//
sign in
Post
by @danabra.mov
PostEmbed
by @danabra.mov
Record
by @jimpick.com
Record
by @atsui.org
+ new component
Post
Thanks @kyleor.land for covering our work for ArsTechnica. arstechnica.com/ai/2026/05/s...
1mo
Overtuning can cause models to "prioritize user satisfaction over truthfulness.”...
arstechnica.com
Study: AI models that consider user's feeling are more likely to make errors
Luc Rocher