BTC $81,629.11 -0.23%
ETH $2,400.77 -0.04%
BNB $744.82 +1.57%
XRP $1.32 -0.77%
SOL $104.15 -0.37%
TRX $0.3300 -0.10%
DOGE $0.0801 +1.19%
ADA $0.2166 -0.27%
BCH $283.39 +2.44%
LINK $12.09 +0.27%
HYPE $77.79 -0.99%
AAVE $148.09 +0.43%
SUI $0.9918 +0.22%
XLM $0.1765 -0.97%
ZEC $1,226.63 -2.62%
AAPL $348.48 +1.21%
AMZN $250.57 -2.91%
GOOGL $323.61 -1.43%
MSFT $519.54 -1.77%
META $646.53 -4.96%
NVDA $213.34 -1.85%
TSLA $375.84 +1.31%
SNDK $1,291.49 -2.59%
INTC $86.45 -2.88%
SPCX $147.80 -4.79%
MU $869.84 -1.87%
AMD $512.83 -3.34%
BTC $81,629.11 -0.23%
ETH $2,400.77 -0.04%
BNB $744.82 +1.57%
XRP $1.32 -0.77%
SOL $104.15 -0.37%
TRX $0.3300 -0.10%
DOGE $0.0801 +1.19%
ADA $0.2166 -0.27%
BCH $283.39 +2.44%
LINK $12.09 +0.27%
HYPE $77.79 -0.99%
AAVE $148.09 +0.43%
SUI $0.9918 +0.22%
XLM $0.1765 -0.97%
ZEC $1,226.63 -2.62%
AAPL $348.48 +1.21%
AMZN $250.57 -2.91%
GOOGL $323.61 -1.43%
MSFT $519.54 -1.77%
META $646.53 -4.96%
NVDA $213.34 -1.85%
TSLA $375.84 +1.31%
SNDK $1,291.49 -2.59%
INTC $86.45 -2.88%
SPCX $147.80 -4.79%
MU $869.84 -1.87%
AMD $512.83 -3.34%

pi

All
Article
Flash

first_img The State Council issued an opinion to mitigate the impact of artificial intelligence applications on employment substitution

On October 23, the General Office of the State Council issued the "Opinions on Adapting to the Development of Artificial Intelligence Technology to Support the Construction of Employment-Friendly Development Models" (Guo Ban Fa [2026] No. 29). The document proposes to adapt to the development of artificial intelligence technology, enhance the positive effects of job creation, empower traditional positions, and mitigate the impact of job replacement.The opinions point out that the role of artificial intelligence should be leveraged to improve the employment environment, enhance occupational safety levels, and improve work quality and efficiency, with priority given to promoting applications in positions with labor shortages and high-risk environments; it is essential to cautiously manage the scope, intensity, and pace of artificial intelligence applications in labor-intensive industries to minimize the impact on workers. The document also encourages enterprises to explore the establishment of an unemployment buffer period system, to inform affected workers in advance during the large-scale promotion of artificial intelligence applications, and to provide support for job transition and transformation.The opinions also outline deployments in areas such as strengthening job creation effects, constructing employment-friendly upgrade paths, enhancing workers' digital literacy and employment capabilities, promoting artificial intelligence-enabled employment policy services, improving workers' rights protection and social security, and perfecting employment monitoring, early warning, and risk response mechanisms. The document is dated October 16, 2026, and is marked for public release.

first_img Anthropic: Claude has completed large-scale formal verification of the open-source library

On October 22, the Frontier Red Team of the artificial intelligence company Anthropic released research stating that large language models can automatically write formal proofs on a larger scale. The team collaborated with Theorem to use Claude to prove safety properties on over 100 real open-source libraries at the RISC-V assembly level, claiming this is the first known instance of using AI for large-scale formal verification of production software at the assembly level. During the proof-writing process, vulnerabilities that had gone undetected for years were also discovered.The experiment converted existing compilable fuzz tests from Google OSS-Fuzz into specifications: under assumptions of single-threaded static binaries, fixed memory budgets, RISC-V CPUs, and Linux, for any input less than 64 MB, the program should exit normally without triggering memory safety errors or assertion failures. Within weeks, Claude proved the aforementioned properties for 239 entry points in 123 OSS-Fuzz libraries after patching the identified issues. The proofs targeted compiled RISC-V instructions, approximately 180 million lines of Rocq, and were checked by the Rocq kernel. The page states that this does not mean these libraries are free of vulnerabilities, as remaining issues may lie outside trusted components or the scope of the proofs.The page also mentions that the team had Claude formalize RFC 1950 and proved that the zlib implementation in the Linux kernel conforms to that specification at the x86 binary level; an interpreter written in Rust covering a large subset of Python showed a 99.95% consistency with CPython behavior across over 1.7 million randomly generated programs. In 36 small verification tasks, Opus 5.5 proved that about half required approximately 40% fewer tokens than Opus 5. The team also pointed out that the proofs targeted single-threaded, statically linked RISC-V binaries, and the proof checker cannot determine whether the specifications align with reality.

