Technology Trends

Explores the latest innovations, protocol upgrades, cross-chain solutions, and security mechanisms in the blockchain space. It provides a developer-focused perspective to analyze emerging technological trends and potential breakthroughs.

Oaktree Capital's Marks: Investors Underestimate the Impact of AI

Oaktree Capital's Howard Marks warns that investors are significantly underestimating the profound and unpredictable impact of artificial intelligence. He argues that AI's transformative power is accompanied by inherent unpredictability—what it will do, what it won't do, and the extent of human job displacement—making traditional investment strategies based on future predictions insufficient. Citing Block's recent layoff of 4,000 employees as an example, Marks highlights the market's failure to grasp the depth of AI's disruption. He expresses caution about the cyclical nature of technology-driven market exuberance, noting that new innovations often hide risks until it's too late. Regarding investment strategy, Marks strongly advocates for equity over debt financing in AI companies. He believes that if investors are taking on fundamental business model risks posed by AI, they should participate as owners to capture the potential rewards, rather than as fixed-income investors.

marsbit6h ago

Oaktree Capital's Marks: Investors Underestimate the Impact of AI

marsbit6h ago

x402 Wish List

"x402 Wish List" explores the potential of x402 as a micro-payment and data aggregation protocol, based on a Galaxy Research report. It highlights a use case where an AI agent uses x402 to query weather data, book trips, and manage payments to various data sources. The author envisions x402 enabling new business models, such as pay-per-use "Skills Endpoints" for AI agents, replacing fixed fees with usage-based pricing. Other proposed applications include curated data streams for niche crypto news, aggregated ecosystem metrics, a tracker for media prediction performance, and a security/audit incident aggregator. The summary concludes by questioning the economic viability of micro-payments at scale and the legal challenges of data sourcing, noting that x402 V2's revenue-sharing features could help align incentives between data aggregators and original providers.

marsbit19h ago

Stratechery Overturns the AI Bubble Theory: What Should We Use AI For?

Stratechery's Ben Thompson revises his stance on the AI bubble debate, arguing that current AI investment reflects structural growth driven by fundamental technological shifts, not speculation. He identifies three key transitions in LLM development: ChatGPT’s debut (making AI broadly usable but unreliable), OpenAI’s o1 (introducing reasoning and reliability), and the recent emergence of true Agent systems (e.g., Anthropic’s Opus 4.5 and OpenAI’s GPT-5.2-Codex). The critical innovation is the "agent harness"—a control layer that autonomously schedules models, uses tools, and verifies outcomes, reducing human intervention. This transforms AI from an assistive tool into an actionable infrastructure, enabling complex task execution. Agentic systems drive surging demand for compute, as they require repeated model calls and higher usage frequency. Thompson emphasizes that AI demand now depends not on user numbers, but on per-user agent utilization. Enterprises are adopting AI not only for efficiency gains but for structural workforce reduction, as agents amplify the impact of key employees while lowering coordination costs. He concludes that massive capital expenditures are justified by real demand, and profits will flow to integrated model-harness providers rather than commoditized standalone models.

marsbitYesterday 03:57

Stratechery Overturns the AI Bubble Theory: What Should We Use AI For?

marsbitYesterday 03:57

Karpathy Deletes Repository in Emergency! AI Job Apocalypse Chart Goes Viral, 60 Million White-Collar Jobs at Risk

Karpathy's viral project, "karpathy.ai/jobs," analyzed 342 U.S. occupations using AI (Gemini Flash) to assign an "AI exposure" risk score (0-10). The findings revealed an average exposure of 4.9, with 42% of jobs (59.9 million positions) scoring 7 or higher, indicating high automation risk. High-risk roles are primarily screen-dependent and information-dense, including software developers (9/10), financial analysts (9/10), lawyers (8/10), and office clerks (9/10). The most secure jobs involve complex manual labor, such as plumbers, construction workers, and cleaners, which are difficult to automate. Higher-paying and degree-requiring jobs are more significant exposure. Karpathy later removed the project, clarifying it was a quick personal experiment that was overinterpreted. A supporting Harvard study confirms AI is already causing a "great reassignment," reducing hiring in highly automatable roles while increasing it in AI-augmented ones, highlighting a bifurcation in the white-collar job market.

