evidoria

← Back to browse

Good practice Imported

Estudo Play AI Essay Grading — Projeto Enem MG Sets Guinness World Record, Minas Gerais, Brazil

Brazil · Belo Horizonte · See the Brazil profile · See the Belo Horizonte profile

Evidence: Descriptive / self-reported Top 73% 40/100 · Ask Evidence Copilot about this practice

Brazilian edtech Estudo Play set a Guinness World Record by having its AI grade over 461,000 handwritten student essays in 30 days for Minas Gerais's public-school Projeto Enem MG, part of roughly 5 million essays the platform processed across Brazil in 2025.

461,000+
Essays graded in 30-day window (Sep-Oct 2025)
~5,000,000
Essays processed nationwide, full 2025 (2025)

Details

Maturity
Scaling
Promoter
Estudo Play; Secretaria de Estado de Educação de Minas Gerais (SEE-MG)
Period
September–October 2025 (Guinness record window); ongoing through 2025
Keywords
K-12, assessment, essay grading, public education

Context

Estudo Play is a Brazilian edtech company whose platform digitises handwritten student essays and scores them against Brazil's national ENEM university-entrance exam criteria. Under Projeto Enem MG, a partnership with the Minas Gerais state public-school network, the platform's AI graded essays at national public-exam scale.

Results

The platform's AI graded more than 461,000 handwritten essays in a single 30-day window during September-October 2025, earning a Guinness World Records title for the most essays corrected by artificial intelligence in one month. Across 2025 as a whole, the company reports processing roughly 5 million essays for public and private students nationwide.

Conclusions

Coverage confirms throughput and the independently verified Guinness record, but the available press does not report inter-rater agreement against human graders, score-validity studies, or student learning-outcome data, and no data-protection or algorithmic-transparency documentation was found for the Minas Gerais deployment.

Implementation
Indicative cost
Low (< €50k)
Time to results
Short (< 1 year)
Staffing & skills
Estudo Play platform/AI engineering team

Conditions for success

  • Partnership with a state public-school network for at-scale deployment
  • Automated digitisation of handwritten essays reducing manual grading burden

Common failure modes

  • No inter-rater agreement, validity studies, or learning-outcome data published
  • No data-protection or algorithmic-transparency documentation found

Commonly funded by

Own resources / municipal budget

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Similar practices you may find useful