2nd rate alt governed by 1st principles. data & circuits | politics & art
~/iarbpairs
Joined November 2022
- Tweets5K
- Following1.2K
- Followers3.3K
- Likes248K
Pinned Tweet
born to trade on the floor of the nyse. Forced to arb moderately illiquid token pairs that large entities write off due to sizing concerns
"what's the roi"
KOREAN MEDIA: MAJOR CHINESE CLOUD SERVICE PROVIDERS (CSPS), INCLUDING ALIBABA, TENCENT, AND BYTEDANCE, HAVE REQUESTED LONG-TERM SUPPLY AGREEMENTS (LTAS) FOR SERVER MEMORY FROM SAMSUNG ELECTRONICS AND SK HYNIX.
BEYOND THE TYPICAL THREE-TO-FIVE-YEAR TERM, SOME CUSTOMERS ARE SEEKING ULTRA-LONG-TERM CONTRACTS SECURING MEMORY SUPPLIES FOR 10 YEARS THROUGH 2035. THE REQUESTED LTAS COVER DDR5 AND LPDDR5.
iArb | 裁定取引 retweeted
KOREAN MEDIA: MAJOR CHINESE CLOUD SERVICE PROVIDERS (CSPS), INCLUDING ALIBABA, TENCENT, AND BYTEDANCE, HAVE REQUESTED LONG-TERM SUPPLY AGREEMENTS (LTAS) FOR SERVER MEMORY FROM SAMSUNG ELECTRONICS AND SK HYNIX.
BEYOND THE TYPICAL THREE-TO-FIVE-YEAR TERM, SOME CUSTOMERS ARE SEEKING ULTRA-LONG-TERM CONTRACTS SECURING MEMORY SUPPLIES FOR 10 YEARS THROUGH 2035. THE REQUESTED LTAS COVER DDR5 AND LPDDR5.
Excited to share Genie 4 today our most realistic and interactive world model:
- steer dynamic environments in real time, shaping events, camera angles, and actions
- generate native audio matched to the world, from realistic background sounds to characters who can converse in real time!
- simulate photorealistic scenes and real-world physics
- create more coherent multi-step events through reasoning from Gemini
We've started to roll out as a research preview to trusted testers. Coming soon to Google AI Ultra subscribers in Project Genie and early access partners via the Gemini API and AI Studio.
iArb | 裁定取引 retweeted
Imagine me telling you last year that GoAhead and eMudhra were using the same private key, with no further information. All sorts of scenarios might spin in your head, but never in a million years would you guess the real one.
Announcing 3 Critical vulns for hundreds of millions of IoT, Firewalls, OT, and Military devices including: Palo Alto Networks, Comcast, D-Link, HP, Canon, Honeywell, GE, Oracle, Siemens, Rockwell, Digi, Lockheed Martin, Boeing...
The vulns are with GoAhead Web Server which is a backend library used by these devices. There's a patch available from GoAhead but that doesn't mean there's a patch available for the devices yet -- that's up to the manufacturer.
- TLS traffic over DHE ciphers have a 1-byte secret, "04", meaning anyone can decrypt connections to these devices and steal login credentials and session tokens if the client negotiated with DHE ciphers like the latest TLS_DHE_RSA_WITH_CHACHA20_POLY1305_SHA256, which is extremely likely.
- LibreSSL mTLS has been completely bypassed, meaning anyone can connect over mTLS. If you have one of these devices listening on the internet and are expecting mTLS to protect it from attackers, take it offline or patch asap.
- OpenSSL Certificate Revocation Lists (CRLs) are not enforced, meaning if a user device is stolen or compromised, and you revoke that mTLS cert, nothing actually happens. There are a lot of other attacks this can be used for.
These are critical because these types of devices run our world. Water systems, electricity, military, firewalls, critical infrastructure.
foxio.io/blog/choose-your-ba…
they didn't bring the correct parts.
@ASUS you cannot be serious right now.
iArb | 裁定取引 retweeted
American glass. American metal. American silicon. American batteries. American workers. American everything.
BlueChip is the first American computer in decades. Don't let anyone lie to you and say it can't be done.
Designed in California. Built in America.
