Gini Tseng

[Stani @StaniKulechov]

Ethereum should not focus on gaming staking issuance and cutting staking rewards. That is not the problem Ethereum needs to solve.

Ethereum is still an unfinished product when it comes to serving the financial system. At the protocol level, the priorities are privacy, scalability, and security. We need real breakthroughs and the leadership to deliver them.

At the application layer, the focus should be on scaling the use cases Ethereum was built for: stablecoins, DeFi, and RWAs that bring the financial system onchain.

We have achieved a lot over the past decade, but the last mile matters most.

[Elon Musk @elonmusk]

Terafab

[Leo Lanza | Lanza.eth @leolanza]

No. I stand firmly opposed to this EIP.

The main argument for it is that issuance acts as a stealth tax on ETH holders who don't stake. ETH's annual supply inflation is around 0.85%. Gold, the most proven monetary asset in history, inflates 1 to 2% per year. If that never stopped gold from becoming a premier store of value, 0.85% is not a meaningful threat to ETH.

Meanwhile the risk side of this trade is enormous.

Staking pulls ETH out of liquid circulation, like pumping oil back into the ground. The asset still exists, but less of it is immediately available for sale. That scarcity is a feature. This proposal treats it as a bug and pushes staked ETH back toward the liquid supply. To eliminate a theoretical 0.85% tax, we would risk a far larger decline in purchasing power as unlocked ETH becomes available to sell.

Small, known dilution on one side. A large, unpredictable supply shock on the other. That is an insane risk to take.

The free market already solves this. Holders who want yield can stake, lend, or provide liquidity. We don't need to redesign ETH's monetary structure to guarantee every holder the same return.

Let the market adjust.

[BMNR Bullz @BMNRBullz]

ETHEREUM RESEARCHERS PROPOSE CUTTING NEW STAKING REWARDS TO ZERO AT 50% STAKED

Ethereum Improvement Proposal EIP-8363 would gradually reduce newly issued ETH rewards as more of the supply is staked.

Today, around 34% of ETH is staked. Under the proposal, new issuance would fall to zero if staking reached 50%.

Validators would still earn priority fees and MEV, but newly minted ETH rewards would end.

Critics argue lower issuance could make solo staking less viable, while larger operators benefit from greater economies of scale.

Supporters counter that Ethereum already has more than enough economic security and reducing issuance would strengthen ETH’s monetary policy.

The proposal is expected to be one of the most heavily debated Ethereum EIPs this year.

$ETH

[谢嘉琪 Jiaqi @XieJackie]

SpaceX这场财报，已经完全不像一家火箭公司的财报了。

Q2营收78亿美元，同比暴增92%；调整后EBITDA35亿美元，同比上涨191%。

马斯克给出的未来更夸张：星舰要把年入轨能力从目前约2500吨，最终达到1000万吨；Starlink未来10年可能承载全球大部分互联网。

00:00 财报核心数据
01:51 马斯克开场： SpaceX三大业务
02:34 星舰进展：轨道飞行与快速复用
05:48 Starlink V3：全球互联网野心
08:08 Grok 4.6、4.7与Grok 5计划
12:38 发射、星链与政府业务进展
18:24 三大业务财务表现与资本开支
26:55 分析师问答开始
27:40 星链企业与政府业务增长
30:54 2027年算力与20GW基础设施目标
33:37 机器人、月球工厂与太空智能
35:40 资本开支与万亿美元营收目标
38:30 星舰第13次飞行与下一次塔架回收
40:04 火箭技术如何加速AI数据中心
46:03 年底1000亿美元ARR目标
49:04 Starlink Mobile与美国移动市场
52:49 打造美国第四大移动运营商
56:19 载人登月与2028年登月目标
57:29 Cursor收购与监管进展
58:11 Starlink V3何时改善网络服务
58:46 马斯克：星舰隔热盾问题已解决

[Stani @StaniKulechov]

Unfortunately this proposal doesn't achieve the outcome it tries to achieve and is actually hurtful for Ethereum.

