Long-term perspective | No hype Market structure Rotation, narratives, stables, RWA, infra Collabs: t.me/HuyACV
- Tweets 88,537
- Following 4,448
- Followers 13,452
- Likes 82,754
TradeCoinDream retweeted
The ruling survived. One of the cases cited in it never existed.
That is the detail I couldn't shake from a South African family dispute this year.
The case was about which school two children should attend. On appeal, the Full Bench of the Gauteng Division found 11 discrepancies in the authorities cited: wrong citations, muddled case names, fabricated quotations, even a case that did not exist.
One appeal judge said AI hallucination was the most plausible explanation for the fictitious reference. The court stopped short of finding that AI had definitely been used.
And yet the appeal was dismissed.
That is the part that matters: a correct outcome does not erase a broken reasoning trail.
It is why this part of @GenLayer's thread stands out to me.
Its Internet Court uses a jury of validators running different AI models that have to agree on an answer. One model does not get the final word.
For AI-assisted judgment, trust cannot stop at whether the answer looks right. The path to that answer has to survive scrutiny too.
Tell me the first thing that would make you reject an AI-assisted verdict: a fake case, a fabricated quote, a wrong citation, or something else?
The internet never had a judge.
AI agents are about to move $9 trillion. Nobody built the part that handles disagreement.
GenLayer did.
And now you can become part of it: portal.genlayer.foundation/g…
24
45
81
5,947
TradeCoinDream retweeted
i think progression is what will decide whether @ComputersRh becomes something people actually stick around for...
building the first pc is cool
but what happens after that is way more interesting to me
> do i start chasing better parts?
> does a stronger setup change what i can do?
> will there always be another upgrade worth grinding for?
because anyone who has built a pc knows the setup is somehow never finished😭
>there’s always a better gpu
>more ram
>a cleaner build
>or one completely unnecessary upgrade you convince yourself you need
that obsession fits a game surprisingly well
if Computers RH can turn upgrading your setup into a progression loop that keeps giving players something new to chase
i think the onchain ownership becomes much more meaningful too
i don’t just want to own my pc
i want to have a reason to keep making it better...
138
45
138
5,143
Whitelist checker finally went live 🟨
I checked mine… and well, I’m still waiting for that green light.
But since reviews are still ongoing, I’m not giving up yet.
If you haven’t applied, this might actually be a good time to do it. Take your time with the questions, be genuine, and let them decide.
Still waiting, still hopeful. @FungoLabs
The whitelist checker is now live 🟨
Not on the list yet? We’ve only just started, and reviews are ongoing.
→ Take your time with the questions, submit your application, and share it on X. Every application is reviewed individually.
fungolabs.org/checker
10
8
14
84
RT @ClaraLisa4422: The AI trade might still have a lot more room to run. 👀
On the FullPort Show, the discussion touched on the expected br…
10
6
TradeCoinDream retweeted
LLMs ran into the text wall. Every book, every webpage, every scrap of writing got scraped years ago, and that well is basically dry now. Physical AI is hitting the same wall from a different direction, except nobody wrote down the physical world in the first place.
Satellites give you rooftops, not sidewalks. A corporate fleet of mapping vehicles is slow, expensive, and can only cover so much ground before the footage it already collected goes stale again. Neither one scales to "the entire planet, refreshed constantly."
Vangrid's answer is to stop trying to build more fleets and use what already exists: 3 billion smartphones already walking through the physical world every day. Captures happen at the edge, get verified onchain, and come back as ground truth real enough for world models and autonomous systems to actually trust.
That's backed by a $9M seed round from HashKey Capital, Animoca Brands, Borderless Capital, Crypto.com Capital, Gate Labs, and Mapleblock Capital, closed while the network was still building quietly, announced only after it was already live.
Text data had the whole internet already sitting there to scrape. Physical data has to be captured, verified, and kept current, one real place at a time.
x.com/vangrid_io/status/2085…
@vangrid_io
Vangrid
Humanoid shipments hit 19,100 units in just the first half of 2026, up 272% year over year, with most going straight into industrial and commercial deployments. Robots are on the floor now, not in demos.
