@EpsilonPM

professional forecaster (gambler) | 300x+ YTD on PMs

 Utah
Joined November 2025
  • Tweets209
  • Following895
  • Followers776
29Media
all good things must come to an end
DGN Quant
@DegenQuantOct 21
running an hft taker strategy on markets that are so heavily subsidized by the exchange that the maker doesn’t even care about getting picked off is such a beautiful thing
11
573
zoomers didn’t choose gambling we just got priced out of every other way to make money
Watcher.Guru
@WatcherGuruOct 17
JUST IN: 🇺🇸 American boomers over 75 years old now have 15 times the median household wealth of Gen Z and Millennials.
4
9
2
112
3,507
Epsilon retweeted
Clive
@Clive_99Oct 16
I fucking hate this Friday morning commute
453
10,426
360
239,779
2,976,121
Can $meta dump already 😭
Lentils@Lentils80Oct 12
I saw a bunch of Muse Wonder outputs from multiple people. This model is such a hot steaming piece of benchmaxxed garbage I couldn't even believe my eyes. Not even Gemini models are this benchmaxxed bro.
1
2
389
This is a slide from Terence Tao's 'Math 2.0' public lecture yesterday evening at Caltech. I am highly confident that, if the cure actually worked, almost any Stage 4 cancer patient would tell you that human comprehension of its mechanism is completely irrelevant to them.
1,375
968
1,225
15,310
4,149,399
Incredibly based
Andrew Curran
@AndrewCurran_Oct 8
Effective November 12th, 2026, abusive behavior towards Claude will be a violation of Anthropic's Usage Policy.
2
8
415
Epsilon retweeted
Bayesian
@Bayesian0_0Oct 1
Three AI safety researchers just left OpenAI
49
86
42
1,088
264,409
kalshi for real world utility 🔥
TroyCuban
@troycubanSep 26
My meth head neighbor was complaining how theres no liquidity for him to hedge up the loot he got from that HVAC unit he destroyed on top of Sam's Club last night
8
655
Epsilon retweeted
Tenobrus
@tenobrusSep 25
real frontier labs are not after consumer tech markets lol cute openclaw is probably worth 5T+ market cap, but is literally nothing vs RSI
Cary Bleasdale@Cary_BleasdaleSep 25
It is entirely possible that in one fell swoop Mark fucking Zuckerberg is about to corner the most lucrative demographic on earth that no other AI model is even aware exists. As in, he might acquire frontier models by BUYING Anthropic in a few years.
27
3
1
386
19,586
rip I guess it’s over, time to sell everything
Gregory Travis
@greg_travisSep 21
Just a reminder that so-called "Recursive Self Improvement" is not a thing that's actually happening with AI, nor is a thing that can happen with statistical inference Recursive self improvement is something that only a thinking machine can do and AI machines (linear algebra equation solvers) don't and can't think
10
1,110
👇
AE News
@aenewsSep 21
Bet my $4K to @EpsilonPM $1K that 2+ Millennium Prize Problems won't be solved by AI by the end of 2026 🤝
3
15
1,353
I expect Muse won’t be the only agent platform for long, am quite bullish on whatever OAI has been vagueposting about
10
737
Ok i’m not worried anymore, P(doom) to 0
First Squawk
@FirstSquawkSep 11
Trump dismisses concerns that artificial intelligence could lead to human extinction.
13
694
Anthropic won’t get anywhere near ASI without the US gov stepping in…
gfodor.id
@gfodorSep 9
They’ll never, ever admit it directly, but it doesn’t take a genius to know what Anthropic’s plans are. Even if some staff can’t admit it to themselves: world domination, via an ASI singleton (see: Bostrom), that front-runs all future human activity to eliminate ex-risk.
7
557
Epsilon retweeted
semi
@SemioticRivalrySep 8
I'm more AI-pilled than >99.99% of the population, but I still got totally destroyed in a bet as to whether a Millennium problem would be solved in the next 16 months, it only took one! AI progress is moving ridiculously fast and it doesn't seem to be stopping anytime soon.
OpenAI
@OpenAISep 8
We’re sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics. The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra. The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.
11
9
1
157
5,180
I think GPT 6 (?) Astra is quite under hyped
15
71
8,078
Epsilon retweeted
DjPolyKek
@DjPolyKekAug 24
E-sports so predictable kek
26
6
1
102
14,431
My prediction is that this happens and they extend quotas. Will be quite surprised if they decide not to.
leo 🐾
@synthwaveddJul 17
Lots of debate at Anthropic rn over whether to go ahead with this still. Should be noted the last extension was also a pretty last-minute call. We'll see
9
1,952
XAI & META (previously on par w/ "S-tier chinese labs") are now frontier US labs btw While their models are not Fable tier, yet, I don't think it's implausible that a much larger pretrain (enabled by their extensive compute) could reach or surpass that level by eoy. tier list credit @cqkten
3
1
15
3,710
Epsilon retweeted
Xeny
@XenyPMJun 17
I finally made it. Turned 3k into 100k. I joined Polymarket around October of last year. At the end of 2025, I gained over $10,000. Now, halfway through 2026, I've turned that 3k into over $100,000! Thank you to @Polymarket for providing this opportunity <3 Next goal: ???
131
123
4
587
44,680
Load more