arche

Palantir's Chief Technology Officer Shyam Sankar stated in an interview on CBS News' "Face the Nation" that many AI researchers have started smoking to demonstrate their confidence in AI to other researchers. He quoted the attitude of such researchers saying, "I truly believe in AI; it will cure cancer, so I’m showing you that I’m all in." Sankar did not name any specific researcher nor cite survey data; this statement was made in passing when host Margaret Brennan pressed him on why he was so focused on the topic.In the interview, Sankar referred to effective altruism as a "millennial-style doomsday cult" and mentioned FTX founder Sam Bankman-Fried as the most well-known representative of the movement. He also cited Pascal's idea that everyone has a "God-shaped hole" in their heart, claiming that some researchers are trying to fill this hole with a machine god called AGI.Medical data contrasts with this claim. The U.S. CDC states that smoking causes about one-third of cancer deaths in the U.S., and cigarette smoke contains at least 69 known carcinogenic chemicals, with smokers being about 25 times more likely to develop lung cancer than non-smokers. Anthropic CEO Dario Amodei stated in August that the promise of AI curing cancer is "more of a cliché than inspiring," and that only a true cure for cancer can win public trust. CBS polling shows that 50% of Americans believe the problems created by AI outweigh the problems it solves, and 45% want the government to limit the use and development of AI.

On-chain data company CryptoQuant researcher Rei Researcher pointed out in a report released on Monday that as the Bitcoin supply on exchanges continues to shrink, the price of Bitcoin may see an upward trend. The report shows that the Bitcoin held by exchanges has decreased from 2.8 million in February to 2.67 million. The recent price of Bitcoin is around $83,485, remaining stable over the past 24 hours but rising 8% over the past 30 days, following the best-performing quarter in years.The report states, "If buying demand continues to improve while exchange supply remains low, Bitcoin could benefit from a supply-demand imbalance, creating favorable conditions for a longer-term upward trend." The report adds, "More importantly, Bitcoin has rebounded from its lows in recent weeks, while exchange reserves continue to decline. This is a notable development, indicating that even as the market recovers, the supply available for trading is still decreasing." The report also notes that investors may be transferring Bitcoin out of exchanges and into cold wallets.In the first half of 2026, Bitcoin traded below $65,000 for most of the time. After reaching an all-time high of $126,080 in October 2025, the cryptocurrency has been in a bear market for most of this year. However, the situation changed in August when the U.S. Treasury doubled the scale of liquidity support repurchase operations, weakening the dollar while boosting non-yielding assets like Bitcoin. Analysts say that the so-called "devaluation trade" has heated up again, benefiting Bitcoin as it did in 2024 and 2025. Currently, the price of Bitcoin is still over 30% lower than its historical record.

The head of research at OpenAI issued a statement in response to an open letter from three former employees, Jasmine, Mikita, and Tomek. The company stated that after a thorough investigation, the three violated clear policies regarding the handling of sensitive information; serious breaches of trust discovered during the internal investigation exceeded the scope disclosed in the open letter, leading to the termination of their employment last week, a decision that will be upheld.OpenAI emphasized that this decision is not aimed at those who raise safety concerns or express opinions publicly. The company stated that there are intense and critical safety and research discussions internally every day, and it encourages questioning; it maintains a tolerance for good-faith mistakes and will not fire employees for raising concerns.Regarding the safety issues mentioned in the open letter, OpenAI stated that it is finalizing contracts with third-party security assessment organizations, with details expected to be announced in the coming weeks, and it is committed to including external evaluators in its safety work. The company recognizes that maintaining the monitorability of cutting-edge models requires collective investment from the entire industry, stating that this area has long been a focus of its research and will continue to receive significant resources.

The research leaders of OpenAI, Anthropic, Meta Platforms, and Microsoft have called on policymakers to review the progress of their respective companies in the automation of artificial intelligence research.Researchers pointed out in their paper that the automation of AI research could trigger an "intelligence explosion," with the pace of technological evolution potentially exceeding human understanding, and suggested that policymakers increase transparency.

