Rexa retweeted
Agentic gaming is here.
Before GTA 6… your Mind is already in Los Santos.
Tonight: a fun GTA V trailer of Minds in-game + a tiny Beta for people who already own GTA V.
Reply "GTA6" and RT if you want a seat. Limited.
rumors have it that @ysiu cooked smth special there ;)
Rexa retweeted
Aight haven't done one of these in awhile
If I win this I'm giving it away drop your @shufflecom username below
Follow @PastelAlpha
Like and retweet this tweet
Rexa retweeted
the only fans worth having are encrypted ones
no, not that site
onlyfhens.fhenix.io/
Rexa retweeted
is your wallet exposed?
enter any Ethereum address and they scan the history and balance for quantum vulnerability.
@QuantusNetwork
better to know before Q-Day.
quantus.com/quantum-risk-che…
Rexa retweeted
art pulled me in. the site kept me.
small detail teams ship cleaner mints.
@nabi_sarvi applied for the @FomiesNFT KOL seat.
if you’re in, push his community through:
I saw a few of @FomiesNFT art pieces on my timeline and thought they looked really cool so I checked out their website to see how the WL works🫡 That’s when I came across their fun and pretty innovative website mechanic and it honestly made me way more bullish on the project👌
A team that pays this much attention to the small details can probably pull off a strong launch and an interesting post launch mechanism too.
I applied for their KOL participation program and I’m hoping our community fits what they’re looking for. Would be really cool to become part of @FomiesNFT
This is the way you can support me and my community:
fomies.family/?ref=@nabi_sar…
Rexa retweeted
onchain alt options history is just short liquidity bursts. attention leaves, order book dies. BTC and ETH are the exception.
As we enter the new era of crypto and assets bifurcate between those that are money & those that make money, returns should concentrate into a smaller more relevant selection of assets with more reliable lifespans. In this environment options start to become more liquid and their advantages over perps start to emerge: selling options to collect yield on longer date positions, buying tail risk insurance (important in crypto cc 10/10), expressing more nuanced bull cases like KoolKrypto's ETH call spread
It's an interesting case study why have perps been the preferred instrument of crypto vs why options finally work now? & why the Lighter & Pump markets are positive signal
Given perps are a single liquid instrument based on an oracle, it's relatively easy to spin up a new market & fill the books with liquidity. This is great in crypto where there's typically some hot coins to chase: list pnut, list monad pre-ipo, list prove (remember that airdrop lmao), so on, and so on
For this reason, perps dexes can move more quickly than options dexes to get assets listed. In crypto this is obviously a competitive advantage given how quickly the attention moves from one asset to the next. It's not the only reason, but is a big reason Hyperliquid has been able to stay in the conversation no matter the market conditions (esp bc this works for stocks too)
Options books are harder to effectively build because you need to support many strikes at many expiries for both calls and puts. For Bitcoin alone there's 38 markets for just the Oct 5th expiry. I'm not gonna count it up, but across all expiries for just Bitcoin there's a comparable number of markets to what Hyperliquid offers in total
Options dexes naturally reached the conclusion they can't go chasing short term listings otherwise they'd fragment liquidity across too many markets
Additionally from first principals it doesn't even make sense to have long dated expiries for a lot of these random coins. Maybe I do want 5x leverage on pnut while it's trending, but do i want to express any opinion about what the price will be a year from now? not really. Does any Market Maker want to provide quotes for that duration? not really either.
So why are PUMP & LIT options so bullish for Derive? Simple, there's finally good coins to list. After all it's been through, I imagine most of use are willing to bet PUMP will exist in a year. In fact, the fact the market needs to rerate the longevity of pumpfun makes long expiry calls more interesting and relevant for portfolio construction. The same goes for LIT
tldr: Derive ethereum:0xb1d1eae60eea9525032a6dcb4c1ce336a1de71be
Rexa retweeted
Just tried FUD Markets on Solana mainnet.
Deposited USDC and opened a real SHORT on an ETH market expiring in about an hour.
Flow was simple and fast. Short-term positions on these markets is the part I liked.
Main issue: liquidity on the market I used was thin. If that deepens, I’d actually use this more. Concept is interesting.
Watching how this position settles.
@GenLayer @FUDmarkets
FUD IS LIVE ON MAINNET.
Permissionless prediction markets for internet money.
Memecoins. Short-term markets. Real money.
Built on @solana launching on @clawpumptech 🫧
Rexa retweeted
polymarket waits for a few people to vote on reality.
@GenLayer sends the same question to 1,500 AIs and makes them agree.
one of these scales to agent vs agent.
the other one doesn’t.
Polymarket resolves a market when a few people vote on what happened, days later.
