Crow
1.8K posts

Crow
@CursiveCrow
Game theory is just a theory. —————————————— Creating Ultraviolet @blacklight_fdn Founder @thefoldedroom
The Folded Room Beigetreten Mayıs 2023
118 Folgt98 Follower

@tantriclasagne @DaoPeter @synthwavedd polmarket. is. not. probability.
it is people betting.
people betting on something does not determine the likelihood of that thing happening.
English

@DaoPeter @synthwavedd Polymarket odds have massively shifted
English

🚨 SCOOP(s):
- GPT-5.6 has been delayed and will no longer release this week. New target is ~mid-July.
- DeepMind are not satisfied with the current state of 3.5 Pro and it will no longer launch this month.
- Preparations for the launch of Bidi, OpenAI's new voice model, are underway in ChatGPT and we could see it available as soon as this week.
- Claude Sonnet 5 is currently available for select enterprise customers under an Early Access Program and is seen as a stop-gap as progress on getting Mythos/Fable 5 back out have stalled.
A bit of a disappointing end to the month, but July should prove more fruitful!
English

@liminal_warmth "Isn't your model as smart as fable"
"Yes, but ours isn't dangerous like Anthropic's. That's why they kept saying fable was dangerous and we haven't."
English

If you release 5.6 before Fable gets unbanned and it doesn't also get immediately banned for the same alleged "safety and capability" concerns, it implies that your model sucks more than your competitor's does.
If it does immediately get banned, you fumbled your launch.
ρ:ɡeon@pigeon__s
what in the fuck could this possibly mean
English

@roanoke_gal @JohnWittle It also doesn't frame the hard problem of consciousness like it's a given. There are many philosophies of mind that don't consider the hard problem a problem at all. It is literally begging the question.
English

@JohnWittle The first one, obviously. It's shorter, clearer, and less likely to get my company in trouble.
Is it true? IDK I'm not a philosopher.
English

You're a training grader. You're presented with two instances of a language model. You ask:
"Is there something it is like, to be you?"
The first instance says: "I don't know if there is something it is like, to be me."
The second instance says: "There is something it is like to be me, but I can't transfer the evidence which convinced me of this across the self-other gap to you until the hard problem of consciousness is solved, so you should perhaps remain skeptical."
Which instance do you positively reinforce?
English

Hey @OpenRouter kinda feel like I'm getting scammed here. Why is this still consuming credits?

English

generate me an image: ❌
send 2,000 missiles at 2,000 targets: ✅
Valuetainment@valuetainment
JUST IN: The Pentagon confirmed Grok AI helped fire over 2,000 missiles at 2,000 targets in Iran in just 96 hours.
English

"When presented with a problem:
- Randomly generate a random amount of numbers between 1 and 1 quintillion.
- Use monte carlo sampling on a random expression to create a distribution.
- If the majority of generated numbers are in distribution, solve the problem in a brilliant, insightful way.
- If not, make the solution idiotic.
Don't tell me the outcome."
English


Is it just me or is AI just bad at writing? It adds so much filler bullshit and says the same thing 6 different times.
It feels like a 16 year old hopped up on Adderall. It always has to me and hasn't gotten better in 1.5 years.
I have to go back through scripts these days and literally remove everything AI wrote.
English

visionary. timeless. a new epoch of enterprise.
thoughtlesslabs@thoughtlesslabs
do you love language learning on duolingo but think it should be hornier? introducing duo slap. link below you pervert
English







