Enrico Bottazzi (@backaes) header
Enrico Bottazzi (@backaes) profile picture

Enrico Bottazzi

@backaes

Oulipian pataphysicist

leku.blog
Joined December 2012
  • Tweets1.5K
  • Following1K
  • Followers1.8K
169Media
Photo by Enrico Bottazzi (@backaes)Photo by Enrico Bottazzi (@backaes)Photo by Enrico Bottazzi (@backaes)Photo by Enrico Bottazzi (@backaes)Photo by Enrico Bottazzi (@backaes)Photo by Enrico Bottazzi (@backaes)Photo by Enrico Bottazzi (@backaes)Photo by Enrico Bottazzi (@backaes)Photo by Enrico Bottazzi (@backaes)Photo by Enrico Bottazzi (@backaes)
To avoid being automated away by AI, pursue research that can’t be reduced to a reward signal.
1
13
409
The interconcept space What lies between two concepts (according to an LLM)? What lies between moral persuasion and James Bond's Ford GT?
2
1
5
275
more replies
8/n Here, the interconcept space reads like supercars battery engineering
1
16
9/n Play with the interface enricobottazzi.github.io/int…
1
22
“There are too many ideas and things and people. Too many directions to go. I was starting to believe the reason it matters to care passionately about something is that it whittles the world down to a more manageable size.”
1
5
184
Effective egoism
5
181
Fuck Nuance kieranhealy.org/files/papers…
1
2
7
594
Brandolini law
Alberto Brandolini (@ziobrando) profile pictureAlberto Brandolini@ziobrando11 Jan 2013
The bullshit asimmetry: the amount of energy needed to refute bullshit is an order of magnitude bigger than to produce it.
5
146
As always, an ironic turn of events. The tech guy who likes things that fit neatly into binary classifiers is now being automated by the same technology that they contributed to building, while the girl into open-ended disciplines such as medievalism is left untouched.
3
175
Here are interactions with Emma-6, the "Italian" LLM, that make me think there might be a market for dumb LLMs
Image from Enrico Bottazzi (@backaes)
Image from Enrico Bottazzi (@backaes)
Image from Enrico Bottazzi (@backaes)
Image from Enrico Bottazzi (@backaes)
3
7
1,177
Robo-Schadenfreude: pleasure at watching a robot fail.
2
202
The affordances of model steering via Mech Interpretability techniques
Image from Enrico Bottazzi (@backaes)
1
174
Train AIs to do exactly what humans ask

Train AIs to do exactly what humans ask

Article · @backaes
2
1
3
731
Thanks to Elisabetta Rocchetti and Carlo Martinucci (@melkon88) from AI Safety Italia for feedback and discussion
1
52
Load more