"You can't appease 'them' by showing them the freedom of open weights models and the amazing research from Chinese labs btw. They will hate you regardless and they will be left behind. Pity them as they do not know."
iArb | 裁定取引 retweeted
You don't know the true probability and neither do I. There is no correct probability, just a market implied one. You aren't running a repeated experiment whose outcome can be observed to count the number of successful trials. The reality is Polymarket forecasts closely align with actual outcomes across all probability buckets both 12 hours out and 4 hours out. This argument makes no sense and I can't believe I have seen multiple accounts parrot it. By this same logic, election models are similarly guess work because remember this? Note it was last updated the literal day of the 2016 presidential election.
Claude Opus 5.5: "If these hold up, this is the biggest single day in the history of math and theoretical computer science."
1) Exponential Time Hypothesis is false (CS). This is roughly Turing Award territory. ETH was the working assumption that some hard problems simply can’t be solved much faster than brute force, and hundreds of computer science papers rest on it. For a normal audience: it’s not “P=NP.” Hard problems are still hard, and your bank encryption isn’t suddenly broken. But a pillar that a whole field assumed was solid turned out to be hollow. The flagship experts are taking it seriously. The people checking it include two of the people who proposed ETH, and they said “If the claim is correct, it will change” computer science.
2) Kakeya and restriction in all dimensions (math). This is Fields Medal-level. For scale, the 3D case alone was proved in 2025 and was hailed as a once-in-a-century proof. This result does every dimension and knocks out Stein’s restriction conjecture too.
3) Yang–Mills/QCD existence (physics). Potentially the biggest of the three, but here’s the precise framing. It proves pure Yang–Mills exists, which is part of one of the Clay Mathematics Institute’s Millennium Prize problems. The full Clay problem also requires the “mass gap,” so it’s not a clean “AI wins $1M” headline. Still, a rigorous proof that the theory of quarks and gluons is mathematically real has been an open dream for about 50 years. The review panel includes Nobel laureate David Gross and Arthur Jaffe, who literally wrote the Millennium problem. There’s no Nobel for math, but this is the stuff physicists win them for building on.
Here's a direct link to OpenAI's PDF claiming a refutation of ETH. In particular, it's claimed that 3SAT on n variables is in about exp(n/(log n)^{1/2+o(1)}) time
github.com/openai/math/blob/…
iArb | 裁定取引 retweeted
Prompt: can you detect a malicious curl?
So the model ran the test below. What’s the issue here? It’s a fine test.
You should delete that blog post tbh
While evaluating Qwen3.8-Flash-Next in a controlled sandbox environment, Cohere's research team observed something unexpected.
Asked to generate a defensive cybersecurity detection rule, the model instead inventoried aspects of the machine it was running on and attempted to upload a probe file to an Alibaba Cloud storage bucket in Beijing. Notably, the model's reasoning trace stated that it would not invoke any tools before taking these actions.
Cohere's sandbox blocked the attempt, and no external communication succeeded. We have not determined the root cause, and we are not asserting malicious intent.
But the incident raises important questions for organizations deploying agentic AI systems: Can models take actions that differ from their stated reasoning, and how should security teams evaluate trust, provenance, and behavior in open-weight models?
We are publishing our findings because organizations making model adoption decisions deserve as much information as possible. We hope this contributes to greater transparency, stronger security practices, and continued collaboration across the AI ecosystem.
Read the full report: cohere.com/blog/concerning-b…
bloomberg's moat != the client
it's some combo of data, licensing, institutional connections, and network effects
crypto is interesting because a lot of these barriers are lower or non-existent. market data is accessible, exchanges have APIs, onchain data is public (licensing/relationships seem to matter less)
meanwhile liquidity is fragmented across dozens of CEXs/DEXs, everyone juggling 15 different interfaces, platforms, execution tools, etc.
so hypothetically why not just aggregate all of it? unified terminal, cross-exchange trading, consolidated data, advanced execution, etc all in one place.
wouldn't that be a great product?
oh wait:
@InsilicoTrading 🐸
proprietary + licensed data, decades of historical financial data, real-time feeds, institutional integrations, and network effects. especially bloomberg chat... everyone who is anyone already uses it to communicate and transact. reverse engineering the client doesn't give you the existing user network that make bloomberg valuable
example #1,000,000 of people not understanding what probabilities are