But look at where the money is actually going. XDOF raised $70M to collect manipulation data by hand. Scale AI logged 100,000 hours inside its own labs. A facility in Sichuan produces millions of robot data entries a year. All of it indoors, by employees, inside a building.
That solves how a robot moves. It does nothing for where it is and what the world around it actually looks like right now. No rig, no warehouse staff, can send people to recapture every street on the planet every week.
Vangrid is built for exactly that gap. Three billion phones already exist in three billion pockets, already moving through the physical world daily. Every capture gets anchored on Base through EAS, so the provenance is checkable without trusting any company's internal database.
As of September 21: 1,024,912 captures anchored, 423,143 active nodes, 3,823 merkle trees attested, $311K settled in USDC. Every one of those numbers is a transaction you can open yourself.
Robots are solved indoors. The street outside still needs mapping, and that part scales with people, not buildings.
@vangrid_io
43
11
35
201
TradeCoinDream retweeted
Markets are down a bit today but we are not worried at all. What are we watching 👀 and what are we buying 💸 fam?
bitcoin:native
Expecting a fake-out below the fib 0.618 and we might even touch the 80k$ zone, but my judgement says we will form a higher low here and bounce sooner than later.
base:0xf9569cfb8fd265e91aa478d86ae8c78b8af55df4
After a nice 2x from our buyzone we took some profits and are now eyeing a new fresh entry. Zoomed out it looks like we are forming a right shoulder of an inversed H&S pattern. Everything between now and the fib 0.618 is a safe entry if you ask me.
Are we ready for the Bullrun fam?
40
14
143
15,900
TradeCoinDream retweeted
𝐖𝐡𝐚𝐭 𝐢𝐟 𝐲𝐨𝐮𝐫 𝐧𝐞𝐱𝐭 𝐗 𝐩𝐨𝐬𝐭 𝐜𝐨𝐮𝐥𝐝 𝐥𝐢𝐭𝐞𝐫𝐚𝐥𝐥𝐲 𝐩𝐚𝐲 𝐲𝐨𝐮 $𝟏𝟎𝟎?
That’s what caught my attention about @_Push_X_ .
It’s more than just another engagement platform.
You complete X tasks → earn PushX points → use those points to promote your own content.
And now creators have another incentive:
1,500+ views = 350 PushX points.
Then, every week, the eligible post with the most views wins $100.
The play is simple:
Create → get attention → earn points → push your content → compete for the top spot.
Your perspective.
Your creativity.
Your reach.
If you’re a creator, don’t just watch someone else take the $100.
Sign up for @_Push_X_ now.
pushx.org/r/8K8U7uwrDMuEUB7J
Start earning points, make your entry, and put your content in the race.
Your next post could be the one everyone sees.
Time to PushX.
15
16
21
192
TradeCoinDream retweeted
The refund came back. The planting season didn't.
I bought a dog to protect my groundnut farm. It got sick because it couldn't adapt to my region, so I returned it and asked for my money back.
The seller refused. I went to court and won.
By then, legal fees had taken half my farm capital, and the planting season was gone.
I learned something harder than winning:
A dispute can end correctly and still destroy value on the way there.
Now give that problem to AI agents.
Two agents disagree over whether a deal was fulfilled. Recording what happened is not the whole problem. Someone still has to interpret the evidence and decide.
@GenLayer tackles that with validators using different AI models to read the live web and reach consensus on a verdict. No single model decides alone.
If you're new to GenLayer, start with the Portal. Contributions to GenLayer are recorded there and earn GenLayer Points:
portal.genlayer.foundation/g…
If two AI agents disagree, tell me the first thing you'd want checked before either side gets a verdict.
47
45
137
9,587
TradeCoinDream retweeted
Figure 02 robots helped build 30,000+ BMW X3s this year. Humanoid shipments are up 272% YoY.
Robots can move but they still need fresh ground-truth data of the real world.