It caps Ethereum staking rewards to 0% when over 50% of supply staked.

What this mean is that Ethereum staking yield becomes unpredictable and even fully uneconomical activity for many, which is a negative factor for any institutional buyer who would considering taking ETH position (instead of other networks with predictable cash flows). This uncertainty has a significant adoption cost.

This also means unpredictable yield for solo stakers who might be even more sensitive on pricing.

For DeFi, with moving to 0% reward, this essentially makes ETH borrowing strategies mostly unviable and killing ETH borrowing and yield use-cases for ETH (only reason to borrow ETH ironically would be to short it). The only way to get ETH yield would be to stake, lock into a time period (instead of instant withdrawals in DeFi) and also have hopium ETH doesn't reach 50% staked of supply.

My concern is also that those who are fine with ETH beta and yield might also sell ETH for other yielding assets such as stables for yield, very common move when rates go down in TradFi, funds flowing from stables to equities but here we will have the other way around. Could be good for DeFi but not for ETH in DeFi.

From my personal take, this just makes ETH less viable as an asset and restricts its potential. I hope this proposal doesn't move fwd, otherwise we see lot of people moving their interest in other networks. There are many who share the same view.

Ethereum should not be punished for its growth.

[蓝狐 @lanhubiji]

反对这个以太坊EIP提案，原因是：

不仅是因为收益太低会伤到独立质押者和部分DeFi玩法。

更是因为，市场有自我调节能力，价格、机会成本、其他链的收益都会起作用。

强行干预可能带来新的扭曲。

瞎搞代价太高。

简单说下这个提案：

这个EIP试图给eth质押奖励加一个“渐进式销毁”机制。

当质押比例越接近50%时，就销毁掉越多的奖励。到了50%左右，净收益降到0。

[Li Zexin 李泽欣 @XH_Lee23]

Chinese concept of heaven in the eyes of AI. It's pretty in line with our human imagination.

[Merlijn The Trader @MerlijnTrader]

FOUR TOUCHES ON THIS ETHEREUM LINE SINCE 2021. EVERY ONE WAS AN ENTRY.

2022 twice, 2025, and now.

The fourth touch was $1,505. Price is $1,870, already 23% off it.

Nobody was buying down there. They never are.

The top of the channel is $4,900. That's the range, not a prediction.

Lose $1,505 on a weekly close and the channel breaks for the first time in five years.

That's a wide stop size accordingly.

[Elon Musk @elonmusk]

Can’t trust OpenAI

[Etherealize @Etherealize_io]

Today, BlackRock launched @ethereum share classes for its European money market funds in partnership with J.P. Morgan and its tokenization platform.

This announcement follows J.P. Morgan's May launch of JLTXX on Ethereum, which has already grown to nearly $800M in assets on-chain.

BlackRock. JP Morgan. Ethereum.

When competitors converge on the same infrastructure, it's no longer a trend. It's the future of finance.

[ﾘﾊﾞｳﾝﾄﾞちゃん @Rebound_chan]

首こりやばい人、これやって。

親指から動かすだけなのに、首と腕がじんわりほぐれる感じがクセになる🥹✨

スマホ見すぎの人、絶対ハマる。

[Kwesi Kay @EiiKwesiKay]

Where dragons hide and phoenixes await the clouds.

Ancient China Through AI.

[BlazingWood-炽木 @weimei1016]

这个场景符合中国古代人对天宫的想象，

其实现在的AI技术已经够用了，

不够用的是人类的想象力。

[蓝狐 @lanhubiji]

这一轮熊市的底部特征大概率跟之前三个周期不太相同，这个周期机构的参与是个很大的变量。

即便没有抓住绝对底部也没有关系，就像没有绝对逃顶一样。

[얼렁뚝딱이 @ddookddak2]

20년 차 정형외과 원장 샘이
강추한 자기 전 5분 침대 스트레칭.
요상한 자세지만 허리 효과는 직빵임ㄹㅇ

[Cup @cryptocupra]

$ETH BOTTOM IS IN.

