RFI response corpus: 178 submissions on AI-powered mathematics innovations collected April–May 2024¶

Element · All elements
Evidence · 6 claims (6 for) · 6 studies (5 qualitative, 1 associational), q2 · 0 of 6 report an effect size · 6 claims rest on one study

Description¶

A corpus of 178 responses to a 21-question Qualtrics RFI launched by Digital Promise and the Bill & Melinda Gates Foundation in April 2024, closing May 20. Submissions described innovative approaches to AI in mathematics teaching and learning; 169 targeted US PK-12. Respondents were mostly for-profit edtech companies (45.0%); 37.1% were early concepts, 35.4% in prototyping/field testing, and 27.5% on market. A panel of 26 reviewers evaluated research-based submissions aligned to field priorities.

Design Implications¶

Context¶

Requirements¶

• Access to the RFI responses shared with funders, which most respondents permitted

Constraints¶

• The report states responses came from innovators in networks connected to Digital Promise and the foundation and "are not necessarily representative of all AI innovations"

Target Learners¶

• PK-12 mathematics students
• Teachers and instructional coaches

Target Learning Goals¶

• Understanding the landscape of AI innovations in mathematics teaching and learning

Claims¶

• Respondents identified algorithmic bias (31.9%), data privacy (27.5%), and technical reliability (27.5%) as the primary risks of AI in education [+W]
• Respondents proposed testing and monitoring (61.9%) and human involvement (22.7%) as leading mitigations for algorithmic bias [+W]
• Fewer than five percent of submissions described AI as opening new opportunities for inclusivity, a gap the report highlights [+W]
• Data privacy mitigations centered on data governance and monitoring (68.9%), while technical reliability mitigations centered on human involvement (36.9%) [+W]
• Personalizing learning (22%) and teacher assistance (25%) dominated project topics, with only 27.5% of solutions on market [+W]
• Respondents anticipated using LLMs most often (23%), followed by NLP (17%), predictive AI (14%), chatbots (13%), and ASR (10%) [+W]

Related Elements¶

Examples¶

-

Key Sources¶

• Noakes, S., Shell, A., Van Nostrand, P., Murillo, A., Ruiz, P., and Liberman, B. (2024, September). AI-Powered Innovations in Mathematics Teaching & Learning: Initial Findings. Digital Promise. https://doi.org/10.51388/20.500.12265/229