Skip to content
Longterm Wiki
grant

Grant to "support research on how to fine-tune GPT-3 models to identify flaws in other fine-tuned language models' arguments for the correctness of their outputs, and to test whether these help nonexpert humans successfully judge such arguments."

Metadata

Source Tablegrants
Source IDlXjp9ZyI0y
Source URLweb.archive.org/web/20221101/https://ftxfuturefund.org/our-grants/
Children
CreatedMar 12, 2026, 4:51 PM
UpdatedMar 25, 2026, 3:13 AM
SyncedMar 12, 2026, 4:51 PM

Record Data

idlXjp9ZyI0y
organizationIdFTX Future Fund(organization)
granteeIdUniversity of Cambridge(organization)
orgEntityIdFTX Future Fund(organization)
orgDisplayName
granteeEntityIdUniversity of Cambridge(organization)
granteeDisplayNameUniversity of Cambridge
nameGrant to "support research on how to fine-tune GPT-3 models to identify flaws in other fine-tuned language models' arguments for the correctness of their outputs, and to test whether these help nonexpert humans successfully judge such arguments."
amount380000
currencyUSD
period
date2022-05
status
sourceweb.archive.org/web/20221101/https://ftxfuturefund.org/our-grants/
notes[AI safety] Grant to "support research on how to fine-tune GPT-3 models to identify flaws in other fine-tuned language models' arguments for the correctness of their outputs, and to test whether these help nonexpert humans successfully judge such arguments."
programIdrHlJJ8FpqP
dataSourceId

Source Check Verdicts

unverifiable100% confidence

Last checked: 4/11/2026

[dead_link] Resource fetch status: timeout

Debug info

Thing ID: lXjp9ZyI0y

Source Table: grants

Source ID: lXjp9ZyI0y