# Сопутствующие статьи по теме AI

a16z Charts of the Week: AI Costs Halved and Usage Doubled This Year, Major Life Milestones for 30-Year-Olds in the US Delayed Across the Board

a16z's Charts of the Week explores four key trends. First, while a "DExit" (Delaware Exodus) narrative exists due to high-profile companies leaving over legal concerns, data shows a more complex reality. Delaware's overall share of U.S. businesses has actually grown, though Wyoming has seen a surge in LLC registrations. Second, AI demonstrates the Jevons Paradox: as the cost to process AI tokens halved this year, usage doubled. Demand for older GPU rentals (H100, A100) is also rising, contradicting predictions of a compute glut. Historical parallels suggest the full economic impact of AI may take time to materialize. Third, AI capital expenditure is massive, comparable to annual U.S. bank lending and significantly larger than U.S. corporate tax income or the military budget of any non-U.S. G7 nation. Fourth, the prediction market Kalshi is outperforming professional forecasters and futures markets in predicting the Federal Funds Rate, providing a valuable high-frequency, probabilistic benchmark. Finally, data shows a stark delay in life milestones for 30-year-olds in the U.S. Since the 1980s, far fewer are living independently, married, living with children, or owning a home. The only exception is college attainment, which has nearly doubled since 1995, though the value of a degree is increasingly questioned.

marsbit03/01 02:49

Anthropic Ban Controversy, OpenAI's Hundred-Billion Financing Dispute: What Is the Overseas Crypto Circle Discussing Today?

In the past 24 hours, the crypto market has witnessed a range of discussions from macro trends to ecosystem-specific developments. Key topics include the controversy over AI and national security boundaries, debates around OpenAI's record $110 billion private fundraising, and the potential impact of AI tools on tech employment. Anthropic refused a Pentagon request to remove safety restrictions on autonomous weapons and mass surveillance in its models, leading to a terminated $200M government contract and a divided public response. OpenAI secured massive funding from investors like NVIDIA and Amazon, though its significant losses sparked concerns about a potential tech bubble. Block's 70% engineering team layoffs, attributed to AI efficiency gains, intensified debates on AI's effect on jobs. In ecosystem news, Ethereum community showed optimism as Vitalik Buterin outlined a roadmap with specific timelines for ZK-EVM integration by 2026-2027. Morpho outperformed AAVE, with some attributing it to simpler governance. Solana integrated with SoFi bank, enabling 13.7M users to hold SOL directly. Base ecosystem saw experiments with AI Agents for trading and social interactions. Other notable events included Paradigm's planned $1.5B fund expansion into AI and robotics, Bitwise's XRP ETF application, and OpenAI firing an employee for insider trading on prediction markets. Hyperliquid emerged as the only profitable Digital Asset Treasury project, and regulatory discussions around prediction markets continued.

marsbit02/28 16:13

Here’s why OpenAI’s $110B raise is a major headwind for crypto

The race for AI dominance is accelerating, particularly among U.S. tech firms, as illustrated by recent FUD surrounding Anthropic. OpenAI’s $110 billion capital raise reinforces U.S. investment in AI infrastructure, which may indirectly pressure liquidity in the crypto market. This comes at an unfavorable time for crypto, as Bitcoin’s 90-day Realized Profit/Loss Ratio has fallen below 1.0, indicating that realized losses exceed gains—a sign of deteriorating investor profitability and tighter liquidity. Historically, such conditions reduce risk appetite and limit capital inflows into crypto. Meanwhile, capital continues flowing into tech equities, giving them a structural liquidity advantage over digital assets, a resilience currently absent in crypto markets.

ambcrypto02/28 16:02

Solana Foundation President Lily Liu Says 'Don't Waste Time with Crypto'—Is the Crypto Industry Really Dead?

