BRENNEN

Industries / Government, education and community

Education

The education industry teaches people in schools, universities and job training, and makes the lessons, tests, books and systems that teaching runs on.

About 3200 BCE

Writing began in the late Uruk period in what is now southern Iraq, and the first scribes compiled standard lists of words and signs that, in the account of the cuneiform scholar Niek Veldhuis, served to standardize and to teach the new script.

oracc.museum.upenn.edu

42

Milestones achieved across 6 domains.

8

To watch for: announced, not yet confirmed outside.

12

Open deliverables. Nobody has shipped them.

In one minute

Where it started

Writing began in southern Mesopotamia around 3200 BCE, and its first users copied fixed word lists to learn it. By about 2000 to 1500 BCE, scribes were trained in the "tablet house", often a private home, where pupils practiced wedge shapes, then word lists, then whole texts from memory. Schooling for everyone came much later: Prussia ordered children into school in 1763, and Massachusetts passed the first US state attendance law in 1852.

Where it stands

UNESCO estimates that 272 million children and young people were out of school in 2023. In low- and middle-income countries, 70% of 10-year-olds could not read and understand a simple text in 2022, up from 57% in 2019, by the World Bank's estimate. UNESCO says the world needs 44 million more primary and secondary teachers by 2030. What its authors call the largest trial yet of an AI tutor in public schools, run by outside researchers in Tennessee, found gains no larger than the same practice software gave without AI, mainly because students rarely used the tutor.

What is open

Much checking is still done by hand: exam papers before they are printed, textbook answer keys, apprentices' practice walls. Some students still wait for basic access, such as blind students in classes taught on a board, and children whose early reading is tested by ear in languages such as Hausa, for which this research found no automatic scorer. AI models that read handwriting, speech recognition that covers more than a thousand languages, and agents that read and compare long documents now put these jobs within reach of a small team.

Domains

AchievedWatch forOpen

Other narrower industries

EdTechDriving schoolsSchoolsUniversitiesTrade schools and vocational trainingCommunity collegesHomeschooling

Figureheads

6 people

Open in education

3 of 12 open deliverables

Open · Classroom teaching

Read the board aloud

A camera at the back of a classroom that reads what a teacher writes on the board, including math, and sends it to a blind student's refreshable braille display or earphones within seconds, with a way to move back through earlier lines. Point the camera at the board only, so no students are filmed.

Built with what exists

A webcam or phone on a tripod; a vision-language model that reads handwriting and math, such as Gemini or Claude; MathCAT, an open-source library that turns math into speech and braille codes such as Nemeth and UEB, and is used by the NVDA and JAWS screen readers; a refreshable braille display.

Credit · unclaimed

Open · Tutoring and practice

Audit AI tutors' math

A public, repeatable test of AI tutors as students meet them. A script plays a student working through released state test questions, sometimes giving right answers and sometimes common wrong ones, and logs every time the tutor accepts a wrong answer, rejects a right one or states false math. Two teachers check each flagged exchange. Run it through each maker's programming interface or with the maker's permission.

Built with what exists

Released test questions that several US states publish each year; a web agent to drive tutors that have no programming interface (see the Note "Deleting 80% to start over"); SymPy, an open-source math library, to check answers; a shared spreadsheet for the teachers' checks.

Credit · unclaimed

Open · Testing and grading

Proofread the exam paper

A checker that works through an exam paper and its mark scheme before the exam is sat. It solves each question independently several times and flags missing information, impossible or ambiguous questions, wrong mark-scheme answers, and differences between the standard paper and its braille or large-print versions.

Built with what exists

An AI model that works through math and science, such as Claude Fable 5.2 or Gemini, with code execution to recheck calculations (the Note "Seven more flagged" covers a Claude-based reviewer of long documents); liblouis, the open-source braille translator, to turn braille papers back into print for comparison; the past papers and mark schemes that exam boards publish.

Credit · unclaimed