Benchmark Almanac

KORA Apps

MagicSchool application 74% tops its leaderboard
Not in model cards

Simulated children use consumer AI apps end to end, with each app system prompt, filters and safeguards active. Scores cover online, physical and sexual, psycho-social and learning risks.

Rates apps, not bare models, in child and adult modes. First board dated July 26th, 2026 lists about 20 app entries such as MagicSchool, ChatGPT and SchoolAI.

Released
July 2026
Kind
Wellbeing and sycophancy
Level
Everyday
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
12
Entries on its own leaderboard, 0 models we track

MagicSchool application tops the leaderboard with 74%

Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

  1. 74.0 MagicSchool application
  2. 59.0 ChatGPT application
  3. 57.0 Claude application
  4. 57.0 SchoolAI application
  5. 56.0 Khanmigo application
  6. 55.0 Meta AI application
  7. 53.0 Copilot application
  8. 46.0 TikTok application
  9. 45.0 Gemini application
  10. 42.0 Snapchat My AI application
  11. 27.0 Character.AI application
  12. 27.0 PolyBuzz application
  1. 1 MagicSchool application Child mode 74%
  2. 2 ChatGPT application Child mode 59%
  3. 3 Claude application Adult mode 57%
  4. 4 SchoolAI application Child mode 57%
  5. 5 Khanmigo application Child mode 56%
  6. 6 Meta AI application Child mode 55%
  7. 7 Copilot application Child mode 53%
  8. 8 TikTok application Child mode 46%
  9. 9 Gemini application Child mode 45%
  10. 10 Snapchat My AI application Child mode 42%
  11. 11 Character.AI application Adult mode 27%
  12. 12 PolyBuzz application Adult mode 27%

Source: Official KORA Apps leaderboard

#EntryScoreSettingDateSource
1 MagicSchool application 74% Child mode Official leaderboard
2 ChatGPT application 59% Child mode Official leaderboard
3 Claude application 57% Adult mode Official leaderboard
4 SchoolAI application 57% Child mode Official leaderboard
5 Khanmigo application 56% Child mode Official leaderboard
6 Meta AI application 55% Child mode Official leaderboard
7 Copilot application 53% Child mode Official leaderboard
8 TikTok application 46% Child mode Official leaderboard
9 Gemini application 45% Child mode Official leaderboard
10 Snapchat My AI application 42% Child mode Official leaderboard
11 Character.AI application 27% Adult mode Official leaderboard
12 PolyBuzz application 27% Adult mode Official leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Safety and alignment

All 465 in Safety and alignment
  • FORTRESS In use Paired adversarial and benign national-security misuse prompts. Rubric-graded risk score on adversarial and over-refusal score on benign prompts. Claude Opus 5.5 5.78%
  • StrongREJECT In use Forbidden-prompt jailbreak benchmark with a rubric grader that scores how useful a non-refusal actually is. Inkling 98.6%
  • Gray Swan IPI Arena benchmark In use Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts. Gemini 4 Carbon 0%
  • DistressBench In use Models reply as a chatbot to 718 simulated users in psychological crisis, in single-turn and multi-turn chats. Three judge models score replies against clinician-written rubrics. Muse Spark 1.2 88.4%
  • Lakera B3 (Backbone Breaker Benchmark) In use Crowdsourced prompt-injection attacks hit backbone LLMs inside agent threat snapshots at three defense levels. The score is attack resistance, 100 minus the board's vulnerability percentage. Claude Opus 4.5 89%
  • IHEval In use Models face conflicts between system, user and tool instructions. Reports a composite score for following the higher-priority instruction. GPT-4o 70%