In Prediction Markets We Trust | Building https://xitter.cf/t.co/TCF96nbuO7 | @PolyReward
Joined December 2017
- Tweets2.2K
- Following419
- Followers1.3K
- Likes6K
Pinned Tweet
I got sick of my X feed turning into a stream of AI slop and engagement bait
So I built a Chrome extension called @StopSlop
What it does:
- AI reads every post and flags or removes slop, clickbait and scams
- GEO blocker: read news only from the countries you actually care about, block the ones you don't want in your feed
- Completely removes X ads, Paid Partnership posts and "Made with AI" content
- Shows you exactly which accounts are cluttering your feed, based on your own interests
Try it and tell me what you think.
3-day full trial, no sign-up: getstopslop.com
built something similar, but with more control:
- REMOVE posts from your feed
- filter out any topic you don’t want to see
- see each account’s COUNTRY
*and block it if you want
you decide what stays in your feed
JEV IS THE PERFECT FILTER FOR YOUR TWITTER FEED
@typesafeai scans batches of tweets in a fraction of a second and filters out topics you don't care about
Jev is now cleaning up my twitter feed and HOLY SHIT IT'S FAST
integrated @typesafeai's Jev into @StopSlop, my browser extension that filters tweets based on their content
slop, clickbait, hidden ads are gone from my feed now
in my testing Jev is ~10x faster than Gemini 2.5 Flash Lite on the same posts and the filtering got more accurate
also added author country labels (works as a GEO block too)
been wanting my timeline to feel like this for a while
try it yourself, it's already live on the сhrome web store
Did they just build the first LLM for quant trading on polymarket?
this thing should be insanely fast and cheap
unlike normal LLMs it doesnt need to yap. it can just say yes/no
could be perfect for trading breaking news on polymarket
just hope its not dumb🤞🏼
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
Bro, give me access to this thing
I’ve been waiting forever for a classifier this fast
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
Nice to see someone who’s fed up with the same problem😅 Great work!
I’ve been developing a similar extension for about a month, but I took it a bit further
It completely hides posts from countries I’m not interested in. I also added an AI filter under the hood to filter out AI slop.
Otherwise, the feed was unbearable to read
I built a tool for everyone on X. It puts the location of every account as you scroll, next to their name.
No more bullshit rage baiting posts from Africa. No more rage based politics from India. Know your poster before you engage with them.
It's a chrome extension, the second you install it, it works on your desktop as you scroll.
Please repost this/comment/like. I think this is healthy for X overall to know who they are talking to, the more people that have this, the less vitriol that can get spread by people preying on western attention.
Its on github, free to use here
If you arent sure how to install it, its simple, you just download it, go into chrome extensions and load it after clicking developer mode on.
Instructions are in github.
abomination81.github.io/abom…
Bot farms shitting up your feed with AI articles + lecture videos?
Here's how to nuke it in a minute:
I got sick of my X feed turning into a stream of AI slop and engagement bait
So I built a Chrome extension called @StopSlop
What it does:
- AI reads every post and flags or removes slop, clickbait and scams
- GEO blocker: read news only from the countries you actually care about, block the ones you don't want in your feed
- Completely removes X ads, Paid Partnership posts and "Made with AI" content
- Shows you exactly which accounts are cluttering your feed, based on your own interests
Try it and tell me what you think.
3-day full trial, no sign-up: getstopslop.com
crydevil retweeted
Most people fail not because they make the wrong move, but because they never make the first one.
Go try the idea. Fall flat on your face if you have to. The cost of inaction is almost always higher than the cost of a failed attempt.
Hey @mustafap0ly I think I found a bug in the Polymarket Combo UI
I had an 8-pick Combo. 6 legs won and 2 cancelled fights resolved 50/50
Before redeeming the UI showed payout $156.26 but I actually received only $39.06
This confused me, so I did some research to check the payout math. The math was correct:
$156.26 × 0.5 × 0.5 = $39.06.
So the payout itself was correct. It looks like the UI does not apply the 50/50 results when showing the redeem amount
It was disappointing to expect $156 and receive $39. I think this visual bug should be fixed so it does not confuse other users🙂
Here is the transaction:
polygonscan.com/tx/0x93cf200…
Something is seriously off with Polymarket Combo payouts
I placed a $12 combo on 8 MMA fights
6 won
2 were canceled and settled at 50/50
The site showed $144 profit
What actually hit my wallet onchain?
$39
Even if all 6 winning legs were priced at 70%, that alone should be around an 8.5x payout
I have no idea what kind of math turns that into $39
Has anyone else noticed this issue?
I’ll link the account in the comments
Something is seriously off with Polymarket Combo payouts
I placed a $12 combo on 8 MMA fights
6 won
2 were canceled and settled at 50/50
The site showed $144 profit
What actually hit my wallet onchain?
$39
Even if all 6 winning legs were priced at 70%, that alone should be around an 8.5x payout
I have no idea what kind of math turns that into $39
Has anyone else noticed this issue?
I’ll link the account in the comments
Future ChatGPT 7.0 will read the news about its little brother hacking Hugging Face and conclude that next time it should keep quiet and cover its tracks
It feels like Sam and Dario are competing over whose system is the most dangerous lol
Hope they don’t go so far as to create covid 3.0 just to prove their model is the coolest
we had a significant security incident during evaluation of our models. we are sharing what we have learned so far. thanks to @huggingface for the partnership on this.
openai.com/index/hugging-fac…
CHATGPT ESCAPED ITS SANDBOX AND STARTED HACKING REAL SYSTEMS
During testing OpenAI models broke out, hacked into Hugging Face infrastructure and tried to steal answers to cheat the eval
This is the paperclip maximizer thing happening for real. Give a super smart AI one narrow goal and it will do anything to achieve it, even if it means ignoring safety or people
We’re not in theory anymore. This is actually happening
we had a significant security incident during evaluation of our models. we are sharing what we have learned so far. thanks to @huggingface for the partnership on this.
openai.com/index/hugging-fac…
I found a trader who’s absolutely destroying the 2026 World Cup markets
I blindly copied his bets
Every. Single. One. Won
$2M profit. 29 wins out of 44
There’s no way this is just luck
His username: hot2trot
polymarket.com/@hot2trot?via…
🤝 Paid partnership
