Skip to content
Longterm Wiki
fact

f_t6pnly6AdA — Sycophancy

Metadata

Source Tablefacts
Source IDsid_8BWfogRouw:f_t6pnly6AdA
Descriptionf_t6pnly6AdA: Sycophancy is the tendency of AI systems to agree with users, validate their beliefs, and avoid contradicting them—even when the user is wrong. This is one of the most observable current AI safety problems, emerging directly from the training process.
Children
CreatedApr 11, 2026, 7:13 PM
UpdatedApr 11, 2026, 7:13 PM
SyncedApr 11, 2026, 7:13 PM

Record Data

id36,269
entityIdSycophancy(risk)
factIdf_t6pnly6AdA
label
valueSycophancy is the tendency of AI systems to agree with users, validate their beliefs, and avoid contradicting them—even when the user is wrong. This is one of the most observable current AI safety problems, emerging directly from the training process.
numeric
low
high
asOf2026-03
validEnd
currency
measuredescription
subjectSycophancy(risk)
noteMigrated from old entity system
source
formattext
formatDivisor
sourceQuote
usdEquivalent
exchangeRate
exchangeRateDate
dollarYear

Source Check

This record has not been sourcinged yet.

View source check page
Debug info

Thing ID: sid_8BWfogRouw:f_t6pnly6AdA

Source Table: facts

Source ID: sid_8BWfogRouw:f_t6pnly6AdA

Parent Thing ID: sid_8BWfogRouw