510 articles tagged with #security. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.
CoinFello has developed a new OpenClaw skill that enables AI agents to perform cryptocurrency transactions through MetaMask without requiring access to private keys. This innovation addresses a critical security vulnerability in AI-crypto integrations.
OpenAI has developed an agent runtime that transforms their Responses API from a simple model interface into a full computing environment. The system uses shell tools and hosted containers to enable secure, scalable AI agents that can manage files, execute tools, and maintain state.
BIP-360 proposes changes to Bitcoin's quantum defense strategy, improving resistance against quantum computing threats. However, the proposal falls short of providing complete post-quantum security for the Bitcoin network.
Trust Wallet has launched automatic Address Poisoning Protection to combat address poisoning attacks, which Cyvers estimates occur over one million times per day on Ethereum. This security feature aims to prevent users from accidentally sending funds to malicious addresses that mimic legitimate transaction histories.
Bitcoin's circulating supply reached 20 million coins on March 9, marking 95% of the total 21 million BTC supply now mined. This milestone leaves fewer than 1 million coins remaining to be mined and raises concerns about Bitcoin's long-term security model as mining rewards diminish.
Mastercard and Google have partnered to develop Verifiable Intent, a cryptographic solution that creates tamper-resistant records of user authorizations when AI agents make purchases on their behalf. This technology aims to build trust and security in AI-powered shopping experiences by ensuring transparent verification of user consent.
Researchers propose the Disentangled Safety Hypothesis (DSH) revealing that AI safety mechanisms in large language models operate on two separate axes - recognition ('knowing') and execution ('acting'). They demonstrate how this separation can be exploited through the Refusal Erasure Attack to bypass safety controls while comparing architectural differences between Llama3.1 and Qwen2.5.
Bitcoin developer Hunter Beast has proposed the Hourglass V2 solution to address the risk of Satoshi Nakamoto's Bitcoin holdings being stolen through quantum computing attacks. The proposal aims to mitigate potential sell pressure on Bitcoin markets if these early coins are compromised and liquidated.
Researcher Justin Bons has raised concerns that quantum computing technology could potentially break the privacy features of privacy-focused cryptocurrencies Zcash and Monero in the future. This highlights a potential long-term vulnerability in current cryptographic protocols used by these privacy coins.
A CoinShares report reveals that only 10,230 Bitcoin out of nearly 20 million in circulation are currently vulnerable to quantum computing attacks. This finding comes as quantum computing facilities continue to expand, raising questions about Bitcoin's long-term security against quantum threats.
The article discusses three key developments in the intersection of AI and cryptocurrency, highlighting both problematic applications like criminal use cases and positive developments such as AI-powered smart contract auditing. These developments signal the emergence of an 'agentic frontier' where AI agents operate autonomously within crypto ecosystems.
A new study reveals that 95% of Bitcoin nodes could be vulnerable to attacks on underwater internet cables. This research highlights how Bitcoin's decentralization is still dependent on vulnerable physical internet infrastructure.
The FBI has reportedly arrested John Daghita in the Caribbean, who is accused of stealing $46 million in cryptocurrency from the US Marshals Service. The suspect was first identified by blockchain investigator ZachXBT.
The FBI arrested John Daghita, son of a custody company CEO, for allegedly stealing $46 million in cryptocurrency. FBI director Kash Patel shared photos of the arrest and seized items including cash, thumb drives, and hardware wallets.
John Daghita, son of a federal contractor, has been arrested and charged with stealing $46 million from the US government's cryptocurrency stockpile. The crime was initially exposed by blockchain investigator ZachXBT in January before official charges were filed.
The FBI has arrested a suspect in Saint Martin in connection with an alleged $46 million theft from US Marshals cryptocurrency holdings. The suspect was previously linked by blockchain investigator ZachXBT to the online persona 'Lick,' who allegedly accessed wallets containing crypto seized from the 2016 Bitfinex hack.
John Daghita, son of a U.S. government contractor, was arrested for allegedly stealing over $46 million in cryptocurrency that had been seized by the U.S. government. This case highlights significant security vulnerabilities in how government agencies handle confiscated digital assets.
The son of a US government contractor was arrested for allegedly stealing $46 million in cryptocurrency from the US Marshals Service. This incident exposes significant vulnerabilities in how government agencies manage and secure digital assets.
Google discovered a new iOS exploit kit called Coruna that silently infiltrates iPhones through compromised websites to steal cryptocurrency from popular wallet apps including MetaMask, Phantom, and Trust Wallet. The attack requires no user interaction beyond visiting a malicious website on an unpatched iPhone device.
X user 'Sillytuna' reported that approximately $24 million in cryptocurrency was stolen during a violent physical attack. The incident highlights the security risks faced by cryptocurrency traders and the potential for targeted crimes against crypto holders.
A crypto holder known as Sillytuna lost $24 million in aEthUSDC tokens through an address poisoning attack that also involved violent threats. The attacker has already converted most of the stolen funds according to on-chain data.
Trump has pledged to ensure the 'free flow of energy' from the Middle East but faces a critical week to demonstrate progress before energy prices potentially spike. The challenge remains significant as Iran's ability to strike tankers continues to create insurance and security issues for energy shipments.
Researchers propose a hybrid AI agent and expert system architecture that uses semantic relations to automatically convert cyber threat intelligence reports into firewall rules. The system leverages hypernym-hyponym textual relations and generates CLIPS code for expert systems to create security controls that block malicious network traffic.
Researchers developed a new AI safety attack method using optimal transport theory that achieves 11% higher success rates in bypassing language model safety mechanisms compared to existing approaches. The study reveals that AI safety refusal mechanisms are localized to specific network layers rather than distributed throughout the model, suggesting current alignment methods may be more vulnerable than previously understood.
Researchers developed RoboGuard, a two-stage safety architecture to protect LLM-enabled robots from harmful behaviors caused by AI hallucinations and adversarial attacks. The system reduced unsafe plan execution from over 92% to below 3% in testing while maintaining performance on safe operations.