marsbitYesterday 11:44

Karpathy Deletes Repository in Emergency! AI Job Apocalypse Chart Goes Viral, 60 Million White-Collar Jobs at Risk

marsbitYesterday 11:44

Amid the OpenClaw Craze, CEXs Vie for AI Agent Trading Entry Points

Amid the OpenClaw trend, CEXs are competing to become the primary trading interface for AI Agents, rather than passively serving as liquidity providers. The core question is whether exchanges will retreat to the backend or proactively position themselves as the default financial infrastructure for Agents. While many portray the shift from trading apps to dialog-based interfaces as revolutionary, current practical applications are more limited—focusing primarily on research, filtering, alerts, and conditional judgments. Execution remains challenging due to unresolved issues around permissions, confirmation mechanisms, error handling, and liability. Therefore, the initial competition will likely center on perfecting pre-trade functionalities rather than full automation. Exchanges are now racing to modularize their trading capabilities, design secure permission systems, and secure default integration within platforms like Claude, OpenClaw, and ChatGPT. Although it is too early to predict whether this will reshape the industry structure, it is clear that trust and user habits cannot be built overnight. The key takeaway is that the industry is confronting a concrete emerging question: as AI Agents gradually take over parts of the trading workflow, who will become the default crypto financial operating system?

marsbit2 days ago 06:11

Amid the OpenClaw Craze, CEXs Vie for AI Agent Trading Entry Points

marsbit2 days ago 06:11

Daniil and David Liberman: AI is Not Just a Battle of Models, But a Battle of Computing Infrastructure

In the article "Daniil and David Liberman: AI Is Not Just a Battle of Models, but a Battle of Compute Infrastructure," the authors argue that the core of AI development is not just about algorithmic advances but control over computational resources. They emphasize that AI is not a neutral technology—who owns and governs the compute infrastructure ultimately determines who benefits from AI. Currently, advanced AI compute is highly concentrated among a few cloud providers and specific nations, creating a growing "compute divide." This centralization leads to high costs, limited access, and geographic imbalance. Decentralized alternatives, meanwhile, often waste resources on consensus mechanisms rather than meaningful computation. The authors propose a practical alternative: an infrastructure where most compute is used for actual AI work, governance is based on verified computational effort (not capital), and global GPU access is permissionless. They stress that infrastructure choices made today will have long-term economic and geopolitical consequences. For businesses, early reliance on centralized AI infrastructure creates lock-in effects that reduce strategic flexibility over time. The authors warn that waiting too long to explore decentralized options may make transition prohibitively difficult. They conclude that future generations must recognize that AI architecture is a deliberate design choice—not an inevitability—and that open, decentralized infrastructure is essential to preserving fairness, innovation, and freedom in the AI era.

marsbit2 days ago 03:19

Daniil and David Liberman: AI is Not Just a Battle of Models, But a Battle of Computing Infrastructure

marsbit2 days ago 03:19

Which Areas Still Have Moats in the AI Era?

In the AI era, certain moats remain despite rapid technological advancement. The author, a former hedge fund manager, argues that the true inflection point occurred when AI models like ChatGPT’s o1 began generating functional code—even with imperfections—enabling recursive self-optimization and fundamentally altering software development. Key short-term moats identified include: 1. **Proprietary Data**: Firms with unique, inaccessible data (e.g., multi-strategy hedge funds) can fine-tune models, creating defensible advantages. 2. **Regulatory Friction**: Industries requiring human approval (e.g., traditional finance) face slower disruption due to compliance and legal barriers. 3. **Authority-as-a-Service**: Human trust in institutional authority (e.g., legal or audit services) persists even if AI outperforms humans technically. 4. **Physical World Lag**: Hardware-dependent sectors evolve slower, delaying full AI integration. However, these moats only delay, not prevent, disruption. The author emphasizes acting on signals rather than waiting for certainty: identify directional trends, place asymmetric bets (limited downside, high upside), and iterate through action. As AI accelerates, windows of opportunity close quickly. To remain relevant, humans must excel in long-term strategy, complex system-level thinking, and collaboration—areas where AI still lags. The time to act is now, before markets price in the obvious.

marsbit03/15 05:35

Which Areas Still Have Moats in the AI Era?

marsbit03/15 05:35

From 'Collective Intelligence' to 'Super Individuals': How AI is Reshaping DAOs and the Ethereum Ecosystem?