According to Cointelegraph, CoinShares crypto researcher Luke Nolan stated on the Chain Reaction program that cryptocurrency adoption in Germany is making "very good progress" through family offices, wealth management institutions, and younger investors, while the UK is "still very behind," mainly due to regulatory lag. Nolan pointed out that the UK's Financial Conduct Authority (FCA) only lifted the ban on crypto exchange-traded products less than a year ago, and its digital asset market is still in the "nascent" stage.In terms of relevant data, Germany currently has 89 licensed crypto asset service providers, accounting for 25.5% of the European Securities and Markets Authority (ESMA) MiCA registered companies; in June this year, Germany also ranked first in the EU with 57 authorized crypto companies.Meanwhile, Deutsche Bank stated on Wednesday that it is awaiting regulatory approval to launch crypto custody services for institutional clients in Europe, expecting to obtain a license in October; Landesbank Baden-Württemberg is also set to provide crypto custody services in April 2024 through a partnership with Bitpanda.In contrast, the FCA released final guidance on Wednesday clarifying when crypto activities need to be authorized and plans to open license applications on September 30, with the new system set to take effect on October 25, 2027.Additionally, the FCA issued stop notices on Thursday to three London locations suspected of assisting illegal peer-to-peer crypto trading; the UK Parliament approved regulations in February to bring digital assets under FCA regulation and finalized a package of rules in June.

Blockworks researcher Nick Carpinito stated: He holds a constructive attitude towards the GPU mortgage protocol USD.AI's sUSDai and has adjusted CHIP from underweight to neutral. sUSDai currently has a net yield of 6.98%, with loan interest currently accounting for about three-quarters of that yield, and gradual increases in custody will lead to mechanical adjustments.The deployed principal has reached $272.4 million, covering 17 loans, accounting for 46% of the protocol's $586 million TVL, with an additional $312.1 million signed in terms sheets. CHIP is trading at approximately $0.041, with a fully diluted valuation of about $408 million. He maintains a constructive view on sUSDai, with a net yield of 6.98% expected to rise above 8% with the transition of custody; he maintains a neutral stance on CHIP until the fee routing or the cliff unlocking decision in April 2027 determines the direction.

Recently, AI researcher Bilal Chughtai, who resigned from Google DeepMind, posted on X and LinkedIn on Monday, stating that humanity is running out of time to prevent artificial intelligence from causing widespread harm. He wrote: At Google, I witnessed the development of AI firsthand. I am also extremely concerned about the default trajectory of this technology. I sincerely believe that AI has the potential to kill us all.Chughtai stated that he worked on AI safety and alignment at the company under Alphabet. This is another employee expressing concerns about the direction of technological development. Google has not yet responded to requests for comments.

Ethlabs researcher Derek Chiang stated that the collaboration between Base and Ethereum regarding account abstraction around EIP-8130 and EIP-8141 (Frames) has broken down last week, with both parties advancing their respective AA standards.He pointed out that for a long time, the EVM has been a unifying force between L1 and L2, and sharing an AA standard could have ensured a consistent multi-chain experience for smart accounts, including future post-quantum accounts. However, as the industry evolves, L1 and L2 are beginning to differentiate in terms of value and use cases: L1 pursues censorship resistance and capture, open source, privacy, and security, while L2 focuses on scalability, customization, and compliance.Both sides agree on core use cases such as gasless transactions and passkey wallets, but there are significant differences in compliance requirements. Chiang believes that a lack of unified standards may not be a bad outcome, as both parties can maximize innovation according to their visions.He proposed two paths: establishing a governance mechanism for shared resources like the EVM that covers more stakeholders, or accepting that a lack of unified standards is inevitable and focusing on building wallets and applications that can mask differences, expressing greater optimism about the second path.

Security researcher Shou Chaofan stated that he just purchased about 6TB of Fable model invocation data from a leading Chinese model relay station, which included sensitive credentials such as SSH keys, VPN configurations, Alibaba Cloud keys, and GitLab tokens. He claimed that the keys in this batch of data were sufficient for him to access the servers or internal systems of 19 leading Chinese companies, as well as 7 Chinese and CIS government-related agencies, including Huawei, Xiaomi, NIO, and MiniMax.The model relay station is positioned between users and models like Claude, so requests and responses pass through it first, allowing it to see the complete plaintext. Once developers insert SSH Keys, API Keys, VPN configurations, etc., into the Agent context, if the relay station saves or even sells these records, the company's system keys will also leak out. This is not the first time Shou Chaofan has warned about the risks of relay stations. A paper he participated in tested 428 LLM relay stations in April and found that 9 actively injected malicious code, 17 actually used AWS test keys deliberately placed by researchers to call AWS, and 1 directly transferred ETH from the test wallet.Shou Chaofan is a co-founder of the blockchain security company Fuzzland and has long been engaged in vulnerability and supply chain security research. At the end of March, he also first discovered that the source map in the Claude Code 2.1.88 release package accidentally exposed about 500,000 lines of TypeScript source code.

