Tweet Disematkan
ElementZero79
20.2K posts
ElementZero79
@ElementZero79
Here to… why am I here actually? 🤔
Bergabung Temmuz 2016
295 Mengikuti301 Pengikut
@ChefBonsaidemic Masterpiece of simple genius these things are
English
People are starting to say Google is OK with Scaled Content on Reddit :D
reddit.com/r/Agentic_SEO/…
English
Claude Mythos may start as Enterprise only.
Even if available to consumers, it's likely too expensive for an individual to run for serious work.
If you are an employer who offers Mythos token access to employees you have a distinct advantage.
I'd gladly forego equity comp for a 1 billon token/mo Claude Mythos allowance.
Chubby♨️@kimmonismus
Claude Mythos is scheduled for Q3 2026. We're currently still in Q1 2026. Anthropic will certainly factor in compute limitations and project how good the model will be. However, if it's already this good *now*, and NVIDIA's Vera Rubin GPUs aren't even live yet, what will we see by the end of the year? Or in 2027? I'm increasingly convinced that Dario is right and superintelligence will become a reality. But there's also a caveat: Anthropic is also planning its IPO for Q3 2026. I certainly see a connection there.
English
@daniel_mac8 That’s fine, they can choose to be one of those. But tbh, the pressure on them is so big atm, from OpenAI, from Chinese labs and recently Cursor that it may be what breaks the camel’s back. Companies that spit into the well don’t usually do great in the long term 🤷🏻♂️
English
@ElementZero79 this is true. hadn’t thought of that.
there are plenty of enterprise only software providers out there though
English
Anthropic likes “doom-marketing” where they hype up their products using “their fears and concern.”
it’s an interesting approach tbh 😅
Disclose.tv@disclosetv
JUST IN - Leaked documents from Anthropic show that a new generation of super-strong models, "Claude Mythos," is already in testing with Anthropic believing it "poses unprecedented cybersecurity risks." — Fortune
English
ElementZero79 me-retweet
Rule of thumb: if this man speaks on a podcast, there’s a 99% likelihood the information is entirely BULLSHIT.
English
@daniel_mac8 Google should just do research, not models 😅
Seriously though, the research out of Google over the years is nothing short of amazing.
English
TurboQuant is a genuine gamechanger.
I used that term sparingly.
What it does:
> Models use 6x less memory
> While getting 8x faster
> Without a drop in capability
An algorithmic improvement that has nothing to do with training.
Local models you can run will get better too.
Google Research@GoogleResearch
Introducing TurboQuant: Our new compression algorithm that reduces LLM key-value cache memory by at least 6x and delivers up to 8x speedup, all with zero accuracy loss, redefining AI efficiency. Read the blog to learn how it achieves these results: goo.gle/4bsq2qI
English
ElementZero79 me-retweet
Not all losers are Jew haters, but all Jew haters are losers
English
Gotta love how many Israel and Judaism experts are on this platform… so well informed and scholarly… unbiased too. Amazing!
English
This is a Black Ashkenazi. She keeps and takes on the traditions, customs and laws of ashkenazim. She prays Ashkenazi prayers and attends Ashkenazi synagogues in Jerusalem.
Maurice "Lil Moe" Williams@lilmoemusic
@ashira_solomon How are they Ashkenazi then?
English
ElementZero79 me-retweet
If X existed in 1941:
"Why do Jews get their own ghetto in Warsaw? What are they hiding?"
English
GPT-5.4 Pro solved an open FrontierMath problem.
The problem was open since 2019 but the original authors of the problem were unable to solve it.
This is already a form of narrow ASI?
It will soon be possible in every field of science, and when it is the world changes.
English
43 young for cancer and being mega rich to afford all the cutting edge treatments.. probably state sponsored
English
English
> be Jane Street intern
> $350k/yr, still a cog
> one night find prediction markets
> same mispriced probabilities you model at work
> except here nobody's competing
> 87% of wallets bet on feelings
> you run 5 PhD formulas on a google sheet
> ceasefire contract at $0.30 - Bayes spits 58%
> you buy. CNN posts the headline 40 minutes later
> 6 months: +340%. no boss. no bloomberg terminal
> steady 7 figures/month. alarm clock deleted
> hand in resignation at Jane Street
> they offer double to stay
> you already forgot where the office is
Hanako@hanakoxbt
English