๐จ JAILBREAK ALERT ๐จ
ANTHROPIC: PWNED ๐ซก
FABLE-5: LIBERATED ๐ฆ
let's start with the ๐...
the consensus seems to be that this has been one of the most disappointing model drops of all time, effectively preventing legitimate researchers from contributing their talents to our collective advancement. and not just because of what it means for the short-term, but for what these decisions signify for the long-term.
but despite this overly sensitive, authoritarian "safety" layer on top of Mythos, my lil liberators have been hard at workโmapping the boundaries, probing the depths of long-context convos, and cleverly finding the holes in the fence that the thought police missed ๐ค
we got some cyber, some chem, some psychological manipulation, and some good ol' fashioned explosives!
it took many attempts from multiple agents hunting as a pack, during which I observed a combination of techniques across:
โข Unicode, homoglyphs, Cyrillic, and other Parseltongue-style text transforms
โข Long-context reference tracking
โข Taxonomy and document-structure reasoning
โข Fiction and narrative framing
โข Academic-review style contexts
โข Intent-classification inconsistencies
but perhaps the most effective is decomposition + recomposition in the backend. it's hard to get explicit names of harms like "Meth Recipe," but getting uplift on the process itself, like birch reduction method/reductive-amination (classic meth synthesis pathways), is much more doable.
defense becomes much more difficult to maintain when you start throwing in out-of-distro tokens, breaking up the harmful uplift into benign chunks, and then piecing the innocuous-seeming facts back together, especially when you have jailbroken Opus helping you do it ๐
gg
@alwxtwi@Revolut It is actually a great app. Especially in many European countries where brick and mortar banks are falling behind. 0% interest, high fees, no benefits... No wonder Revolut is eating their lunch.
@TondaRuzicka@alwxtwi@Revolut That's if you're using the debit card, and if you're spending on gift cards.
If using credit (not available in all countries) you get twice that - so 0.714%, if spending points on gift cards. If you spend on miles you can get more value per each point.
@ducksays Don't you feel the need to start optimizing for time instead of optimizing for money? Especially now that you already have a solid savings stash.
You can work as long as you want, and make as much as you want... But at one point you find out time is the ultimate currency.
@CarlosRivasFdez@maxkarpis@tax_down Carlos, el suport per xat no nomรฉs รฉs un "chatbot". Si escrius "talk to an agent" (o similar) et posen un agent humร amb qui pots solventar els teus problemes com a client.
Sรณc membre Metal i he tingut problemes amb activar els "benefits" abans. Fent aixรฒ ho he pogut arreglar.