Emily Richardson
Region: US
Friday 05 June 2026 23:52:50 GMT
1858
168
22
12
Music
Download
Comments
Electricbike :
So just a system prompt. Lol, you can’t get different values without training. And you can’t train a frontier model unless you have the compute so really it’s just an agent harness you built. Anyone can do this. And you should be happy with Claude. It has great values
2026-06-06 00:00:44
2
dalyshenanigans :
HELL YEAH
2026-07-01 00:11:49
0
KellyHatcher :
I love this, but I’m a little curious if this can cause a confirmation bias output
2026-06-23 18:09:48
0
:) :
This is so dope Emily!! So proud!
2026-06-06 14:17:23
2
Emmy AI + Health :
Yesss please! Thank you 🫶🏼
2026-06-06 21:06:05
0
Corbyn Quick │AudHD│Sub Nuke :
This sounds really cool. I've been doing something very similar with mine. But not near as built
2026-06-29 02:27:01
0
Veronica | Paradigm Shifts :
Most of my essays on AI and Web3 highlight the glaring disparities in values. Thanks for building something good.
2026-06-29 12:13:20
0
bakabakamason :
you're an actual legitimate superhero
2026-06-22 14:31:35
1
Jillian Pezet :
get it girl 💗
2026-06-30 19:55:48
0
M.J. Muyot | Author ✨ :
Sounds really interesting, look forward to learning more about this…just joined the waitlist.
2026-06-07 20:40:57
1
Jen Creates AI :
Immediate follow!
2026-06-23 13:38:28
0
Formerlyknownasprofessorbuell :
Love this! 🥰
2026-06-07 16:23:50
1
Ryan Manning :
What’s the training data? If you didn’t use any existing models.. like what’s the corpus of data for pre training?
2026-06-24 14:33:33
0
Jason Van Orden :
Very effective video. I love this. In addition to embedding your values into an AI, do you find it also necessary to counter the inherent biases baked into the underlying models we use?
2026-06-07 01:06:15
0
Jimmy Bornamann ❌ :
Well done can’t wait
2026-06-07 00:03:09
0
Audi :
very very cool
2026-06-07 03:39:54
0
MollyAlexandra :
💪❤️
2026-06-06 15:47:34
0
To see more videos from user @emilyknwn, please go to the Tikwm
homepage.
