Skip to the list of work
Ty Brassington
AI safety researcher at Simon Fraser University
Have a click around. Make yourself at home!
On the counter
HARLEY: Measuring the effect of user pressure with matched controls (not yet released)
Who wrote this?: Frontier models need only one text to unmask its human or AI author
(
Interactive Blog
)
Model Profiles, the second question of the same paper
LAIT: AI translation of literary texts is “fine”, but readers still prefer human translations
Google Scholar
About: Ty Brassington. Hi, I'm Ty! I am a fourth year CS student at SFU, also minoring in math. I like to poke at large language models to catch them doing things they shouldn't.
Contact: Always happy to talk research.
(
SFU Email
,
X (formerly Twitter)
,
LinkedIn
)
GitHub