Kent reports fewer homeless households after targeted support

AI system

System: Xantura AI-enabled predictive-analytics platform used for homelessness prevention
Model: Not assessed
Release: Not assessed
Application developer: Xantura
Role: Not assessed
Configuration: Not assessed
Inputs and outputs: Not assessed
Autonomy: Not assessed
Workflow: Maidstone combined local data with data shared by Kent County Council, including social-care information. The comparison group did not receive proactive support because the housing team lacked capacity.
Human oversight: Not assessed
Relation to AI: The presenter describes an AI-enabled platform that combines council data and flags households for human support. The prevention comparison concerns the combined targeting-and-support program. Because support was capacity-constrained and not randomized, AI’s separate contribution to the difference is not established.
AI contribution: Not assessed

Outcome identifiers: Not assessed
Benefit: Kent reports under 1% homelessness among proactively supported households versus 40% in an unsupported comparator. The groups were not randomized.
Category: Reduced Risk
Status: Reported benefit
Beneficiary groups: People. Organizations.
Beneficiaries: Not assessed
Reach: Not assessed
Distribution and access: Not assessed
Duration: Not assessed
Expected?: Yes
Costs and harms: Not assessed

Participants: Developer: Xantura. Operator: Maidstone Borough Council and Kent County Council with participating district and borough councils.
Jurisdictions: Benefit: United Kingdom.
Ownership and control: Not assessed
Dependencies: Not assessed
Cross-border connections: Not assessed
Geopolitical implications: Not assessed

Conditions for replication: The presenter states that protected characteristics are not included in the live prediction model. The source does not establish causal conditions for reproducing the prevention result.
Replication attempts: Not assessed
Later findings: Not assessed
Availability: Not assessed
Institutional decision: Not assessed
Grounds: Not assessed
Action and result: Not assessed
Success criterion: Not assessed

Assessed benefits: People. Organizations.
Depth: People: 4. Organizations: 2.
Reasons: People: The council reports that proactive support prevented homelessness in a high-risk cohort. The groups were not randomized, and AI’s separate effect is unknown. Organizations: The reported prevention program improves targeting and preventive results for councils. The source does not supply sample sizes or an independent evaluation.
Observed or intended: Reported benefit
Rubric: Not assessed
Assessor and date: Not assessed
Alternative assessments: Not assessed
Fill: The deployer's account

Kind: Not assessed
Related entries: Not assessed
Field: Housing and homelessness prevention.
Purpose: Identify households at risk of homelessness so that councils can offer proactive support.
Dates: Source checked: 22 October 2026. Updated: 22 October 2026.
Stage: Not assessed
Version: SHA-256: ee4a7830fdaf…