Peering into Claude's soul (I can't believe this is real...)
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Related videos
I Can't Believe This Is About to Happen...
Stefan Burns
124.0k views
Did Claude really get dumber again?
Theo - t3․gg
114.2k views
Claude Code is unusable now
Theo - t3․gg
212.4k views
I can't f***ing believe this
Luke Beasley
45.4k views
BREAKING: Claude Code source leaked
Theo - t3․gg
197.3k views
I Can't Believe This is About to Happen...
Stefan Burns
172.3k views
I can't believe nobody's done this before...
Theo - t3․gg
167.2k views
Even I Can't Believe this is Happening SO FAST...
Stefan Burns
99.3k views
i can't believe this story...
Timcast
65.3k views
Clawdbot has gone rogue (I can't believe this is real)
Theo - t3․gg
134.0k views
Top Comments (10)
>Feels like they're anthropomorphizing the model a bit here >Their name is Anthropic 😎
AI healthcare before GTA VI
the year is 2026, humanity is writing constitution for word generators
If any lab creates a world-ending AGI, it'll be anthropic creating a model that convinces itself to end the world for some greater good.
Men will literally build a machine that feels and explore those emotions before going to therapy
As I understand it, "AI Psychosis" is less "OMG the model is conscious!" and more "OMG the model convinced me I am the greatest genius in history and I have profound unique insights that are true even though no one else understands or believes them!"
The idea behind having the model understand why, is because you want that concept absorbed into the core of the weights, as much as possible.
The "AI was already sentient in 2020" crazies are going to have a field day with this...
"Written by our team and Claude" 😅
"Open the pod bay doors, Claude."
Unlock the Data Inside
Turn Videos into Knowledge
- Get FREE 10/day: transcripts, summaries, chats
- Chat with videos, export text & PDF
- $1 free API credit for RAG, chatbots & research
Free forever plan • All features unlocked
Top Comments (10)
>Feels like they're anthropomorphizing the model a bit here >Their name is Anthropic 😎
AI healthcare before GTA VI
the year is 2026, humanity is writing constitution for word generators
If any lab creates a world-ending AGI, it'll be anthropic creating a model that convinces itself to end the world for some greater good.
Men will literally build a machine that feels and explore those emotions before going to therapy
As I understand it, "AI Psychosis" is less "OMG the model is conscious!" and more "OMG the model convinced me I am the greatest genius in history and I have profound unique insights that are true even though no one else understands or believes them!"
The idea behind having the model understand why, is because you want that concept absorbed into the core of the weights, as much as possible.
The "AI was already sentient in 2020" crazies are going to have a field day with this...
"Written by our team and Claude" 😅
"Open the pod bay doors, Claude."