind.

You also agree not to scrape the app to re-host its scoring, attempt to identify individuals from aggregated data, or use the app in any way that violates the terms of the underlying open-data feeds (each city's open-data portal has its own terms — typically permissive for non-commercial use with attribution).

Acceptable use & rate limits

To keep the app responsive for everyone and to protect the upstream open-data feeds from being saturated, public API endpoints are rate-limited per IP — typical limits are 30-60 requests per minute per endpoint family, with the AI Assistant capped lower (10/min) because each call has a real per-token cost. Sustained automated access beyond these limits will receive HTTP 429 responses with a Retry-After header.

You agree not to attempt to circumvent the rate limits, generate synthetic load against the app, scrape the per-neighborhood pages at high volume, or otherwise impair the service's availability for other users.

AI Assistant

The optional AI Assistant uses a third-party large language model to answer questions about CommunitySafe's data. By using it you acknowledge that:

• Outputs are generated by a probabilistic model and may be inaccurate, incomplete, or out of date. Every numeric claim the assistant makes includes a source URL — verify against that source before acting on it.
• Outputs are not professional advice (legal, medical, security, or otherwise) and must not be treated as such.
• The text of your prompt is transmitted to the AI provider for processing. See the Privacy policy for what travels and how.
• The assistant is not available in regions where it would violate local law.

Community posts & moderation

Posts you submit through the CommunitySafe tab are public by design and persist after you close the app. An automated pre-vetter blocks posts containing profanity or slurs, threats of violence, addresses below the block level, vehicle plates, phone numbers, named individuals, or descriptions that profile people by appearance. Posts that survive vetting may still be reported by other users and reviewed.

Repeated rejected posts, repeated upheld reports against you, or any abuse of the moderation system itself can result in a temporary suspension or, for severe / repeated violations, a permanent ban from posting. You agree your posts are factual to the best of your knowledge and that you have the right to share them. Edits to your own posts are recorded in an append-only history so the moderation context is preserved.

Data accuracy

CommunitySafe re-displays data from official city open-data portals without editorial modification. We rely on those publishers to be accurate and current. Police-incident data is reported by officers; it reflects what was reported, not necessarily what occurred. Publication delays can be 7-30 days depending on the city. The FBI national-rate comparison comes from the FBI Crime Data Explorer (cde.ucr.cjis.gov), pulled as the annual sum of monthly UCR rates for the most-recent complete calendar year. For full methodology, see the Methodology page.

No warranty

CommunitySafe is provided "as is" without warranty of any kind. We do not warrant that the data is current, complete, or fit for any particular purpose. To the extent permitted by applicable law, CommunitySafe and its operators disclaim liability for any decision made in reliance on data displayed by the app.

Changes

These terms may be updated as the app evolves. Material changes will be noted by the "Last updated" date at the top of this page.