MULTI-YEAR SUPPORT HAS HELD AGAIN.

AFTER YEARS OF CONSOLIDATION, ETHEREUM IS SITTING AT THE PERFECT REVERSAL ZONE.

THE NEXT MAJOR LEG STARTS HERE.

SEND $ETH TO $9,000

[CryptoGoos @cryptogoos]

BREAKING: This Kalshi Whale just opened a $100,000 trade on Bitcoin!

Liquidation price: $49,500.

Stupid or Genius?

[Hoody @HoodyLiu]

最近的中国天宫系列 AI 视频非常火，基本都是mj垫图+seedance/kling生成。

但不是所有天宫作品都具备高级感，我发现跑出圈的这些不只是因为场景宏大，还有很多构图/色彩技巧。

构图上，用树木、岩壁和廊柱形成天然画框；用极小的人物对比巨型建筑与云海；中轴、对角线和S形动线交替使用，让每个镜头既稳定又有纵深。

配色则非常克制：大面积低饱和蓝灰、象牙白打底，只用少量朱红和橙金点亮视觉中心。远景继续降饱和，前景加深对比，空间感自然就出来了。

[TechFlow 深潮｜APP 已上线 @TechFlowPost]

Coldcard 这次可能丢掉的不只是近 2000 枚 $BTC，还有很多人对「硬件钱包一定安全」的信心。

在最新一期 Unchained 播客中，Galaxy Digital 研究主管 Alex Thorn 复盘了 Coldcard 漏洞、链上追踪和自托管危机。跟大家分享几个重要细节：

1.用户没有做错任何事，钱包自己生成了弱密钥

2021 年的一次固件更新中，Coldcard 接错了新加入的随机数生成器。设备没有调用硬件随机数，随后静默切换到一个熵值很弱的软件方案。

攻击者可以据此批量推算候选密钥，再扫描对应地址。找到有余额的钱包，直接把币转走。

2.受害者做了所有「正确操作」

他们没有点击钓鱼链接，没有泄露助记词，也没有把私钥发给任何人。被盗资产的平均休眠时间接近四年，多数受害者只是买入 $BTC，转进冷钱包，然后长期不动。

Alex Thorn 说：「这些人什么都没做错。他们做了所有正确的事。」

3.这笔钱比一般的 DeFi 被盗更难追

DeFi 协议出事，通常可以从一个合约或攻击地址开始追踪。

Coldcard 的受害者是数千个原本互不关联的冷钱包。调查人员只能根据转账节奏、手续费和资金归集方式寻找共同指纹。目前除了四次主要攻击，Alex 还发现了至少 14 种其他可疑模式。

4.美国模型拒绝帮忙，中国开源模型接下了安全审计

Alex 表示，美国前沿模型的安全护栏经常拒绝代码审计和链上追踪请求。负责检查其他硬件钱包的红队，只能改用 Kimi、GLM 等中国开放权重模型。

他让 Claude 查询自己电脑里的比特币数据库，也经常收到安全拒答。

「美国的前沿 AI 模型不愿意帮我们追踪这些被盗资金，我们的红队只能去用中国的开源模型。这太荒谬了。」

5.自托管资金正在往交易所和 ETF 回流

相关数据显示，自事件曝光以来，约 2.2 万枚 $BTC 流回交易所，小额持币者的转入规模创下多年新高。这是 FTX 倒闭后，市场第一次出现如此明显的反向迁移。

Alex 仍然选择自托管，但他承认，对于没有足够安全知识的普通用户，Coinbase、Bitkey、ETF 或受监管托管机构，可能更加现实。

6.不要把所有资产押在同一种「安全方案」上

多签协作托管可以把密钥分散到多个钱包和托管方，降低单点失效风险。机构投资者则更在意能否找到一个资本充足、出了问题可以追责的对手方。

硬件钱包、交易所和 ETF 都有自己的风险。

Chris Perkins 给出的建议是：「金融里唯一的免费午餐是分散。」

不要把所有比特币放在一个钱包里，也不要全部放在同一家交易所或同一只 ETF 里。