Shears et al v. First Bank

On February 5, 2026, a landmark class action was filed in the U.S. District Court for the Eastern District of Michigan. The plaintiffs, led by Detroit small-business owner Jane Shears, allege that First Bank’s proprietary AI-powered mortgage and credit underwriting system, known as “RiskWise,” systematically discriminates against African-American and Latino applicants. The complaint asserts that the algorithm relies on non-traditional data proxies—such as zip-code-level retail vacancy rates and social media activity—which correlate closely with race and result in disproportionate denials for equally creditworthy minority borrowers. Data scientists retained by the plaintiffs found that applicants from majority-minority neighborhoods were 40% more likely to be rejected than white counterparts with identical credit scores, revealing a modern form of redlining. The suit charges First Bank with violating the Equal Credit Opportunity Act and the Fair Housing Act, and seeks declaratory and injunctive relief, compensatory damages, and full algorithmic transparency.

As of June 2026, the case remains in active litigation. In April 2026, the district judge denied First Bank’s motion to dismiss, ruling that the statistical evidence of disparate impact was sufficient to survive pleading standards and that the plaintiffs had plausibly alleged that less discriminatory alternative models were available. The court also denied the bank’s motion to compel arbitration, finding the electronic agreement’s opt-out mechanism unconscionable. The class was preliminarily certified, and the parties are currently in discovery, with the plaintiffs seeking access to the RiskWise source code and training data.

The central dispute is whether an opaque, third-party-developed lending algorithm that produces racially disparate outcomes constitutes illegal discrimination under the Equal Credit Opportunity Act, even absent explicit discriminatory intent. The case tests the legal standard for “disparate impact” in automated decision-making, the discoverability of proprietary machine-learning models, and whether a bank can be held liable for biased outputs of a system it did not build but actively deployed.

The lawsuit has sent shockwaves through the financial and tech industries, accelerating calls for federal AI accountability legislation. It prompted the Consumer Financial Protection Bureau to issue an advisory bulletin on algorithmic fairness in credit, and multiple major banks voluntarily launched internal audits of their AI lending tools. The case has become a focal point for civil rights organizations and “responsible AI” advocates, highlighting the tension between financial innovation and anti-discrimination law.