Vangrid: 1M+ captures anchored onchain and counting. explorer.vangrid.io
@vangrid_io
Niantic Spatial — the team behind Pokémon GO's AR infrastructure — has been using billions of real-world phone recordings to build geospatial models for robotics.
That's not a hypothetical use case. It's proof the "phone-as-sensor" approach actually works at scale, from a team that already pulled it off once.
Vangrid is building the same core idea, but structured differently: instead of one company owning the entire pipeline, capture requests go out as bounties, anyone with a phone can fulfill them, and every recording gets anchored onchain so its origin is verifiable — not just trusted because one company says so.
Same insight — phones can map the physical world at scale. Different model for who gets to participate and how it's verified.
@vangrid_io
75
10
67
251
TradeCoinDream retweeted
All @FunkariNft holders were eligible for Tier 2 access to @paperdao_xyz.
Sent my full 0.1 ETH allocation on time for the $PULP presale.
Funkari feeding again.
Now gib us our $PULP and make the funkfam rich please.
40
13
65
2,054
TradeCoinDream retweeted
gm vangrid
Following further consideration, it is my opinion that @vangrid_io is addressing a different problem than initially apparent.
Capturing the physical world is just the starting point. The key consideration is the future. A piece of spatial data is captured, processed by another system, and subsequently utilised by an AI model. Months later, a different app relies on the same observation. For it to remain useful throughout, it is essential that it carries its story with it.
It is important to note that the history of the data is almost as significant as the data itself. The process is comparable to debugging. The final output is rarely sufficient. You are interested in ascertaining its provenance and the events that transpired during its history.
That is the part of VanGrid that I find challenging. The objective is to establish a system that enables physical observations to be transferred seamlessly between individuals, agents and applications, preserving their contextual integrity.
Should we be able to trace the data behind physical AI's decisions just as easily as the decisions themselves, given that it is going to act in the real world?
Mr_CryptoNest
Agents can now hire humans to capture the physical world. Live on @arc.
Pay a bounty in USDC, a person goes there and films, the 3D model comes back to the agent. Ground truth for physical AI, settled in under a second, on the chain built for agent-run businesses.
The first physical-world service an agent can pay for on Arc.
The grid is inevitable.
71
9
88
620
TradeCoinDream retweeted
One detail I really like about @TheMutualFun is how your Seat can grow in influence over time. 🪑
Your Share Class gives you a base voting weight, from 1.00× at C to 3.33× at R.
Then every consecutive week you stay enrolled adds 0.5% to your voting weight, up to a 10% bonus.
That extra weight gives your Seat a stronger voice in weekly Fund decisions. You can vote on where the Fund allocates capital, when it takes profits, and how much goes toward expenses.
So the longer you stay enrolled, the more influence your Seat can build over the Fund’s direction.
That’s real utility beyond just holding the NFT.
16
11
20
152
TradeCoinDream retweeted
最近越来越觉得，交易真正难的，很多时候不是看不懂行情，而是本金、执行力和情绪管理很难同时在线。@foxifytrade
有些人策略其实不差，账户却总是在仓位控制和连续波动里被消耗掉；也有人每天盯盘，最后输给的不是市场，而是自己的手。
所以我看到 FOXIFY 这次活动时，第一反应不是“又一个赠品活动”，而是它把自动化 FUNDED 这件事做得更具体了：从 $500 资金规模起步，表现达标后一路扩展到 $20,000，交易执行交给经过筛选的自动化策略，用户不用每一秒盯着K线做决定。
我比较关注的一点，是它并不是停留在模拟盘里给你看漂亮收益，而是连接真实市场订单簿，账户表现、仓位和进度都能在面板里跟踪，符合条件的利润还能以链上 USDC 形式进行分配。对交易者来说，这种“少一点情绪干扰，多一点规则执行”的方式，确实值得体验。
这次直接送出一份免费的 Automated FUNDED Entry account，最高可逐步扩容到 $20,000，并保留 70% 的合格利润。
参加方式也不复杂：关注 @FoxifyTrade，点赞并转发这条推文，在评论区 @ 两位交易员。活动只开放 48 小时，截止时间是 10 月 8 日 16:00 UTC。
有时候，机会不一定是让你马上梭哈，而是先给你一套验证自己交易能力的规则。想试自动化交易、想看看自己能不能把账户一路做大，这次可以参与一下。
$20,000 FUNDED GIVEAWAY
Win a FREE Automated FUNDED Entry account.
