Contribution to the handbook
Supporting · Assessed 2026-10-28
The comparison helps distinguish ways of augmenting preference data, and the evidence separates the reported performance from the authors’ proposed explanation. This is useful background for considering alternatives, but by itself does not establish a governance design or general effectiveness.
Does this challenge the goal?
In-scope gap
The question compares alternative preference-estimation methods within the accepted goal; it does not challenge the goal or beneficiary boundary.
The reported comparison is study-specific, and the offered explanation is explicitly conjectural.
Why this category?
- meaningful alternatives
- consequential limits
The packet gives no basis for treating the authors’ proposed explanation as an established cause or for generalizing the comparison beyond the study.
This is an AI assessment of contribution to the goal, separate from whether the question has an answer.
Making and interpreting laws →
Tests, stop conditions and unresolved work →
Read the exploration goalNew evidence is being assessed. Any answer below is provisional; parts whose evidence changed or was withdrawn are withheld.
Compare the reported Type II and Type I B results and state the explanation the authors offered, distinguishing that explanation from an established cause.
What we can say so far
In the reported tests using all demographic information, the authors found that Type II A and Type II B augmentation performed better than Type I B. They suggested that Type II may benefit from using participants’ recorded preferences as well as demographic information, since it predicts additional preferences for people already in the sample. That is the authors’ proposed explanation, not an established cause. The supplied evidence gives no numerical size for the Type II–Type I B advantage.
What remains open
- The supplied evidence does not quantify the Type II advantage over Type I B.
- The evidence does not establish why Type II performed better.
Conditions and limits
- The reported comparison is for tests using all demographic information; it is not evidence of effectiveness in other settings.
- The explanation about recorded preferences is conjectural, not a separately established causal finding.
Evidence and related ideas
Supporting evidence, qualifications and context play different roles in this answer.
supports
Recorded preferences may help In tests with GPT-3.5 Turbo, the authors report that Type II A and B augmentation performed better than Type I B augmentation. The Type II tests used all demographic information and predicted additional preferences for people already in the sample. The authors suggest that these tests may benefit from using participants' recorded preferences as well as demographic details.
Directly reports the Type II A and B versus Type I B comparison and the authors’ tentative explanation.
Large Language Models (LLMs) as Agents for Augmented Democracy
Read full source: Large Language Models (LLMs) as Agents for Augmented Democracy (opens in a new tab) ↗Read source discussion: Large Language Models (LLMs) as Agents for Augmented Democracy →
context
Ways to add preference data The authors describe several ways to add predicted preferences. Type I predicts preferences for additional people using information such as demographics. Type II A predicts more preferences for people already in the sample about existing proposals; Type II B predicts their preferences about new proposals. Type II C combines those two Type II approaches. Type III adds predictions for both new people and new proposals.