Policy & Practice | Fall 2025

where agencies struggle to address challenges proactively. n Moreover, the CFSR’s reliance on traditional data-collection methods poses further limita tions. Manual input and traditional analysis techniques are prone to errors, biases, and inefficiencies, affecting the accuracy and reliability of review outcomes. The com plexity and time-consuming nature of manual analysis can delay the identification of critical trends and emerging issues. These limitations can hinder child welfare agencies from fully understanding their per formance and determining effective improvement strategies. The CFSR’s periodic nature, complexity, and unrepresentative sampling methods can hinder the effectiveness and timely improve ments of state programs. Facilitating CFSRs Through AI and Machine Learning Integrating artificial intelligence (AI) and machine learning into state processes offers a promising solution to many of the challenges faced during the CFSR. AI can analyze vast amounts of data quickly and accu rately, reducing the time, manpower, and financial investment required for comprehensive evaluations. This alleviates resource constraints smaller agencies faced, allowing them to participate effectively in the review process without diverting attention from direct service delivery. By employing advanced algorithms, states can enhance the accuracy and reliability of review outcomes, cap turing the nuanced realities of child welfare practice more effectively. Continuous monitoring facilitated by machine learning allows for real-time evaluation and timely identification of ongoing issues and trends, positioning agencies to implement improve ments swiftly, benefiting children and families without delay while fostering a culture of continuous improvement. Leaving behind the traditional and manual approach to data collection and analysis and adopting a com prehensive and automated approach empowers agencies to identify areas

for improvement, implement effective interventions, and ultimately improve outcomes.

tools serve as an “extra set of eyes” to provide valuable insights, questions, and opportunities for improvement. By reviewing not only case contacts, but also assessment tools, court nar ratives, and provider documents, it generates feedback to workers while they are engaged with families, facilitating reflection and action to improve performance at both indi vidual case and aggregate levels. n Timeline View for Comprehensive Content Review: A timeline view allows workers or supervisors to review case content by date, including AI summaries of case contacts and analyses of safety, permanency, and well-being. Users can filter by content type, choosing to view specific contacts, such as those with the child or other participants, face to-face contacts in the home, or office meetings with the mother. In the same screen, users see a summary of the contact narrative, including par ticipants, assessments, and the date, time, and location of the contact. By providing real-time feedback, advanced case review tools, and AI-driven insights, RedMane’s Expert Practice Advisor helps workers engage more effectively with families, making informed decisions that promote safety, permanency, and well-being. This innovative approach not only supports continuous improvement within the workforce but also facili tates the delivery of timely, responsive, and high-quality services. A Call to Action It's time for policymakers, child welfare agencies, and stakeholders to unite in the adoption and integration of AI and machine learning technologies. By investing in these advancements, we will have a transformative impact on child welfare services, helping every child receive the care and support they deserve. Let us take decisive action now to pave the way for a future where technology drives excellence in child welfare, benefiting children, families, and communities nationwide. Paige Rosemond , MSW, is the Director of Innovation at RedMane Technology.

A Model Solution: RedMane’s Expert Practice Advisor

RedMane Technology’s Innovation Team, a group combining decades of experience in both human services and software development, is deter mined to harness the power and promise of AI and machine learning to answer the challenges of the CFSR. The Expert Practice Advisor is a tool designed to enhance states’ ability to assess and address child welfare practice and performance by offering the following functionality: n Analyzing Every Case Contact: RedMane’s Expert Practice Advisor analyzes each case contact to determine if safety, permanency, and child well-being—the founda tional components of child welfare practice—have been assessed. These determinations are supported by detailed explanations based on policy and practice guidelines. The tool aligns its analysis with the ACF’s first CFSR framework from 2001. n Building Trust Through Transparent Analysis: The Expert Practice Advisor increases trust among workers, supervisors, and leadership by providing clear expla nations for its assessments. This transparency helps stakeholders understand the rationale behind each evaluation, fostering confi dence in AI-driven reviews and supporting their adoption. n Providing Real-Time Feedback for Continuous Improvement: The Expert Practice Advisor delivers immediate feedback on every contact, allowing staff throughout the hierarchy to review casework promptly, incorporate findings into future contacts, and improve practice continuously. With this dynamic feedback loop, there is no longer a need to wait and react to court, citizen review, departmental, or federal case review processes. n Offering Enhanced Case Review Tools: RedMane’s Expert Practice Advisor uses machine learning and AI to review practices in real time. These

7

Fall 2025 Policy & Practice