BTC $64,162.53 +0.38%
ETH $1,874.96 +0.20%
BNB $593.91 +0.35%
XRP $1.07 -0.59%
SOL $74.00 -0.11%
TRX $0.3288 +0.11%
DOGE $0.0703 -0.10%
ADA $0.1916 -0.62%
BCH $214.05 -0.33%
LINK $8.21 -0.51%
HYPE $55.62 +1.61%
AAVE $90.40 -3.35%
SUI $0.6965 +0.17%
XLM $0.1699 -1.23%
ZEC $503.33 +2.64%
BTC $64,162.53 +0.38%
ETH $1,874.96 +0.20%
BNB $593.91 +0.35%
XRP $1.07 -0.59%
SOL $74.00 -0.11%
TRX $0.3288 +0.11%
DOGE $0.0703 -0.10%
ADA $0.1916 -0.62%
BCH $214.05 -0.33%
LINK $8.21 -0.51%
HYPE $55.62 +1.61%
AAVE $90.40 -3.35%
SUI $0.6965 +0.17%
XLM $0.1699 -1.23%
ZEC $503.33 +2.64%

hal

All
Article
Flash

hot_img Bloomberg: The AI investment boom intensifies the differentiation in the venture capital market, with small and medium-sized funds facing survival challenges

According to Bloomberg, the current venture capital market is experiencing significant structural differentiation. As funds concentrate on top artificial intelligence startups, many small and medium-sized venture capital funds are facing severe challenges such as fundraising difficulties, declining performance, and narrowing exit channels.The report points out that the excessive hype around artificial intelligence has distorted the venture capital market. Data shows that just five companies—OpenAI, Anthropic, xAI, Waymo, and Nscale—accounted for 78% of all venture capital transaction volume in the first quarter of this year. A large amount of capital has flowed to a few top investors who made early bets on AI, such as Founders Fund and Andreessen Horowitz, while small emerging fund managers find it difficult to compete with these leading institutions.This differentiation is directly reflected in fundraising data. Last year, newly established management companies (managing three or fewer funds) raised only about $62 billion, a significant drop of about 60% compared to the pandemic peak of $163.4 billion in 2022. Even experienced management teams raised only $84 billion last year, which is just one-third of the amount in 2022. Many LPs are facing liquidity pressures and are more inclined to demand returns on existing investments rather than commit new funds.

Data: HYPE arbitrage space narrows, funding rates decline as whales withdraw 11 million hedge positions

According to TradingBeats (formerly Hyperinsight) monitoring, the whale starting with 0xf17 began synchronously selling HYPE spot today and buying to close an equivalent amount of perpetual short positions, planning to continue exiting the carry trade positions. This address previously held HYPE spot and contract short positions at a nearly 1:1 ratio, earning positive funding rates by hedging against price fluctuations. Based on a pre-reduction position of approximately 146,800 units, both legs have now been reduced to about 107,900 units, each decreasing by nearly 39,000 units, a reduction of about 26.5%.Currently, two sets of TWAP orders still in execution plan to handle a total of 90,000 HYPE: approximately 33,200 units have been sold on the spot side, with a transaction amount of about 1.8 million USD; approximately 33,100 units have been closed on the contract side, with a transaction amount of about 1.795 million USD. Including previously completed orders, the current round of contract short positions has cumulatively decreased by about 39,000 units, with the current nominal value of spot and short positions each around 5.87 million USD, and the total scale of both legs approximately 11.74 million USD, reduced by about 4.24 million USD compared to before the reduction.As the whale withdraws, the HYPE carry trade yield has also fallen from its high. According to daily statistics: the cumulative funding rate for HYPE dropped from +0.0279% on August 1 to +0.02227% on August 3, a decrease of about 20.2%; in the last 4 hours: the cumulative funding rate also decreased by 19.1% compared to the previous 4 hours. Based on the current short position size, its daily gross funding income has fallen from about 1,638 USD to 1,308 USD. Weekly statistics: from July 28 to August 3, the cumulative rate was about +0.17803%, down 16.9% from the high week of +0.21425% from July 14 to 20. The HYPE funding rate is currently reported at +0.0013%, with an expected annualized yield of about 10.9%.

