Arbitrum officially announced that users can now privately deposit, send assets, and query balances through the privacy protocol NullMask.[Source]
11/1, 10:05
OKX Founder Star: Will Strengthen Security Self-Audits, Already Initiated LLM Penetration Testing Over a Year Ago
OKX founder Star addressed user concerns regarding exchange security, pledging to strengthen self-inspections, and disclosed that OKX had implemented large lang...[Source]
10/22, 04:08
Slowmist: LiteLLM Targeted Again by Supply Chain Attack, Malicious Version 1.107.0 Can Steal Cloud Credentials and AI API Keys
SlowMist monitoring reveals that LiteLLM has once again been targeted by a supply chain attack. The malicious version litellm@1.107.0 affects PyPI packages and ...[Source]
10/16, 10:51
Vitalik Launches "Paranoid Mode" Privacy Solution for AI Agents, Integrating zkAPI, Tor, and Local LLM Prompt Rewriting
Ethereum co-founder Vitalik Buterin (@VitalikButerin) posted on X, integrating zkAPI, Tor, additional cover traffic, and local LLM prompt rewriting into a daemo...[Source]
Ostium Announces Phase Two Compensation Plan: Offering USDC Installment Payouts and Equity Conversion Options
Ostium has released an update regarding the design of the Phase II loss compensation plan for Liquidity Providers (OLP). This plan offers two compensation mecha...[Source]
9/26, 21:57
Iran's Foreign Minister says the opening of the Strait of Hormuz depends on the fulfillment of conditions.
Iran's Foreign Minister Araghchi stated that the opening of the Strait of Hormuz depends on whether his proposed conditions are met, and that Iran will not make...[Source]
9/21, 06:43
Alibaba appoints Liu Dayiheng as the new head of the Qwen LLM team.
According to The Information: Alibaba has appointed Liu Dayihen as the new head of the Qwen LLM team. (Jin10)
9/3, 14:57
Meta Releases CORAL Framework to Optimize Recommendation Systems Using Closed-Loop LLM Agents
Meta published the paper CORAL, proposing an LLM-native agent harness for closed-loop optimization of production-grade recommendation systems. Through closed-lo...[Source]