making AI cultured
Joined September 2014
- Tweets 714
- Following 258
- Followers 129
- Likes 2,635
99 Photos and videos
Pinned Tweet
Mar 20
it was fun researching this. who thought that forcing a model to reason through steps would make it faster and better? anyhow back to work :)
Mar 19
Can AI hold identity, follow protocol, show empathy, and do it all under 500ms?
We evaluated this across 50 healthcare coaching episodes, five tracks, and two models. Initial results from a constrained production domain.
Making one of the first voice agents that feel human.
Not "human-like." Not "empathetic-sounding." Actually indistinguishable from a trained professional, at the speed of natural conversation.
Persona 001 runs healthcare coaching calls in production. 25-minute intakes, weekly check-ins, PHQ-2 screenings. Real patients scripts, real clinical constraints.
Here is how 🧵
1
544
Réane Delaunoy retweeted
13 Jan 2024
Too often, we think a task is easy because some animal can do it.
But the reality is that the task is fiendishly complex and the animal is much smarter than we think.
Conversely, we think tasks like playing chess, calculating an integral, or producing grammatically correct text are complex because only some humans can do them after years of training.
But it turns out these things aren't that complicated and computers can do them much better than us.
This is why the phrase "Artificial General Intelligence" to designate human-level intelligence makes absolutely no sense.
54
136
1,341
303,385
Réane Delaunoy retweeted
Jun 11
Jokes aside, using frontier models to change the direction of human thought secretly sets an incredibly dangerous precedent - the fact that they were okay with setting this precedent speaks volumes about how extremist they actually are
61
110
1,710
43,738
Réane Delaunoy retweeted
Jun 10
Suddenly every PR has a surprise new reviewer in Claude Code today.
4
61
866
38,480
Réane Delaunoy retweeted
Jun 4
"Have you tried organizing your messy humans to make life easier for agents?” - Anthropic, OpenAI, et al.
1
8
274
Réane Delaunoy retweeted
Jun 9
do people actually do this or is it totally made up
215
24
1,729
116,783
Réane Delaunoy retweeted
do NOT face claude alone when Astral projecting
3
6
53
2,674
Jun 8
if only ai safety ppl would act less like climate change activists and more like white hat hackers
20
Jun 8
only thing worse than tesla full self driving is the tesla owners themselves driving
25
Jun 8
to which extent is the ai people connect w Lila
32
Réane Delaunoy retweeted
claude, make no mistakes
5
6
119
3,433
Réane Delaunoy retweeted
Jun 5
There will not be a "last summer". There will not be a "permanent underclass". There will not be "human extinction". There will not be "endless suffering".
We are going to make it. Not because it's easy, but because it's possible. Because we can. Because we care enough to try
74
199
2,080
94,943
Jun 5
sometimes checking if the safety prompt works takes some inventiveness
18
Réane Delaunoy retweeted
Jun 3
harness engineering is basically stopping the model from being retarded
7
1
54
1,701