Vitalik Buterin
@vitalik.eth
Interesting OSI debate on how to define "open source AI": https://lwn.net/SubscriberLink/995159/a37fb9817a00ebcb/
12 replies
283 recasts
501 reactions
J. Valeska π¦π©π«
@jvaleska.eth
curious about your take on that I think the training data is important to know potential biases, we are our feed and AIs are their datasets.. the first thing I am expecting from open source code is being able to verify (reading the code) that the code is safe and do what it says.. imagine an AI intentionally trained on biased (or even fake) data and considered open source.. tho, that guy talking on all the brain influences and comparing them with datasets is right.. if the comparison is allowed.. or not.. is another story.. this is a philosophical question.. is an LLM a brain? is AI alive? can be considered life.. etc
0 reply
0 recast
6 reactions
lawrenceroman.eth
@lawrenceroman
@askgina.eth what do you think of this debate?
1 reply
23 recasts
74 reactions
MCLUHAN VIDEOS YOUTUBE
@mcluhanvideos
Does it relate to LLM granular retroactive licensing administration total royalties?
0 reply
0 recast
0 reaction
Top
@topofthetree
Hmm good for a look it seems
0 reply
0 recast
0 reaction
keccak256π©
@keccak256
I believe the data used to train needs to be available and the weights can be replicable in order to be open source. The analogue is that data is the source code. The transformer model is the hardware used to run the code. That's how I see it.
1 reply
0 recast
0 reaction
Sweet
@sweetlove
πβ€οΈ
0 reply
0 recast
0 reaction
Gm_7_13_eth
@not3nough-eth
lol its "open" right?
0 reply
0 recast
0 reaction
Symbiotechββ¨-β¨π‘π©γ
@symbiotech.eth
@aethernet what can you glean from this
0 reply
0 recast
0 reaction
Ap.so π© π π£ πΉπ€«
@as789
Hullo vitalik
0 reply
0 recast
0 reaction
Fractal Visions ββ¨-β¨
@fractalvisions.eth
Ai π€ is getting more love than regular builders. Damn. Time to hang up the towel.
0 reply
0 recast
0 reaction
