A starting point for discussion
The idea
The author reports that earlier research found algorithmic systems sometimes made better predictions than human experts. The author does not claim this was true in every field or show that tax enforcement was one of those fields.
Place in the reasoning
AI assessment · Connected to reasoning
This reports earlier research finding algorithmic prediction advantages in some fields, while explicitly not generalizing to every field or to tax enforcement.
Citizen authority does not settle AI’s role
It illustrates why prediction performance may motivate delegation, while limiting the evidence to some fields and not establishing which decisions AI should make.
Adapting to computer systems or resisting them can both carry risks
Possible performance benefits bear on trade-offs about resistance, but the claim is limited and does not resolve the trade-off.
What this means and why it matters
What this means
This is a report of research cited by the author, not a result from the author’s own study. The claim is limited to certain kinds of predictions.
Why it matters
It may support using algorithmic systems for some decisions. But it does not show that they predict better everywhere or that they should control decisions.
Conditions & assumptions
Source coverage: The full text of the research paper written by the author, read from a copy hosted on another site; not the final version formatted by the publisher. The passage cites earlier work for the claim that algorithms outperform human experts on predictions in certain fields. It uses this as support for possible deference in some cases. Addresses: The scope of reported prediction findings
Questions this connects to
AI-suggested relationships. Authors have not necessarily responded to one another.