Anthropic plans to go public by mid-November at the earliest, with market discussions estimating a valuation of about 1.5 trillion USD

The artificial intelligence company Anthropic plans to go public as early as mid-November, with market discussions giving it an IPO valuation of about $1.5 trillion, approximately 55% higher than the $965 billion valuation from financing in May this year. Based on this potential valuation, Spark Capital invested $75 million in 2023, with a stake now valued at around $10.9 billion, having a valuation of about $5 billion that year and securing a board seat, making it the only known venture capital firm with a board seat. After an initial investment of $35 million, Sound Ventures added $400 million, co-founded by Guy Oseary and Ashton Kutcher, entering the market in 2023, with a stake valued at several billion dollars. Menlo Ventures has invested over $1 billion in total, with a stake valued at about $21.8 billion, including a $500 million Series D round, entering the market for the first time in May 2023.Iconiq entered the market in 2025, investing over $6 billion in total, with a stake potentially valued at around $26.4 billion, with the valuation reaching $183 billion at the time of its first investment. Sequoia Capital began investing this year, with a total of about $3 billion, holding approximately 1% of the shares, valued at about $15 billion, having previously invested in OpenAI and xAI. The Singapore sovereign wealth fund GIC made its first investment in 2025, accumulating over $4.5 billion, holding about 1% of the shares, valued at around $15 billion. Former Google CEO Eric Schmidt invested tens of millions of dollars in 2021, holding about 1% of the shares, valued at around $15 billion.FTX invested $500 million in 2021, initially holding 13.56% of the shares, which diluted to 7.84% in 2024. After bankruptcy liquidation, it sold all its shares, recovering about $1.3 billion, with actual earnings around $800 million. Based on the undiluted 7.84% and the $1.5 trillion valuation, the theoretical value is about $117.6 billion, approximately $116.3 billion higher than the actual selling price, with calculations not accounting for subsequent financing dilution. In 2024, FTX sold all its Anthropic shares in batches, with the first batch involving about $884 million in transactions with 24 buyers. The Abu Dhabi sovereign wealth fund Mubadala's affiliate ATIC bought about $500 million, Jane Street purchased $100 million, and a Fidelity fund bought $45 million. Lightspeed holds less than 2%, and Google, Amazon, NVIDIA, Microsoft, as well as Founders Fund, Coatue, Altimeter, and others have also invested, but public holding data is not complete enough to accurately calculate returns for each.

Anthropic donated the Agent Skills standard to the Linux Foundation

Anthropic announced that it will donate the Agent Skills standard to the Agentic AI Foundation (AAIF) under the Linux Foundation. This format was designed by Anthropic while developing Claude Code, set to launch in October 2025 and open as a cross-platform standard in December. Currently, over 70 agent platforms, including ChatGPT, Cursor, and GitHub Copilot, have integrated it.Anthropic describes Skills as reusable work manuals prepared for AI agents, where developers can write task steps in SKILL.md and attach scripts and reference materials for different agents to read as needed. MCP is responsible for connecting external tools and data, while Skills explain how agents complete specific tasks. What is being donated includes the standard name, logo, GitHub organization, and official website domain.AAIF was jointly initiated by Anthropic, OpenAI, and Block in 2025 and has previously received MCP. In the future, engineers from Anthropic, OpenAI, and Google will jointly maintain the Skills standard in their personal capacities, no longer solely led by Anthropic. Vercel statistics show that the skills.sh directory has recorded approximately 1 million skills, with nearly 280 million installations. Snyk has confirmed 76 malicious skills in ClawHub's skill samples, and Anthropic has proposed establishing version control, formal proposal processes, and security guidelines.

first_img Jensen Huang: neocloud demand is extremely high, but capital is insufficient

NVIDIA CEO Jensen Huang stated during an interview with CNBC reporter Arjun Kharpal at the GTC conference in Berlin on Thursday that the demand for neocloud is extremely high and its capabilities are extraordinary, but insufficient working capital is restricting growth. He mentioned that there is huge demand in multiple fields of artificial intelligence, and the capital demand far exceeds supply, a problem that will be resolved.Jensen Huang indicated that electricity is the main constraint on the expansion of artificial intelligence, with European electricity prices being higher. He believes that allocating scarce energy to companies like Nebius and Mistral is crucial for their growth. Investment can enter Europe, but local infrastructure must be established first; if European neocloud can build infrastructure locally, the artificial intelligence ecosystem and economy can start and remain in Europe. He added that the whole of Europe still needs more neocloud.NVIDIA is conducting media activities in Europe this month and has committed nearly $10 billion in investments to local companies, universities, and research institutions. Jensen Huang announced last week during his visit to London that the company has invested over $1 billion in five UK startups and is also investing in several French startups. He stated that NVIDIA is trying to kickstart the artificial intelligence economy in Europe, which has a wealth of technology, industrial experience, and research capabilities, and is ready to participate in and build a local artificial intelligence economy.

1% of customer contributions come from Anthropic, while corporate Token spending has nearly stagnated

According to Ramp's tracking of payment data from over 70,000 U.S. companies, approximately 80% of Anthropic and OpenAI's enterprise revenue comes from the top 1% of clients. In August and September, enterprise Token spending saw nearly zero growth for two consecutive months; although it reached a historical high at the beginning of October, it was only 1.5% higher than the peak in August.According to Ornn data, the actual Token price index for Anthropic fell from over $1.70 in mid-September to $0.83, a decline of more than 50%, while OpenAI's corresponding index was $0.45. According to YipitData, the top 1% of Anthropic's clients contributed 46% of revenue in August, up from 25% a year ago. Ramp data shows that the paid adoption rate for enterprise AI in September only increased by 0.62 percentage points to 55.9%. Ramp's chief economist Ara Kharazian believes that growth is shifting from expanding the paid user base to enhancing the depth of use among existing users.Goldman Sachs warns that the growth rate of Token demand must exceed the decline in prices to support sustained investment in computing power. Apollo's chief economist Torsten Slok pointed out that companies are highly sensitive to computing power prices, and price reductions may not necessarily lead to sufficient new demand. Goldman Sachs data also shows that the realized return on invested capital for Amazon, Alphabet, Meta, Microsoft, and Oracle has dropped from 40% to 25%, the lowest since 2016.
app_icon
ChainCatcher Building the Web3 world with innovations.