This year, I’m going to try an experiment. I’m going to use this blog in notebook mode, posting very short shitposty things at a higher frequency.
Let’s kick things off with this screenshot of a prompt I tried in Dall-E this morning, inspired by a conversation about the implications of LxMs being really bad at repeating things exactly or maintaining invariants across responses (such as a series of images that feature the exact same object). Like humans, and unlike traditional computers, LxMs are very bad at generating highly deterministic and reproducible behavior. Modulo random-number seeds at the start of a blank-slate (empty context) generation attempt for a fixed-weights model. Based on these results, I have reached no conclusion on whether or not AI has Buddha nature.
About Venkatesh Rao
Worth noting that even at zero temperature (the ML term for “introduced variability”, and the equivalent to that zero seed you were mentioning) many ML models suffer inconsistency in results. This is likely because ML is a big pile of FLOPS, and FLOPS are a little nondeterministic even on CPU, and more-so on GPU (where most ML FLOPS are run).
“LxMs being really bad at repeating things exactly or maintaining invariants across responses”.
No zazen, no enlightenment. Teach it to bake the same shitpost loaf everyday until it learns to nature of raw to baked shitposting. Or is that what you are doing?
Plus a Robotic priest named Mindar, and a book.
“I, ROBOMANCER: JAPAN, BUDDHISM, AND ARTIFICIAL INTELLIGENCE
“Writer Ang Kia Yee reviews roboticist Masahiro Mori’s classic book “The Buddha in the Robot”, while reflecting on the text’s contemporary relevance to today’s late capitalist, geopolitical arena.
…
” With these provocations in mind, I would like to pivot from the U.S.-China binary by turning to The Buddha in the Robot, a 1981 book written by robotics engineer Dr. Masahiro Mori [21].
··· “In August this year, Kyoto’s Kōdaiji Temple began using a robotic priest to deliver sermons and impart wisdom about Buddhism to visitors[22]. Named Mindar, the humanoid was designed to evoke the Buddhist Goddess of Mercy [23]. Almost four decades after his seminal text, Mindar makes literal Masahiro Mori’s proposal in The Buddha in the Robot, that within AI is the Buddha-nature. Cementing the link were these remarks by Kōdaiji’s head monk Tensho Goto,”…
…
https://so-far.xyz/weekly/i-robomancer-japan-buddhism-and-artificial-intelligence