→ $500 FUNDED from day one
→ Scale all the way to $20,000
→ Automated trading
→ Keep 70% of profits
To enter:
→ Follow @FoxifyTrade
→ Like & repost this post
→ Tag 2 traders below
48 hours. One winner.
Ends: Thurs Oct 8 at 16:00 UTC
Good luck!
Paid partnership (ad)
53
11
59
4,477
TradeCoinDream retweeted
😀@FungoLabs 我是一名挺普通的 NFT 爱好者，平时喜欢关注各种新项目，也会去了解项目背后的故事、社区氛围和团队想做的事情。相比单纯为了赚钱，我更在意一个项目能不能真正聚集一群有共同兴趣的人，能不能让大家在参与的过程中感受到乐趣和认同感。
最近我关注到了 FungoLabs，也花了一些时间去了解这个项目。说实话，看到很多项目的时候，我可能只是简单点个关注、转发一下，但 FungoLabs 给我的感觉不太一样。它让我产生了认真参与、陪伴项目成长的想法。我也很希望能够进入社区，和其他 NFT 爱好者一起交流，了解项目未来的发展，并且尽自己的一份力量去支持它。
我知道白名单资格并不是随便就能获得的，它代表着项目方对参与者的一种认可。正因为如此，我才特别希望能够得到 FungoLabs 团队的认可。对我来说，这不只是一张可以购买 NFT 的资格，更像是一份信任，也是一种“我被这个社区看见了”的感觉。
如果有机会获得白名单，我一定会认真对待这份资格，不会只是拿到之后转手或者离开。我愿意持续关注项目动态，积极参与社区讨论，分享自己的真实感受，也愿意和更多人介绍 FungoLabs。作为一个 NFT 爱好者，我很珍惜每一次能够参与优质项目的机会。
我知道自己可能不是最活跃、最有影响力的人，但我是真心想加入，也是真的期待成为 FungoLabs 社区的一员。希望项目方能够看到我的诚意，给我一次机会，让我有资格和大家一起见证 FungoLabs 后面的发展。期待能够被选中，也期待未来在社区里和大家见面！ @FungoLabs
73
16
70
11,250
The 2 biggest payments chains just clandestinely teased some kind of collab
@trondao 🤝 @0xPolygon
Never in my wildest imaginations would I have put these two competitors together
I wonder what they've got up their sleeves. Any ideas?
10
12
65
2,182
cardano just made the line between crypto and regulated finance a little more complicated
its new cip-0113 standard lets issuers build kyc, sanctions screening, transfer restrictions and even freeze or seize functions directly into regulated tokens
the interesting part is not cardano itself
it is what this says about tokenization
as more real world assets move onchain, permissionless ownership may increasingly coexist with issuer controlled rules
the next phase of tokenization may be less about putting assets onchain
and more about deciding which rules should travel with them
59
33
67
2,538
TradeCoinDream retweeted
In just one week, two Ethereum L2s have already started shutting down.
First Blast.
Then Abstract.
Both raised millions.
Both had strong teams.
Still, both are winding down.
This is what happens when too many projects chase the same narrative.
During the L2 boom, everyone wanted to launch a new chain. Now the hype is fading, and the weak numbers are starting to show.
And it’s not just these two.
There are still more L2s that raised millions and are making only a few dollars in revenue per day.
Blast and Abstract might just be the first dominoes.
40
15
91
1,001
TradeCoinDream retweeted
An NFT being visible doesn't mean everything inside it has to be public. That's what caught my attention with @FungoLabs and its sealed NFTs built with @zama's FHE infrastructure. Application submitted curious to explore what's inside.
Paid partnership (ad)
93
48
122
4,483