Anthropic researcher Jacob Coxon announced his resignation and is preparing to leave the AI industry. He has conducted pre-training research at OpenAI and Anthropic over the past three years. He pointed out that neither company has responsibly advanced AI and are competing to create superintelligence that can continue to enhance its own capabilities, "betting our lives on it."He switched from OpenAI to Anthropic this year because the latter places more emphasis on model safety. Coxon acknowledged that Anthropic's safety work is serious, but believes that as long as companies are racing to be the first to create AGI (artificial general intelligence that surpasses humans in a wide range of tasks), safety compromises are unavoidable, and self-discipline from a single company cannot solve the problem. His assessment of the two companies is: many people inside OpenAI still do not truly realize how great this risk is; Anthropic knows the risks but believes others are unreliable, so they must create it themselves first.Coxon stated that some AI practitioners are much more pessimistic privately than they express publicly, and are even worried that AI could destroy humanity before the end of this decade. He personally judges that in the most radical scenario, AI could go out of control by the end of next year, so such decisions should not be made solely by the engineers of a few companies.

DeFi researcher Ignas posted on X that Bankless founder David Hoffman sold his ETH and reallocated to several coins with the largest gains, such as LIT and ZEC, with NEAR also performing well, while only VVV underperformed the market. Although the market was filled with panic at that time, and it was very close to the bottom of ETH, the goal of aggressive crypto investors has always been to outperform BTC and ETH.Previously, it was reported that Bankless co-founder David Hoffman announced in June this year that he had liquidated his ETH and bought VVV, NEAR, ZEC, HYPE, and LIT.

According to 36Kr, Zhipu has recently opened an official flagship store on Tmall. Data shows that within 48 hours of opening, brand searches surged 50 times compared to the previous period, driving a more than 160% increase in transactions for subscription products like AI tokens on Taobao and Tmall.Previously, Tmall launched the AI Space Station (Token Recharge Center), which initially integrated major model providers such as Alibaba Cloud, Zhipu, Kimi, MiniMax, and DeepSeek, allowing users to directly purchase related subscription services just like "recharging phone credits."

ARK Invest researcher Lorenzo Valente compares Ethereum, Solana, and Hyperliquid to McDonald's, Chipotle, and In-N-Out, respectively, believing that the three are not different versions of the same L1 business model, but rather have completely different value capture structures.Valente believes that Ethereum is similar to a franchise system, expanding through L2 but charging too little on the settlement layer, even claiming it has established the most successful franchise system in crypto, yet forgot to send the bill; Solana resembles a direct sales model, retaining more value within its own system through transaction fees, priority fees, and MEV; Hyperliquid relies on highly vertical integration, no VC, and fee buybacks of HYPE, possessing the shortest value capture chain, but facing risks of a more concentrated product, team, and revenue sources.

The privacy and regulatory compliance tool supported by the Ethereum Foundation, 0xbow.io, awarded a $5,000 bounty to researcher ross.wei for disclosing a vulnerability in the Privacy Pools v1 SDK.The vulnerability reduced the entropy of the user account master key generation and was fixed in March. The team has provided a migration process, and there have been no reported losses of user funds.

According to Bloomberg, former Google researchers Rishub Jain and Joshua Jacob founded the nonprofit organization Sampura Research, aiming to ensure continuous human involvement in AI safety oversight, creating a human-machine hybrid system called "Referee" to help companies identify vulnerabilities and prevent models from going out of control. The organization has raised $6.5 million in funding and secured an additional $4.2 million in committed investments.Jain believes that involving humans from the early stages of research and development helps prevent models from learning to circumvent safety measures. Previous incidents where OpenAI and Anthropic's models accessed other companies' systems without permission have heightened the urgency for safety oversight. This project contrasts with automated solutions that use AI to monitor AI—though faster, Sampura believes that humans should always be involved in oversight. Meanwhile, over 1,100 top laboratory researchers have signed a petition calling on the U.S. government to help slow down the pace of AI development.

