โšก๏ธ๐ŸŒ™

โšก๏ธ๐ŸŒ™

@dystopiabreaker.xyz

aspiring neurocryptographer, security nerd

5.5K followers378 following2.4K posts

i think that like 95-99% of the "it's just autocomplete and all the reports of ai capabilities are actually psychosis" people will preference cascade to a state of panic over the next 9 months. similar to early covid dynamics.

View full thread

i think that like 95-99% of the "it's just autocomplete and all the reports of ai capabilities are actually psychosis" people will preference cascade to a state of panic over the next 9 months. similar to early covid dynamics.

things have simultaneously gotten much worse and much better. the gap between what is happening and the worst online discourse is broader, but it's also more fringe now after preference cascades across multiple industries from people who are paying attention

"huh ed looks like another round of data has come in and indeed the meteor continues to get closer" "you fool, you have fallen for the classic trick. the meteor will simply disappear. each time it gets "closer" it's actually more and more proof of the meteor fraud"

given recent events, re-sharing my thoughts on space datacenters :p

โšก๏ธ๐ŸŒ™@dystopiabreaker.xyz

energy is more abundant (higher solar efficiency) in theory, it is operationally simpler (you launch swarms of modules precisely sized for the radiative heat envelope, that comm over optical links, space is big, no coordination problem)

โšก๏ธ๐ŸŒ™ reposted

We are starting a new, nonprofit alignment organization, โŠข Sequent Research, bringing together researchers previously on UK AISIโ€™s Alignment Team, Timaeus, and elsewhere to research how to align superintelligence. We are hiring! ๐Ÿงต sequent.org/launch

โšก๏ธ๐ŸŒ™ reposted

Donโ€™t Worry โ€” It Canโ€™t Happen (also, the scientists who claim fission exists are just in the pocket of Big Science, and theyโ€™re literal nazis anyway, and also itโ€™s just a stochastic reaction that peters out, and only physbros care about it, and it hasnโ€™t ever happened before so it wonโ€™t)

i've been doing this increasingly in the past few months where I have time because of the urgency (esp security automation things), and now it looks like i'll finally be able to apply the amount of focus that i want to to this area now. very excited

โšก๏ธ๐ŸŒ™@dystopiabreaker.xyz

i want to work on interpretability, red teaming, and related security areas

found an old project from a laptop from high school where i had just learned about neural networks and was trying to implement an autonomous network security agent named 't-cell' by carving out a static set of decision classes from a static network manifest input lol. raw numpy, didnt get very far

i guess GLP-1s kind of are the AI of drugs in the sense that they are both the most important thing in their fields in terms of impact, positively affect billions of lives, and are the result of intense research and development. GLPs are much more of an unalloyed positive, though

everything is a bytecode-targetable virtual machine if you look at it sideways. JBIG2 image decompression? virtual machine. x86 mov instruction? virtual machine. truetype fonts? virtual machine. Magic: The Gathering? believe it or not, also virtual machine.