Employers increasingly use automated tools for sourcing, screening, assessment and performance management. The legal framework is developing quickly and the underlying principles are not new.

Disparate impact. A tool that screens out protected groups at a disproportionate rate must be job-related and consistent with business necessity. That the vendor built it, and that the employer does not know how it works, is not a defense.

Validation. Selection procedures should be validated against job performance under recognized methodologies, and the employer should obtain and review the vendor’s adverse impact analysis rather than accept an assurance.

Disability. Tools measuring reaction time, speech patterns or facial expression may screen out individuals with disabilities and require an accessible alternative and a clear route to request accommodation.

Local requirements. Some jurisdictions require bias audits of automated employment decision tools, published summaries, and advance notice to candidates. Others require disclosure of the use of the tool and of the characteristics assessed.

Privacy and biometrics. Statutes governing biometric identifiers impose consent, retention and destruction requirements with statutory damages.

Practical controls. Contractual audit rights and indemnity from the vendor; human review of adverse decisions; documented business justification; and retention of the inputs and outputs, which will be the evidence.