Trendaavat aiheet
#
Bonk Eco continues to show strength amid $USELESS rally
#
Pump.fun to raise $1B token sale, traders speculating on airdrop
#
Boop.Fun leading the way with a new launchpad on Solana.

Omer Goldberg
AI is the best demoware ever invented.
You can hack something in a weekend, post a screen share, and go viral.
But turning that into a product people trust is brutal. The first time users lose trust, they’re gone.
The gap between demo and production is massive. For all the AI investment so far, there are few breakout apps beyond frontier labs and coding tools.
Closing that gap takes two capabilities:
1. Building a killer app - this is hard enough.
2. Bending an LLM to your will - this is breaking new ground, as we're all learning how to master them.
It’s hard.
But if you can do both, the opportunity is enormous.

Chaos Labs14.8. klo 22.30
Anyone can build an AI demo. Few can ship an AI product, especially in finance.
@omeragoldberg, CEO @chaos_labs, and @diogomonica, GP @HaunVentures, discuss the gap between AI prototypes and production and why finance raises the stakes.
2,61K
Control > Borrowed Security
Stripe and Circle launching L1s is just the start.
For years, the assumption was that enterprises would choose Ethereum, Solana, or another public L1 for their security. The fully decentralized future was the utopia we hoped for.
In reality, when I talk to large banks and fintechs, this is not what drives their decisions.
They are already regulated and audited. If they own the network, they see that as a stronger security guarantee. They choose the validators, control the upgrade process, and know exactly who is operating the infrastructure.
If something goes wrong, there is direct recourse. Enterprises are liable to their customers, and the business will take the hit if they act in bad faith.
And if you are moving trillions in stablecoin volume, no public network can offer meaningful economic security anyways.
If decentralization and economic security are not priorities, what's left?
Valid reasons to launch on a public L1 is distribution and DeFi interoperability. If you need that audience and those integrations, it can make sense. But for fintechs that simply want to move stablecoins, FX, or other RWAs faster and cheaper, and can bridge to other chains when necessary, owning the L1 eliminates value leakage and gives full control.
Oracles and interop messaging protocols will be critical in breaking the data silos between private enterprise chains, public networks, and the broader web. Oracles and bridging protocols will grow in their sophistication over the next years to meet the requirements of neobanks, fintechs, etc.
Decentralization will be a spectrum, will be interesting to see the distribution over time.
5,73K
The White House and SEC are proposing major updates to how we think about the integrity of market data in an onchain economy.
Both the recent WH Report and SEC’s ‘Project Crypto’ call for updates to Reg NMS, the rulebook for U.S. market data, to handle tokenized securities and other RWAs.
Their answer? Amend it.
And, per the White House, consider using oracles, aggregators, and DeFi infrastructure to collect bids, offers, and other market data.
Reg NMS ensures that traditional markets remain fair and competitive by requiring brokers to route orders to the best available price, mandating consistent data across venues, and preventing any single exchange from dominating the market.
Applying this to crypto doesn’t require reinventing the wheel. It requires connecting the data that’s already out there, allowing us to extend the infrastructure and protocols already built.
At @chaos_labs, we’re building infrastructure that solves for this. Not just serve as a price oracle for onchain assets, but rather for any asset, aggregating data from all relevant sources.
Next-gen oracles will stitch together real-time, tradfi, onchain, and general market data across protocols, agents, exchanges and more.
The Digital Assets Report is 158 pages. It mentioned oracles on 4 pages.
It's a start :)


4,4K
Generative Finance by @diogomonica
Banks and fintech sell you the financial products they have, not the ones you need.
Today, the overhead of creating financial instruments is significant.
But what if it was instant?
We know we can turn language into code ->
We know we can turn code into financial protocols ->
What's next?
Natural language -> personalized financial instruments
> "perfectly hedge my portfolio"
> "rebalance after the latest market move"
> "set a limit order, if market volatility increases 2%"
> ...
The possibilities are endless.
True AFI is not only democratizing intelligence, but allowing anyone to act on it.

Haun Ventures2.8. klo 02.17
GENERATIVE FINANCE
Much of the current conversation around AI and crypto focuses on distant possibilities. @diogomonica references biologist Stuart Kauffman’s concept of “the adjacent possible” (the set of opportunities that lie one step beyond the current reality of a system) to understand what realistically comes next.
In a conversation with @chaos_labs founder @omeragoldberg, he explains that crypto’s most immediate adjacent possible is already taking shape in the form of Generative Finance.
3,75K
Amazing piece on @ethena_labs by the research team.
Covering @aave risk exposure including:
- exchange failures
- collateral stress
But highlighting risk isn't enough
We review mitigations with @chaos_labs Oracles, setting price floors and params, preventing bad debt.

Chaos Labs18.7.2025
1/ Stress Testing Ethena: A Quantitative Look at Protocol Stability
We published a research paper on @ethena_labs evaluating how USDe and sUSDe perform under exchange failures, collateral stress, and the resulting risk to @Aave.
Here are our key findings.

5,27K
You can’t build a truth-seeking AI if you can’t trace where the truth came from.
Model bias is a training, tuning, and real time search problem.
For models, agents, and RAG systems, provenance is mandatory.
No source tracing means no credibility.
No lineage means no trust.

Chaos Labs17.7.2025
AI models learn what’s most repeated, not what’s true.
@NEARProtocol co-founder @ilblackdragon and our CEO @omeragoldberg break down why data provenance, source reputation, and community curation are critical to building trustworthy AI.
4,57K
Great AI x crypto convo with @ilblackdragon.
AI 'Scaling laws' are intriguing, but this isn't Moore's law in hardware.
Progress is far less predictable.
Domain-specific RL will be the next step function.
We're seeing its impact internally.
Excited to share more soon.

Chaos Labs15.7.2025
“I don’t like calling them scaling laws. They're not actual laws.”
@ilblackdragon, co-author of Attention Is All You Need and co-founder of @NEARProtocol, joined our CEO @omeragoldberg to discuss why future AI breakthroughs may come from improved training and formal reasoning rather than larger models.
2,89K
Johtavat
Rankkaus
Suosikit
Ketjussa trendaava
Trendaa X:ssä
Viimeisimmät suosituimmat rahoitukset
Merkittävin