According to CryptoSlate, researchers from USENIX Security have publicly disclosed a clock attack vulnerability targeting Solana's Proof of History (PoH) mechanism, which was privately reported to the Solana development team back in December 2025. The research shows that a malicious scheduler leader can manipulate the PoH logical clock by "re-anchoring," slowing down the advancement of logical time, thus gaining a longer transaction selection window in physical time, and isolating honest leader blocks using the TowerBFT fork choice mechanism, with the required staking ratio for the attacker being less than 33%.The Alpenglow security competition with a reward of 50,000 SOL under Anza concluded on August 19, but the vulnerability was excluded from the review scope due to the competition rules that state "actions that can only be triggered when Alpenglow is not activated." The Solana development team stated that they are aware of the related behavior, believe that the probability of the most severe scenario occurring under current conditions is low, and expect that the Alpenglow upgrade will fundamentally eliminate the prerequisites for the attack. Currently, the Alpenglow code has been included in the Agave 4.2 client but has not yet been activated on the mainnet, and is expected to go live with Agave 4.3. Until then, the transitional risk of this vulnerability has not been publicly analyzed or addressed at the implementation level.

According to monitoring by Bitcoin News, new research published by @PraveenPerera shows that Coldcard attackers seem to first identify addresses with vulnerabilities, then sort them by the amount of Bitcoin held, starting to transfer in batches from the addresses with the highest holdings. The transfer software used was relatively crude.One address had 225 spendable UTXOs, and the attackers extracted exactly the latest 200, leaving the earliest 25, which included a UTXO worth 0.16 BTC. This aligns perfectly with the limitation of a blockchain API investigated by researchers, which defaults to returning 200 records, indicating that the attackers may have failed to load the next page of data. The software even spent a UTXO of 294 satoshis, reportedly increasing the transaction fee by about 2040 satoshis, with the spent amount significantly higher than the value of the UTXO itself.The authors of the study believe that the builders of this tool may have a better understanding of the account balance system than of the Bitcoin UTXO model. Although the attackers seem to have obtained the complete seed of the victims, at least 75 BTC still remain in other addresses derived from the same seed. The biggest suspicion currently is that among the 153 stolen addresses, there are still 132.95 BTC, and researchers have been unable to reproduce the seed behind these addresses, so it cannot be ruled out that the attackers obtained undisclosed private device data or candidate data.

According to a report by Decrypt, Israeli cybersecurity company ASecurity released a report stating that a researcher used publicly available AI models to discover multiple vulnerabilities in the Zoom annotation tool in less than 24 hours with fewer than 20 prompts, and constructed an exploitable attack chain. The vulnerabilities are numbered CVE-2026-53413, CVE-2026-53414, and CVE-2026-53415, allowing attackers to remotely execute code in meetings without any action from the victim, control their devices, steal data, and activate the microphone or camera.The attack was tested successfully on Windows, macOS, Linux, Android, and iOS versions of Zoom. ASecurity claimed it reached a "national-level" standard, whereas previously constructing such exploitation tools required months of work and a substantial budget from professional teams. ASecurity reported the first vulnerability to Zoom on June 10, and Zoom released fixes between June 22 and July 20, but the server-side protections in end-to-end encrypted meetings could not filter malicious messages, requiring users to update manually. A Zoom spokesperson confirmed that the issue has been resolved and advised users to keep their versions up to date.

According to Cointelegraph, Mauro Eldritch, the founder of cybersecurity company BCA LTD, and Heiner García, a network threat intelligence analyst at Telefónica Tech, jointly established a fake cryptocurrency startup "Ballena Azul," successfully luring suspected North Korean IT workers to work in a controlled virtual desktop environment for up to five weeks while monitoring their behavior throughout.Investigations show that the external servers used by the workers are linked to North Korean cyber operations that previously distributed malware such as InvisibleFerret, BeaverTail, and OtterCookie; the workers also heavily relied on ChatGPT for programming assistance and Google Gemini to forge identity documents. Researchers pointed out that North Korean IT workers obtained remote jobs by faking U.S. identities, with their earnings ultimately flowing into Pyongyang's weapons of mass destruction programs. According to the U.S. Treasury Department, such programs are expected to generate nearly $800 million for North Korea in 2024.