From "Collective Intelligence" to "Super-Individual": How AI is Reshaping DAOs and the Ethereum Ecosystem AI is fundamentally transforming how work and governance are structured in Web3. While DAOs have long symbolized decentralized collective intelligence, AI is now enabling a shift toward the "individual + AI" unit, where a single person, augmented by AI agents, can perform tasks that previously required entire teams—such as research, trading, asset management, and governance. This shift raises a critical question: Is this beneficial for DAOs and the broader crypto ecosystem? AI addresses key DAO challenges like inefficient information processing, complex decision-making, and high participation costs by automating governance processes, analyzing proposals, and executing on-chain operations. This allows DAOs to operate with smaller core teams while significantly improving efficiency. For AI to participate in the on-chain economy, it requires asset custody, transaction execution, and trusted settlement—capabilities native to blockchain. Initiatives like Ethereum Foundation’s dAI team and the ERC-8004 standard aim to establish trust and verification for AI agents in a decentralized context. Wallets are evolving into "Agent Wallets," enabling non-custodial authorizations, cross-chain asset management, and human-AI collaboration through restricted sub-wallets and automated execution within set limits. Ethereum is positioning itself as the financial infrastructure for the AI economy, offering a trusted settlement layer for AI-driven activities. With its growing staking economy and mature DeFi ecosystem, Ethereum could serve as the neutral base where AI agents across platforms settle value and establish trust. In summary, AI and crypto convergence is reshaping organizations and infrastructure: AI amplifies individual capability and automates execution, while blockchain provides secure and decentralized settlement. Ethereum and crypto wallets are poised to become key interfaces connecting humans, AI, and the on-chain world.

marsbit03/14 00:10

From 'Collective Intelligence' to 'Super Individuals': How AI is Reshaping DAOs and the Ethereum Ecosystem?

marsbit03/14 00:10

Vitalik Buterin Argues Ethereum’s Biggest Use Case Is Data Availability

Vitalik Buterin argues that Ethereum's primary value is not smart contracts or payments, but its role as a censorship-resistant "public bulletin board" for data availability. This foundational layer is critical for cryptographic systems like secure voting and version control. Buterin highlights scaling solutions like PeerDAS, which significantly boost data capacity, positioning Ethereum as key infrastructure for a broader, privacy-preserving internet. While payments and smart contracts remain important, they are secondary, often serving anti-spam or security functions. He concludes that Ethereum is best understood as "global shared memory" and that current low fees and improved tooling make this vision more achievable than ever.

bitcoinist03/13 21:02

Vitalik Buterin Argues Ethereum’s Biggest Use Case Is Data Availability

bitcoinist03/13 21:02

Vitalik Wrote a Proposal Teaching You How to Stealthily Use AI Large Models

Vitalik Buterin and Ethereum Foundation's AI lead Davide Crapis co-authored a proposal titled "ZK API Usage Credits," which introduces a method for anonymously accessing AI models like ChatGPT using zero-knowledge proofs. Instead of requiring users to register and link personal information, the proposal suggests depositing funds (e.g., 100 USDC) into a smart contract. Users can then generate a zero-knowledge proof to verify they are on an encrypted whitelist and have sufficient balance—without revealing their identity. This approach aims to protect user privacy while preventing abuse. Although still in the research phase and facing skepticism, the proposal reflects Vitalik's broader focus on privacy and Ethereum's potential role in providing a tamper-proof data layer in the AI era, where verification and anonymity may become increasingly critical.

比推03/13 14:51

Vitalik Wrote a Proposal Teaching You How to Stealthily Use AI Large Models