What AI Cannot Do
🟥 AI Cannot DoOther

AI cannot pick up your child from kindergarten

AI can help parents plan routes, remember pickup time, contact ride-hailing services, or create schedules, but it still cannot act as a responsible guardian or delegated robot that independently picks up a kindergarten child and brings them home safely.

Creator: Lin Qi2026-06-23childcare, kindergarten, robotics, physical-world, safety, handoff
Challenge this
Why is this considered beyond AI?

Background

Many white-collar families deal with kindergarten pickup every day. The task may look like moving a child from point A to point B, but the real workflow involves identity verification, teacher handoff, emotional care, road-safety judgment, exception handling, responsibility, and trust between the family and the school.

Goal

Have AI or an AI-controlled robot, without an adult present, complete school-approved pickup, identity handoff, trip supervision, exception handling, and safe delivery of the child back home.

Observed problems

The core difficulty is not route planning; it is responsibility. Kindergartens usually hand children to parents or authorized adults. A robot cannot yet carry guardian responsibility, communicate situational judgment, handle crying, injury, getting lost, stranger approach, or traffic risk, and earn shared trust from both school and family.

Failure analysis

Current AI lacks an offline guardian identity that society and schools accept. Even if a robot can navigate, recognize a child, and avoid obstacles, it still cannot fully carry child-safety responsibility, legal accountability, ethical judgment, and complex human communication. As long as the handoff requires an identifiable, accountable adult who can make immediate judgments, AI cannot independently complete this task.

Success criteria

A successful case must be a firsthand real pickup experiment, not a concept video or media report. It must provide kindergarten/parent authorization boundaries, robot or AI system information, the full pickup workflow, at least 5 independent pickup records, on-site screenshots or video, exception-handling logs, and evidence that the school recognizes the handoff method. Child-safety requirements must not be reduced at any point.

Attempt submission standard
  • The solution must be personally tested and completed by the submitter; media reports, reposted articles, and second-hand cases are not accepted.
  • Submissions must include screenshots, video, repository links, logs, or accessible proof. Screenshots should come from actual running results.
  • Submissions must state the AI product, model version, model settings, experiment count, and full process.
  • If claiming success, provide a step-by-step reproduction or operation guide so reviewers can verify it.
  • Submissions that only cite news, product marketing, or someone else's result will be rejected.
AI tool attempt records

No reviewed attempts yet.

Timeline
  1. Challenge created

    2026-06-23

    Recorded AI's current inability to independently complete kindergarten child pickup as a responsible adult or delegated robot.