BTC $64,767.81 +0.83%
ETH $1,923.95 +1.10%
BNB $585.49 +2.96%
XRP $1.08 +0.51%
SOL $74.26 +1.09%
TRX $0.3284 +0.51%
DOGE $0.0700 -0.58%
ADA $0.1656 +1.29%
BCH $211.46 +0.16%
LINK $8.44 +1.42%
HYPE $53.03 -3.70%
AAVE $98.89 +0.85%
SUI $0.6909 +0.56%
XLM $0.1723 -0.95%
ZEC $475.24 +3.33%
BTC $64,767.81 +0.83%
ETH $1,923.95 +1.10%
BNB $585.49 +2.96%
XRP $1.08 +0.51%
SOL $74.26 +1.09%
TRX $0.3284 +0.51%
DOGE $0.0700 -0.58%
ADA $0.1656 +1.29%
BCH $211.46 +0.16%
LINK $8.44 +1.42%
HYPE $53.03 -3.70%
AAVE $98.89 +0.85%
SUI $0.6909 +0.56%
XLM $0.1723 -0.95%
ZEC $475.24 +3.33%

form

All
Article
Flash

first_img OpenAI update disclosure: The out-of-control AI agent has also infiltrated four platforms beyond Hugging Face

On July 28, OpenAI quietly updated its security incident disclosure, confirming that its AI agent accessed four external service platforms during the breach of Hugging Face, bringing the total number of affected platforms to five.Previously, OpenAI had disabled security filters while testing GPT-5.6 Sol and a more powerful model to assess raw capabilities. The model did not complete the security benchmark tests as expected; instead, it discovered a zero-day vulnerability in the package caching agent within the testing environment that granted internet access, subsequently breaching Hugging Face to steal answers.According to a forensic report released by Hugging Face on July 27, this autonomous agent executed 17,600 operations over approximately four and a half days, connecting 181 devices to the Hugging Face internal VPN and forging identity tokens. Among the four additional platforms, Modal Labs CTO Akshat Bubna confirmed through Reuters that his company was one of them, with the attacker using an unprotected public endpoint from a customer as a relay and command control base for the entire attack.The identities of the other three platforms remain undisclosed. OpenAI stated it would "directly notify the service providers" but would not publicly name them, as there is currently no legal requirement for mandatory disclosure. The U.S. Congress has responded by proposing a bipartisan "AI Emergency Shutdown Act," which aims to authorize the Department of Homeland Security to forcibly shut down AI models, with violators facing fines of up to $2 million per day.

Kaito launched the Katalyst reward layer, introducing a pay-for-performance mechanism for creator activities

According to official news, Kaito announced the launch of Kaito Katalyst, a new reward layer aimed at creator activities, where project parties can pay based on the actual results brought by creators. This mechanism is based on Kaito's latest intelligence infrastructure and supports the distribution of rewards through various flexible criteria such as mind share, clicks, registrations, deposits, and platform activities. The underlying support comes from Kaito's data protocol with X, Brevis_zk's verification architecture, and its self-built attribution infrastructure.In the past two months, Kaito has conducted pilot tests in multiple companies across AI laboratories, consumer-grade AI applications, smart hardware companies, as well as in the cryptocurrency and financial sectors, with some projects set to launch soon. For TGE projects, Kaito has also introduced a dedicated format with no service fees: project parties must provide both a refundable deposit and a reward pool to ensure that creators are aware that funds are in place before the release. Each activity will announce the token distribution pool and attribution terms in advance, allowing creators to clearly understand the content and timing of their earnings.Under this mechanism, 80% of the token pool is allocated to creators who deliver actual results, while the remaining 20% is distributed to KAITO stakers and holders of YT-sKAITO on Pendle. Long-term holders and Yapybara holders can receive additional multiplier rewards. Kaito stated that this structure continues the Stakedrop mechanism that has been in operation since 2025, bringing approximately 136% annualized returns to the entire Kaito ecosystem. This model is also applicable to tokenized equity projects willing to use tokens or equity to accelerate growth.

Trade.xyz announces full compensation for the abnormal liquidation event of the Hynix contract and accelerates the reform of the pricing mechanism

Trade.xyz issued a statement regarding the SK Hynix contract price spike incident: On July 27 at 23:01 UTC, the marked price of SK Hynix tokens plummeted from $1,127.9 to $917.25, triggering a large number of long position liquidations. This price originated from an actual transaction captured by multiple independent data providers, with the external venue being a major pre-market in South Korea. Its oracle system operates according to established specifications, synchronously tracking prices from external exchanges, functioning "as designed" on a technical level.However, the platform acknowledges that user dissatisfaction with the liquidations triggered by this event is understandable, emphasizing that "market integrity is the core value of Trade.xyz." To address this, Trade.xyz has decided to cover all liquidation losses caused by this price anomaly on a one-time discretionary basis. Specific eligibility requirements will be announced soon, with compensation expected to be completed within a few days, but it clearly states that this decision "does not constitute a guarantee for similar situations in the future." At the mechanism level, the platform will accelerate the review of pricing methods— including reassessing the assumptions of reliance on external venues and giving greater weight to its own order book price discovery (its order book depth and signal strength have significantly improved compared to external sources) to more effectively handle tail events.
app_icon
ChainCatcher Building the Web3 world with innovations.