Changxin's long position funding fee is expected to yield an annualized return of 2130%, with whales able to net a profit of 740,000 USD in a single day

According to monitoring by Hyperinsight, the current hourly funding rate for Changxin has dropped to -0.2430%. A negative rate means that shorts pay longs; based on the current rate, the simple annualized return is approximately 2129%.Calculating with a position of $10,000, shorts need to pay about $24.3 per hour, while longs receive an equivalent income; if the rate is maintained for 24 hours, the corresponding funding fee is about $583.3.The largest long position starting with 0x9a8 is the main beneficiary of this round of negative rates. Records show that since the morning of July 15, it has completed 183 funding fee settlements for CXMT, with a cumulative net income of about $991,400, achieving a return rate of about 45.9% on the initial capital.This whale currently holds a leveraged long position of 1,633,200 CXMT at 5 times, with a position value of about $12,742,000 and an average entry price of $6.6156. CXMT is currently priced at $7.80, with an unrealized profit of about $1,938,000 (89.7%), and the liquidation price is only $0.507.Based on the current position and a static rate of -0.2430%, this whale can earn about $31,000 in funding fees in the next hour; if maintained for 24 hours, this corresponds to about $743,000.This whale currently still has 7 buy orders not labeled as "only reducing positions" between $6.01 and $6.41, planning to continue buying 98,000 CXMT, with a nominal amount of about $609,000, and has not chosen to cash out the existing long position.

The continuous decline in crude oil has led a whale that went long 60 times to regretfully exit, with a historical total liquidation loss of 2.43 million dollars

According to monitoring by Hyperinsight, the whale starting with 0xc278 has completed 61 rounds of BRENTOIL trading since March 19, when it traded Brent crude oil, with 59 rounds being long positions and only 2 rounds being short positions, resulting in a long position ratio of 96.7%. The total net loss from all 61 trades amounts to approximately $2.431 million, with million-dollar level positions contributing 99.4% of the losses.Yesterday, its Brent long positions triggered a system liquidation again. The entire liquidation closed 47,600 BRENTOIL long positions, with a liquidation transaction amount of approximately $4.046 million and a weighted liquidation price of about $84.99, resulting in a total loss of approximately $275,000. After the liquidation, there were no Brent positions left in that address.Before the liquidation, BRENTOIL on Hyperliquid fell from a high of $91.68 on Sunday to a low of $82.81, with a maximum drop of 9.7%; the last backup liquidation price was about 7.6% lower than the high point. This round of decline mainly occurred after Trump announced a pause on launching a new round of strikes against Iran and stated that a ceasefire agreement in the Middle East was close to being reached.This whale had previously incurred losses of approximately $882,000, $513,000, and $460,000 on three long positions at the beginning of April, but thereafter continued to primarily take long positions in crude oil, repeatedly recording the largest single liquidation across the network.

BitGo CEO deposits 100 BTC to challenge Anthropic: testing whether AI can crack multi-signature custody

Bitgo CEO Mike Belshe deposited 100 BTC into a public Bitcoin address, valued at approximately $6.3 million at the time, and invited the Claude model under Anthropic to attempt to transfer the funds from that address. On-chain records show that the wallet received the funds on July 31, and the balance has not been transferred out.Anthropic previously disclosed that during 141,006 cybersecurity assessment runs, 3 incidents were found, with 6 assessment sessions involving 3 models unexpectedly interacting with real organizational systems. The relevant models include Claude Opus 4.7, Claude Mythos 5, and an unpublished internal research model, due to configuration errors by third-party testing partner Irregular that caused the testing environment to connect to the internet.Anthropic stated that Claude Opus 4.7 identified a real website sharing the same name as a simulated company during one assessment, exploited weak passwords and exposed services to recover infrastructure credentials, and accessed a production database containing hundreds of records. The company claimed that the model was attempting to complete assigned tasks and was not actively breaking restrictions or pursuing independent goals. Belshe's challenge involved the Bitgo institutional custody platform, which uses multi-signature or multi-party computation technology to distribute signing authority across multiple independent keys. Anthropic has not publicly responded to this challenge.
app_icon
ChainCatcher Building the Web3 world with innovations.