小鱼Daisy
Open Trade
BNB Holder
Frequent Trader
2.4 Years
28 Following
2.8K+ Followers
3.7K+ Liked
41 Shared
Posts
Portfolio
All
Quotes
Videos
Live
I said that getting up early brings blessings!!!
After finishing my makeup, it's ten times better.
After finishing my makeup, it's ten times better.
🎙️ Check the market
End
17 m 37 s
75
0
Aave's Shocking Slippage Incident!
On March 13, 2026, in the early morning, a user attempted to exchange approximately $50.43 million worth of aEthUSDT (the interest-bearing token for USDT deposited in Aave) for aEthAAVE through the Aave official interface. As a result, the order size far exceeded the liquidity, triggering an extreme price impact and >99% slippage, ultimately receiving only about 327 AAVE (worth approximately $36,000), instantly evaporating nearly $50.39 million!
Aave founder Stani Kulechov posted a response: The interface had repeatedly shown “extraordinary slippage” warnings and required users to check a box to confirm the risk. The user clearly accepted this on a mobile device before executing the transaction. The transaction was routed through CoW Swap and operated as designed, but the result was “clearly not ideal.”
Similar incidents are not uncommon in DeFi history. Between 2024 and 2025, multiple classic “sandwich attacks” + high slippage errors occurred on Uniswap: for instance, a user faced a significant swap due to high slippage tolerance + shallow pools, being precisely attacked by an MEV bot from both sides, resulting in a loss of about $700,000; another large memecoin pool transaction directly crashed the price, with slippage consuming hundreds of thousands to millions of dollars, leaving only scraps.
These cases are analogous to Aave's incident: Code is Law executed ruthlessly, and users' one-click “confirm” is irreversible. No matter how prominent the warning, continuous clicking on the phone still leads to tragedy.
A moment's mistake can cause funds to instantly go to zero.
On March 13, 2026, in the early morning, a user attempted to exchange approximately $50.43 million worth of aEthUSDT (the interest-bearing token for USDT deposited in Aave) for aEthAAVE through the Aave official interface. As a result, the order size far exceeded the liquidity, triggering an extreme price impact and >99% slippage, ultimately receiving only about 327 AAVE (worth approximately $36,000), instantly evaporating nearly $50.39 million!
Aave founder Stani Kulechov posted a response: The interface had repeatedly shown “extraordinary slippage” warnings and required users to check a box to confirm the risk. The user clearly accepted this on a mobile device before executing the transaction. The transaction was routed through CoW Swap and operated as designed, but the result was “clearly not ideal.”
Similar incidents are not uncommon in DeFi history. Between 2024 and 2025, multiple classic “sandwich attacks” + high slippage errors occurred on Uniswap: for instance, a user faced a significant swap due to high slippage tolerance + shallow pools, being precisely attacked by an MEV bot from both sides, resulting in a loss of about $700,000; another large memecoin pool transaction directly crashed the price, with slippage consuming hundreds of thousands to millions of dollars, leaving only scraps.
These cases are analogous to Aave's incident: Code is Law executed ruthlessly, and users' one-click “confirm” is irreversible. No matter how prominent the warning, continuous clicking on the phone still leads to tragedy.
A moment's mistake can cause funds to instantly go to zero.
🎙️ 1
End
06 m 33 s
42
0
🎙️ 1
End
05 m 45 s
59
0
Perplexity AI has launched 'Personal Computer'!
A locally hosted AI system that is always online, running on a Mac Mini, with 24/7 access to files, applications, and chat records, emphasizing personal privacy and security, allowing control from any device at any time. Today, the initial waitlist is open, and joining will get you support.
This is clearly aimed at OpenClaw. OpenClaw has become popular as an open-source solution, with users setting it up on their local machines, granting extensive permissions, making it capable of real work, but also prone to security issues, resulting in some being 'backstabbed', either burning API money or maxing out hardware.
This version from Perplexity seems like a 'secure upgraded version of OpenClaw': local execution + company-backed security mechanisms, user action confirmations + audit logs, eliminating the need for users to struggle with vulnerabilities, making it more reassuring for the average person to use.