Lily Liu, President of the Solana Foundation, sparked controversy by echoing a statement from OpenClaw founder Peter Steinberger: "don't waste time with crypto." Steinberger’s remark was a direct response to ongoing harassment from crypto speculators who promoted a fraudulent token falsely tied to his project, causing significant financial losses for investors. His frustration is aimed at the predatory and scam-prone aspects of the industry rather than the technology itself. Liu’s repetition of the statement, however, was met with criticism and confusion, given her leadership role in a major blockchain foundation. Many interpreted it as a negative signal toward the crypto industry’s current state, though others suggested it might be ironic commentary on its speculative excesses. The incident reflects broader trends: talent and capital are shifting toward AI, which is seen as being in a value-creation phase with clearer opportunities. However, the crypto market has historically absorbed capital and attention when other tech sectors mature and returns diminish. The current downturn is part of a natural cycle, and the industry’s long-term viability will depend on what remains after speculative projects are filtered out.

marsbit02/28 12:32

OpenAI Is Turning AI into a Nuclear Arms Race That Ordinary People Can't Afford

In a record-breaking funding round, OpenAI has secured $110 billion, raising its post-money valuation to $840 billion. This investment, led by Amazon, NVIDIA, and SoftBank, marks the largest-ever private tech funding and signals a new phase in the global AI race—one defined by extreme capital concentration and geopolitical significance. The scale of funding dwarfs the GDP of many mid-sized nations and equals nearly half of NVIDIA’s annual revenue. It also accounts for more than half of all AI startup funding in 2025, accelerating an industry-wide arms race in compute, talent, and model development. This capital influx, however, risks widening the gap between giants and smaller players, potentially stifling innovation and increasing market consolidation. Strategic investors are not merely providing capital: Amazon’s $50 billion commitment includes an eight-year, $100 billion cloud expansion deal. SoftBank’s $30 billion staged investment serves as both a hedge and a bridge for future sovereign wealth entrants. NVIDIA’s $30 billion replaces an earlier partnership promise and effectively locks up its advanced GPU supply, creating a closed loop that sidelines competitors. Despite ChatGPT reaching 900 million weekly active users and 50 million paid subscribers, OpenAI’s burn rate remains high. It spent $0.62 for every dollar earned in 2025, with cumulative cash burn projected to hit $1150 billion by 2029. At the same time, its market share is eroding amid rising competition from Google’s Gemini and Musk’s Grok. Facing mounting financial pressure, OpenAI is eyeing a potential IPO in Q4 2026. The offering could mark either the peak of the AI investment bubble or the beginning of the AGI era—but for now, the world watches as OpenAI races against capital, competition, and time.

marsbit02/28 11:46

AI Won't Achieve Technological Equality; It Will Only Reward the Right People

AI will not achieve technological democratization but will instead reward those with the right qualities. While new technologies like AI lower the entry barrier (raising the floor), they disproportionately elevate the ceiling—widening the gap between median and top performers. This follows a power law: democratized tools often lead to aristocratic outcomes, as seen in music (Spotify), photography, and software. In an era where AI makes execution cheap and accessible, true differentiation comes from taste—a relentless commitment to excellence even in unseen details—acting as proof of work. Unlike the previous SaaS era, where distribution was king, the AI era values deep, systemic insight and the ability to recognize and build toward unseen possibilities. Founders who succeed will be those who combine technical depth, aesthetic rigor, and the patience to play a long-term game. They will focus on business-critical domains (like payroll or compliance) where complexity, data accumulation, and high stakes create durable moats. Short-term speed matters, but lasting value comes from decade-scale compounding of quality, trust, and nuanced understanding. The future will see a bifurcation: a crowded field of replaceable point solutions and a handful of dominant platforms that capture most value. AI doesn’t change the rules—it reveals and accelerates them.

