AI Landlord Screening Tool Ends Discriminatory Practices After Settlement

SafeRent AI Screening Tool Settlement: A Landmark Case Against Discriminatory Practices

SafeRent, a leading AI screening tool employed by landlords to evaluate potential tenants, has reached a significant settlement to reform its practices concerning applicants utilizing housing vouchers. The U.S. District Judge Angel Kelley has granted final approval for a settlement amounting to approximately $2.3 million, aimed at combatting discrimination against tenants based on income and race.

The Background of the Case

This landmark settlement arises from a class-action lawsuit filed in Massachusetts in 2022. The lawsuit asserted that SafeRent's scoring system disproportionately affected individuals using housing vouchers, particularly harming Black and Hispanic applicants. Allegations included violations of Massachusetts laws and the Fair Housing Act, which is designed to prevent housing discrimination.

The Controversial SafeRent Scoring System

According to the initial lawsuit, SafeRent's algorithm assigns potential tenants a SafeRent Score, using various factors such as credit histories and debts that are unrelated to rental history. This score influences landlords' decisions on rental applications. However, critics highlighted the opaque nature of this scoring system, as landlords often lacked insight into how scores were calculated.

Impact on Minority Groups

The lawsuit revealed that SafeRent's scoring system tended to assign lower scores to Black and Hispanic tenants as well as those relying on housing vouchers, leading to unfair denials of housing applications. This pattern raised serious questions about the transparency and fairness of algorithm-driven tenant evaluations.

Key Changes Mandated by the Settlement

Under the terms of the five-year settlement agreement, significant changes will occur regarding how SafeRent operates:

• No SafeRent Scores will be displayed for applicants using housing vouchers.
• Landlords will no longer receive recommendations to accept or deny applications based solely on SafeRent Scores.
• Rental applications from housing voucher users will now be evaluated based on the entire applicant’s record, rather than a single score.

Statements from Advocacy Groups

Shennan Kavanagh, the director of the National Consumer Law Center, states that "credit scores and scores modeled similarly, such as SafeRent Scores, draw on information that has only been tested at predicting repayment of credit obligations. There is no evidence such data is predictive of tenants paying rent." This sentiment echoes the growing concern over the reliance on AI algorithms in housing and its implications for fair housing practices.

Distribution of Settlement Funds

The funds raised from the settlement will be allocated to rental applicants in Massachusetts who faced housing challenges due to the biases in SafeRent's tenant scoring system. SafeRent, while maintaining that its scoring system complies with laws, acknowledged that extended litigation would detract from its core mission of providing landlords with essential tenant screening tools.

The Bigger Picture

The SafeRent case highlights a broader trend of algorithm-driven property management tools facing scrutiny and legal challenges regarding their impact on housing equity. In August, the Department of Justice took action against RealPage, claiming its algorithmic pricing software contributed to inflated rents.

Conclusion

As the conversation surrounding technology and fair housing continues, the SafeRent settlement serves as a crucial reminder of the potential for systemic bias in AI systems and the need for transparency in tenant screening processes. This legal decision marks a significant step towards fostering a fairer housing environment for all individuals, especially those utilizing housing vouchers.