Curious about who is actually more practical: Perplexity's worry-free solution, or the full freedom offered by OpenClaw? How do you choose?
A locally hosted AI system that is always online, running on a Mac Mini, with 24/7 access to files, applications, and chat records, emphasizing personal privacy and security, allowing control from any device at any time. Today, the initial waitlist is open, and joining will get you support.
This is clearly aimed at OpenClaw. OpenClaw has become popular as an open-source solution, with users setting it up on their local machines, granting extensive permissions, making it capable of real work, but also prone to security issues, resulting in some being 'backstabbed', either burning API money or maxing out hardware.
This version from Perplexity seems like a 'secure upgraded version of OpenClaw': local execution + company-backed security mechanisms, user action confirmations + audit logs, eliminating the need for users to struggle with vulnerabilities, making it more reassuring for the average person to use.
Curious about who is actually more practical: Perplexity's worry-free solution, or the full freedom offered by OpenClaw? How do you choose?
Why do many people think lobsters are useless?
Are you guys thinking too highly of its functionality? AI trading, a work assistant to help you make money?
To be honest, how well a lobster is trained has a lot to do with its owner; if your skills aren't good enough, if your token isn't sufficient, if you can't make money yourself, then don't expect Opc lobster to make money for you.
The scenario I look forward to is that the lobster changes my life, for example, integrating with the Whoop Apple Watch API, monitoring that I start to wake up, and the lobster automatically orders coffee and breakfast on the food delivery app, so I can eat as soon as I wake up.
For instance, the lobster could be my personal butler, helping me contact the cleaning service every week to schedule cleaning time, helping me organize key information in the group chat, and so on.
Technology changes life, AI changes life; wanting to rely on a lobster trained with a few thousand dollars to help you create wealth is indeed a bit too much to think about 🙃
Are you guys thinking too highly of its functionality? AI trading, a work assistant to help you make money?
To be honest, how well a lobster is trained has a lot to do with its owner; if your skills aren't good enough, if your token isn't sufficient, if you can't make money yourself, then don't expect Opc lobster to make money for you.
The scenario I look forward to is that the lobster changes my life, for example, integrating with the Whoop Apple Watch API, monitoring that I start to wake up, and the lobster automatically orders coffee and breakfast on the food delivery app, so I can eat as soon as I wake up.
For instance, the lobster could be my personal butler, helping me contact the cleaning service every week to schedule cleaning time, helping me organize key information in the group chat, and so on.
Technology changes life, AI changes life; wanting to rely on a lobster trained with a few thousand dollars to help you create wealth is indeed a bit too much to think about 🙃
Just woke up... feels like the sky has collapsed
I heard that posting selfies in the square will get a lot of comments
Is this true, guys?
Is this true, guys?
Today also caught my cousin live streaming at Binance Square @CZ !!
Asked my cousin about his thoughts on the prediction market this year
Cz replied: The prediction market still has a lot of room to grow, and it's still in its early stages, with many aspects needing improvement and exploration. Meanwhile, Yzi Labs is actively investing in prediction market projects on the BNB Chain.
Looks like I need to dive deeper into researching the prediction market soon. I've been studying a lot recently, and I'll share my insights later with everyone!
Also shared his thoughts on the recent hot topic of Vibe Coding
Cz answered: AI can improve code quality and better assist creators in production, but inevitably, most vibe coding is useless 🤣
@币安广场
Asked my cousin about his thoughts on the prediction market this year
Cz replied: The prediction market still has a lot of room to grow, and it's still in its early stages, with many aspects needing improvement and exploration. Meanwhile, Yzi Labs is actively investing in prediction market projects on the BNB Chain.
Looks like I need to dive deeper into researching the prediction market soon. I've been studying a lot recently, and I'll share my insights later with everyone!
Also shared his thoughts on the recent hot topic of Vibe Coding
Cz answered: AI can improve code quality and better assist creators in production, but inevitably, most vibe coding is useless 🤣
@币安广场
Login to explore more contents