marsbit02/28 10:12

Behind OpenAI's $110 Billion Financing: The Game Between Amazon and Microsoft

OpenAI has secured $110 billion in new funding at a $730 billion pre-money valuation, with major investments from Amazon ($50 billion), Nvidia ($30 billion), and SoftBank ($30 billion). Notably, CEO Sam Altman’s public acknowledgment included Microsoft—an existing investor and partner—immediately after Amazon, signaling strategic alignment shifts. A key insight lies in two technical distinctions: “Stateless API” and “Stateful Runtime Environment.” Stateless API, the current mainstream model, handles single-request tasks without retaining context, but faces commoditization and margin pressure. In contrast, Stateful Runtime Environment supports persistent, autonomous AI agents capable of complex multi-step workflows, representing the future of enterprise AI adoption. Microsoft’s existing agreement ensures Azure remains the exclusive cloud provider for OpenAI’s Stateless API, securing present-day revenue streams. Meanwhile, Amazon’s expanded $100 billion partnership with OpenAI focuses on co-developing Stateful Runtime Environments via AWS Bedrock, positioning AWS as the infrastructure backbone for next-generation AI agents. This dual-cloud strategy enhances OpenAI’s leverage, reducing dependency on Microsoft while pitting Amazon and Microsoft in competition for future AI dominance. OpenAI gains negotiating power by diversifying its infrastructure partnerships and aligning each cloud giant with distinct—yet complementary—AI futures.

Odaily星球日报02/28 08:37

Cheaper to mine, harder to profit – The new reality for Bitcoin miners!

ambcrypto02/28 08:02

What Brought GPT and Claude Together? Opposing the Pentagon?

The article discusses the unexpected alignment between rival AI companies OpenAI and Anthropic, driven by their shared ethical stance against the U.S. Department of Defense's demands. Anthropic, the maker of Claude, had signed a $200 million contract with the Pentagon but insisted on two red lines: no mass surveillance of U.S. citizens and no autonomous weapons without human oversight. When the Pentagon demanded unrestricted use, Anthropic refused, citing ethical concerns. In a show of solidarity, over 400 employees from OpenAI and Google signed an open letter supporting Anthropic’s position. OpenAI’s CEO also internally affirmed similar principles. However, this unity was short-lived. After Anthropic held its ground and rejected the Pentagon’s ultimatum, it was labeled a "supply chain security risk," effectively barring it from all federal contracts. Meanwhile, OpenAI secured the Pentagon contract by accepting less stringent terms, agreeing not to engage in mass surveillance or autonomous weapons but without pushing for additional legal safeguards. The piece highlights the political and ideological dimensions of the conflict, noting that Anthropic’s stance was perceived as "woke" and ideologically driven, while OpenAI’s more pragmatic approach was rewarded. The outcome signals the high cost of resisting government pressure in the AI industry and raises questions about the real-world value of ethical principles when faced with political and economic consequences.

marsbit02/28 05:47

Paradigm's New Arithmetic: When Crypto Can't Hold $12.7 Billion, AI Becomes the Answer

Paradigm, a major crypto-focused VC managing $12.7 billion in assets, is raising a new $1.5 billion fund to expand into AI, robotics, and frontier tech. This shift follows a contraction in its crypto-only strategy—its third fund was $850 million, down from $2.5 billion in 2021—reflecting a lack of sufficiently large and early-stage crypto opportunities. The 2022 FTX collapse, which cost Paradigm $278 million, prompted internal reevaluation. By 2023, the firm had quietly removing “crypto” and “Web3” from its website, signaling a broader investment focus. Co-founder Matt Huang later clarified that Paradigm remains excited about crypto but sees AI as too significant to ignore. Paradigm’s move isn’t a full pivot to AI; rather, it targets the intersection of AI and crypto. Investments like $50 million in AI infrastructure firm Nous Research and the development of Tempo—a stablecoin payment platform—highlight this strategy. The firm believes AI agents will require programmable money and on-chain execution, creating synergies between both fields. The new fund also serves a narrative purpose: offering LPs a compelling growth story amid crypto’s concentration of capital and AI’s dominance in venture funding (61% of global VC investments in 2025). Paradigm aims to leverage its crypto expertise to capture value at the convergence of AI and decentralized technologies.

marsbit02/28 04:16