Introduction
The Institutional Accountability Framework (IAF) evaluates whether responsibility and authority remain properly aligned when digital systems interact with minors.
Different groups encounter this question from different positions:
Parents experience outcomes without control
Schools hold responsibility for environments they do not fully govern
The sections below describe how the same structural conditions appear from each perspective.
For Parents
Parents are legally and practically responsible for their children’s wellbeing.
However, when children interact with digital systems - particularly on school-issued devices, parents often do not have:
Visibility into interactions
Ability to reconstruct what occurred
Ability to intervene in a timely and informed manner
What Has Changed
Historically, tools used by children were:
Observable (books, assignments, software outputs)
Reviewable after use
Controlled by adults at the point of interaction
With generative and interactive systems:
Outputs are generated in real time
Interactions are not consistently logged, retained or accessible
Experiences may differ between users and sessions
This reduces the ability of parents to verify or understand what occurred.
Structural Condition
This creates a separation:
Responsibility remains with parents
Control over interactions and outputs is partially delegated to systems and institutions
The framework identifies this as a misalignment between responsibility and authority.
What the Framework Evaluates
From a parent perspective, the framework asks:
Whether interactions be observed or reconstructed
Whether there is clear accountability for system behavior
Whether defined boundaries exist for when and how systems are used
Whether institutional backing has been fulfilled
These are governance questions, not content questions.
Questions Parents need to ask their schools today.
1. Insurance Review
For each defined student-facing generative or conversational AI use case, has the district’s insurer explicitly reviewed and confirmed coverage?
2. Logging / Reconstruction
Are all student prompts and corresponding AI outputs fully reconstructable by authorized administrators when generative AI tools are used on district-issued devices or district-managed student accounts, whether used at school or off campus?
3. Vendor Liability Allocation
If a vendor disclaims responsibility for AI-generated outputs, how is liability contractually allocated to protect the district, students, and taxpayers?
4. Default Device Posture
What is the default configuration of generative AI tools on district-issued student devices?
Are they (a) disabled by default, (b) disabled by default but enabled only during supervised instructional use, or (c) enabled for general student access?
If configurations differ by grade level, please indicate those differences.
Ask for answers in writing. Policy and IT responses do not typically answer governance questions. These questions require legal counsel / risk management review and response.
For Schools
Schools are responsible for the environments in which students learn and interact.
What Has Changed
Traditional educational tools:
Produce consistent, reviewable outputs
Operate within defined instructional boundaries
Are directly observable by educators
Are typically provided within vendor and institutional structures where responsibility for product performance is defined and understood
Student-facing generative systems:
Produce non-deterministic outputs
Operate through conversational interaction
Are not directly observable by educators during student interaction
Are often provided under vendor terms that disclaim responsibility for outputs and define use on an “as is” basis
Structural Condition
Schools may be responsible for outcomes arising from systems that:
They do not fully control
They cannot fully observe
They may not be able to reconstruct after use
This creates a governance gap between:
Institutional responsibility
Operational visibility and control
What the Framework Asks
From a school perspective, the framework asks:
Can the institution reconstruct interactions as needed?
Are responsibility and liability clearly defined and institutionally confirmed?
Are systems operating within bounded, observable, and supervised contexts?
These questions align with how schools evaluate risk in other domains.