Contribution to the handbook
Priority · Assessed 2026-10-27
This distinction materially affects whether human participation or review can meaningfully check an algorithmic decision. The evidence offers reasons not to treat a human presence or tool use as proof of understanding, while leaving the threshold unresolved.
Does this challenge the goal?
In-scope gap
The question distinguishes tool assistance from a person’s understanding, a relevant issue for citizen participation and meaningful oversight within the existing goal.
The supplied passages raise the distinction but do not establish how much understanding is necessary for effective participation or review.
Why this category?
- consequential safeguard limitation
- meaningful distinction for oversight
- testable uncertainty
The article’s standard is a normative argument, not a measured finding that all tool use prevents understanding; its examples do not settle whether some tools can support understanding.
This is an AI assessment of contribution to the goal, separate from whether the question has an answer.
- Does using a tool for part of a task mean that a person understands how it reached its result?
- A human presence may not be enough
- Understanding the reasons
- Using a tool is not understanding
- Human involvement may not be enough
- Better performance, less understanding
- Thinking with external tools
- Some systems are hard to explain
- Understanding may be the deeper problem
- Possible in principle, difficult in practice
- Participation and understanding matter
- A possible public decision risk
- Reasons for choices are unclear
Authority, accountability and resilience →
Tests, stop conditions and unresolved work →
Read the exploration goalDistinguish a tool’s contribution to a task from the user’s understanding of the process that produced its result.
What we can say so far
No. Danaher’s calculator example distinguishes a tool’s contribution from the user’s understanding: a person can rely on a tool to perform steps without consciously understanding those steps. The extended-mind view allows tools to participate in thinking, but that participation alone does not establish that the user understands how the tool produced a result.
What remains open
- The supplied material does not establish when, or how, using a tool might help a user understand the process that produced its result.
Conditions and limits
- This is an interpretation of Danaher’s manuscript, based on the full readable text of an accessible mirror rather than the publisher typeset version.
- The calculator example illustrates the distinction; it does not establish that users never understand tools or that the same account applies to every tool.
Objections to consider
- The extended-mind view complicates a simple separation between tool use and thinking, but the supplied passage itself notes that incorporating a tool into a mental process does not show conscious understanding of the tool’s reasoning.
Evidence and related ideas
Supporting evidence, qualifications and context play different roles in this answer.
supports
Using a tool is not understanding The author uses a calculator to show that relying on a tool does not necessarily mean consciously understanding the steps it performs.
Directly illustrates that relying on a tool’s contribution does not necessarily involve consciously understanding its steps.
The Threat of Algocracy: Reality, Resistance and Accommodation
Read full source: The Threat of Algocracy: Reality, Resistance and Accommodation (opens in a new tab) ↗Read source discussion: The Threat of Algocracy: Reality, Resistance and Accommodation →
qualifies
Thinking with external tools The extended mind view holds that thinking can include tools outside the brain. The author explains that this view draws on functionalism, the idea that mental states depend on their role in a network, and uses an email folder as a possible extension of memory.