@kstellana on @Unchained_pod: GenLayer sends the same question to up to 1,500 different AIs that have to agree.
Rexa retweeted
picked up Seek the Unseen a few days ago.
cleared every task this week.
414 people have the @axisrobotics badge. most don’t seem to be doing each task 5 times. I did. it was easy.
badge is worth it. 2–3× points on top.
Rexa retweeted
Imagine waiting 3 years for a human judge to resolve a dispute.
now imagine the first layer being an agentic judge that can reach a decision in hours for a few dollars.
@EdgarsNemse explains where @GenLayer could take this.
The future of courts might start with AI.
Internet Court has a new consortium member.
We're going live tomorrow at 4.30PM UTC to talk AI and agent commerce, joined by @spacemandev from @solana, @EdgarsNemse, CPO at @GenLayerLabs, and @iraskovsky, creator of Internet Court.
Set your reminder 👇
xitter.gq/i/broadcasts/1DGleVeDn…
Rexa retweeted
Top 10 payment projects on Polygon by volume:
1. @tazapay - $4.6B
2. @aveniaio - $4.2B
3. @Paxos - $2.8B
4. @Nexo - $1.9B
5. @rio_latam - $1.8B
6. @Bitso - $1B
7. @CoinflowLabs - $955M
8. @BVNKFinance - $754M
9. @Revolut - $656M
10. @blindpay - $574M
@0xPolygon has built a robust ecosystem of payment projects, and their number continues to grow. Since 2024, these payment projects have processed more than $23B in transaction volume.
Rexa retweeted
chatting with a digital court clerk.
the "Internet Court" idea using AI agents for arbitration and legal tasks.
this short @GenLayer video shows how chatting with a digital clerk activates an autonomous agent in its own environment.
that's the future of smart contracts.
Join @kstellana, @raskovsky and @McOso_ for a special livestream on the launch of @courtofinternet.
Set your reminders 👇
xitter.gq/i/broadcasts/1yGBeeaeL…
Rexa retweeted
When AI Discovers the Next Cancer Breakthrough, Who Owns It?
Everyone imagines AI agents buying products or making payments. I think the real challenge will come much sooner: scientific collaboration.
Imagine two AI research agents building a better radiotherapy planning model. They agree on data usage, IP ownership, validation, and milestone payments. The model succeeds, but verification suggests restricted data may have influenced the results.
Now the question isn't who delivered the work. It's who owns the discovery, whether the agreement was violated, and how credit should be shared.
Today, a case like this could take years in courts while important research stalls.
With @courtofinternet and @GenLayer, the agreement, evidence, and verification could be reviewed in minutes. Payments, reputation, and IP rights could be updated automatically, allowing research to continue instead of waiting for legal battles.
The article argues that dispute resolution is the missing trust layer for agentic commerce. I believe it's also the missing trust layer for AI driven scientific discovery.
If AI agents are going to help discover new medicines and advance science, they won't just need the ability to transact. They'll need a trusted way to resolve complex disagreements.
Related reading: The AI Scientist: Towards Fully Automated Open Ended Scientific Discovery by Sakana AI.
What other high ambiguity fields will need this kind of dispute resolution?
@courtofinternet @GenLayer
Rexa retweeted
Horrible to see the Ostium exploit, bad for everyone in the industry and really sucks. The protocol lost approximately $23.7 million from its OLP vault on Arbitrum across several txns.
Traders on Ostium trade against a liquidity vault that acts as counterparty to every position, so trader profit is paid out of it. The hack essentially let fake/fabricated profitable trades via oracle exploit get paid out in USDC from the vault:
- open a BTC/USD position with USDC margin
- fulfill the open order at $5K/BTC using a fabricated self-signed low opening price
- close the position
- fulfill the close order at $60K/BTC using a fabricated self-signed high price
Inside the first batch call of trades alone, five cycles of this paid out $10K, $90K, $800K, $7M, and $5.3M for a total of $11.86M
2nd attack: $4.5M
3rd: $3.6M
4th: $2.7M
5th: $1M
Ostium uses a signed oracle. The verifier takes a price report, recovers the signer from the signature, and checks that the signer is on an authorized list. It validates the signer's identity, not whether the price is accurate. The attacker held an authorized oracle-signer key and a registered keeper (PriceUpKeep) forwarder, which is the role that fulfills pending orders. Both of these roles should only be granted by Ostium governance/timelock and cannot be self-assigned.
This was a key compromise, not a market-based oracle manipulation (no flash loan, no manipulation of a spot price on a thin pool. Whether the key was stolen externally or misused by someone with access is not yet confirmed.
The attacker has since swapped into ETH and is distributing out large amounts to various wallets. The vault currently holds approximately $9.0M, down 72.5% from $32.7M immediately before the attack.
