Construct · 1 fields · 1 citations

alignment tax: a standalone construct

One name. One construct. A landmark on the map.

● anchor · a standalone construct Nearest construct by meaning·not indexed
alignmenttax1 CONSTRUCTAlignment taxASKELL · AI SAFETY

A single standalone construct; nearby constructs are located against it in the Collider. Tap a field for its definition, unit, and measures.

AI safety

any loss of capability or extra cost incurred by making a model aligned with human intentions, compared with an unaligned model.

In AI safety, alignment tax refers to any loss of capability or extra cost incurred by making a model aligned with human intentions, compared with an unaligned model. This meaning comes from Askell (2021) in arXiv.[1]

Unit
Recovery means
Measured by

Next

Run it through the Collider, or take the citations with you.

See how alignment tax sits against neighbouring constructs in the Collider, or export every reference on this page in one click.

Open in ColliderExport BibTeXReport this entry

Nearby constructs

helpful, honest and harmless criteria1 fields · 1 citationsanchor · a standalone constructalignment6 fields · 12 citationsjingle · one name, six constructscapability4 fields · 8 citationsjingle · one word, four constructs

References

  1. Askell, A., Bai, Y., Chen, A., Drain, D., Ganguli, D., Henighan, T., Jones, A., Joseph, N., Mann, B., DasSarma, N., Elhage, N., Hatfield-Dodds, Z., Hernandez, D., Kernion, J., Ndousse, K., Olsson, C., Amodei, D., Brown, T., Clark, J., McCandlish, S., Olah, C., & Kaplan, J. (2021). A General Language Assistant as a Laboratory for Alignment. arXiv. https://doi.org/10.48550/arxiv.2112.00861

Cited on this page?

Confirm your construct, or correct us.

If your work is cited behind one of these entries, you know better than we do whether we captured the meaning right. Confirm it, suggest an edit, or dispute it, and your name goes on the entry as an author who has checked it, linked to your ORCID.

Claim your construct