Anthropic leak all there source code,not an April joke

[snippsat]

Read more about here.
Funny looking at some of the code as eg prevent user to use bad word in Prompt,
good old Regex👀 is used(so i guess AI🫣 is not good for all).
Here is the Function from userPromptKeywords.ts rewritten in Python.

import re

NEGATIVE_PATTERN = re.compile(
    r"\b("
    r"wtf|wth|ffs|omfg|shit(ty|tiest)?|dumbass|horrible|awful|"
    r"piss(ed|ing)? off|piece of (shit|crap|junk)|what the (fuck|hell)|"
    r"fucking? (broken|useless|terrible|awful|horrible)|fuck you|"
    r"screw (this|you)|so frustrating|this sucks|damn it"
    r")\b"
)

def matches_negative_keyword(text: str) -> bool:
    lower_input = text.lower()
    return bool(NEGATIVE_PATTERN.search(lower_input))

Test.

>>> word = 'damn it'
>>> matches_negative_keyword(word)
True
>>> # Can bypass with a typo or other words usage
>>> word = 'dammn it'
>>> matches_negative_keyword(word)
False
>>> word = 'screw you'
>>> matches_negative_keyword(word)
True
>>> word = 'screw all'
>>> matches_negative_keyword(word)
False

[snippsat]

Some funny stuff in spinnerVerbs.ts
Playful🎉 loading-message words used for loading UI text in Claude Code.
Here in Python that take out 3 words,one favorite Flibbertigibbeting(being silly, flighty, or engaging in aimless chatter).

import random

SPINNER_VERBS = [
  'Accomplishing', 'Actioning', 'Actualizing', 'Architecting', 'Baking', 'Beaming', "Beboppin'", 'Befuddling', 'Billowing', 'Blanching',
  'Bloviating', 'Boogieing', 'Boondoggling', 'Booping', 'Bootstrapping', 'Brewing', 'Bunning', 'Burrowing', 'Calculating', 'Canoodling',
  'Caramelizing', 'Cascading', 'Catapulting', 'Cerebrating', 'Channeling', 'Channelling', 'Choreographing', 'Churning', 'Clauding', 'Coalescing',
  'Cogitating', 'Combobulating', 'Composing', 'Computing', 'Concocting', 'Considering', 'Contemplating', 'Cooking', 'Crafting', 'Creating',
  'Crunching', 'Crystallizing', 'Cultivating', 'Deciphering', 'Deliberating', 'Determining', 'Dilly-dallying', 'Discombobulating', 'Doing', 'Doodling',
  'Drizzling', 'Ebbing', 'Effecting', 'Elucidating', 'Embellishing', 'Enchanting', 'Envisioning', 'Evaporating', 'Fermenting', 'Fiddle-faddling',
  'Finagling', 'Flambéing', 'Flibbertigibbeting', 'Flowing', 'Flummoxing', 'Fluttering', 'Forging', 'Forming', 'Frolicking', 'Frosting',
  'Gallivanting', 'Galloping', 'Garnishing', 'Generating', 'Gesticulating', 'Germinating', 'Gitifying', 'Grooving', 'Gusting', 'Harmonizing',
  'Hashing', 'Hatching', 'Herding', 'Honking', 'Hullaballooing', 'Hyperspacing', 'Ideating', 'Imagining', 'Improvising', 'Incubating',
  'Inferring', 'Infusing', 'Ionizing', 'Jitterbugging', 'Julienning', 'Kneading', 'Leavening', 'Levitating', 'Lollygagging', 'Manifesting',
  'Marinating', 'Meandering', 'Metamorphosing', 'Misting', 'Moonwalking', 'Moseying', 'Mulling', 'Mustering', 'Musing', 'Nebulizing',
  'Nesting', 'Newspapering', 'Noodling', 'Nucleating', 'Orbiting', 'Orchestrating', 'Osmosing', 'Perambulating', 'Percolating', 'Perusing',
  'Philosophising', 'Photosynthesizing', 'Pollinating', 'Pondering', 'Pontificating', 'Pouncing', 'Precipitating', 'Prestidigitating', 'Processing', 'Proofing',
  'Propagating', 'Puttering', 'Puzzling', 'Quantumizing', 'Razzle-dazzling', 'Razzmatazzing', 'Recombobulating', 'Reticulating', 'Roosting', 'Ruminating',
  'Sautéing', 'Scampering', 'Schlepping', 'Scurrying', 'Seasoning', 'Shenaniganing', 'Shimmying', 'Simmering', 'Skedaddling', 'Sketching',
  'Slithering', 'Smooshing', 'Sock-hopping', 'Spelunking', 'Spinning', 'Sprouting', 'Stewing', 'Sublimating', 'Swirling', 'Swooping',
  'Symbioting', 'Synthesizing', 'Tempering', 'Thinking', 'Thundering', 'Tinkering', 'Tomfoolering', 'Topsy-turvying', 'Transfiguring', 'Transmuting',
  'Twisting', 'Undulating', 'Unfurling', 'Unravelling', 'Vibing', 'Waddling', 'Wandering', 'Warping', 'Whatchamacalliting', 'Whirlpooling',
  'Whirring', 'Whisking', 'Wibbling', 'Working', 'Wrangling', 'Zesting', 'Zigzagging',
]

chosen_words = random.sample(SPINNER_VERBS, 3)
print(chosen_words)

Know all 3 or none👀

['Wibbling', 'Doing', 'Flibbertigibbeting']
>>> 
['Gitifying', 'Proofing', 'Germinating']
>>> 
['Gusting', 'Forming', 'Boondoggling']

[SleipnirTheHorse]

Is it true that AI isn't really Artificial Intelligence but a self optimizing algorithm?

[snippsat]

For models like the ones from OpenAI or Anthropic...ect,the core model is usually not self-optimizing.
It is more like This:

• A huge model is trained on a lot of data(all legal 🫣👀)
  
• Training adjusts billions of parameters
  
• After training, the model is mostly fixed at inference time when answering you ☃️
  

So during normal use, it is not sitting there independently improving itself from each conversation.
It is generating the next token based on patterns learned earlier.

The self-optimizing part mostly applies to the training process, not the model’s day-to-day operation after deployment.
Self-optimizing algorithm usually refers to a system that adjusts its internal parameters to improve performance based on data or feedback.