BTC $81,016.78 -2.45%
ETH $2,368.36 -4.75%
BNB $721.71 -2.48%
XRP $1.32 -3.47%
SOL $102.90 -3.21%
TRX $0.3312 -0.69%
DOGE $0.0800 -4.68%
ADA $0.2220 -5.14%
BCH $264.61 -4.53%
LINK $12.09 -5.89%
HYPE $79.78 -1.54%
AAVE $150.60 -5.49%
SUI $0.9933 -4.95%
XLM $0.1807 -4.08%
ZEC $1,101.56 -10.32%
AAPL $345.88 +0.85%
AMZN $250.14 -3.81%
GOOGL $342.97 -4.01%
MSFT $531.85 -2.37%
META $728.10 -4.78%
NVDA $217.18 -5.20%
TSLA $361.35 -3.83%
SNDK $1,343.17 -7.29%
INTC $95.06 -6.59%
SPCX $147.80 -6.04%
MU $880.53 -6.57%
AMD $556.75 -7.60%
BTC $81,016.78 -2.45%
ETH $2,368.36 -4.75%
BNB $721.71 -2.48%
XRP $1.32 -3.47%
SOL $102.90 -3.21%
TRX $0.3312 -0.69%
DOGE $0.0800 -4.68%
ADA $0.2220 -5.14%
BCH $264.61 -4.53%
LINK $12.09 -5.89%
HYPE $79.78 -1.54%
AAVE $150.60 -5.49%
SUI $0.9933 -4.95%
XLM $0.1807 -4.08%
ZEC $1,101.56 -10.32%
AAPL $345.88 +0.85%
AMZN $250.14 -3.81%
GOOGL $342.97 -4.01%
MSFT $531.85 -2.37%
META $728.10 -4.78%
NVDA $217.18 -5.20%
TSLA $361.35 -3.83%
SNDK $1,343.17 -7.29%
INTC $95.06 -6.59%
SPCX $147.80 -6.04%
MU $880.53 -6.57%
AMD $556.75 -7.60%

mal

All
Article
Flash

first_img Microsoft launched its self-developed AI chip Maia 300, with the key small chip using TSMC's 2-nanometer technology

According to a report by Central News Agency citing Reuters, Microsoft announced the launch of its third-generation self-developed AI chip, Maia 300, with plans to deploy over 1 million units next year to support internal AI services. The small chip responsible for key computing functions uses TSMC's 2-nanometer process. The Maia 300 can handle code writing and perform complex business tasks on behalf of users. The previous generation of chips has been used to run Microsoft's internal AI models and the OpenAI models provided through Microsoft Azure cloud computing services.Microsoft is taking a different technological approach compared to NVIDIA. The NVIDIA AI accelerator uses a large computing chip surrounded by high-bandwidth memory chips; the Maia 300 core processing unit is composed of small chips, with the small chips responsible for key computing functions using TSMC's 2-nanometer process, while other small chips use the lower-cost 3-nanometer process to handle data transmission out of the processor and transfer data to other locations within the system.The Maia 300 will be installed in cabinets, with each cabinet accommodating 96 chips and consuming about 167 kW of power. In contrast, NVIDIA's latest GB300 AI server cabinet is configured with 72 GPUs and consumes about 142 kW of power. Microsoft stated that it plans to deploy the Maia 300 on a large scale starting next year, reaching several gigawatts. Microsoft previously indicated that it expects its capital expenditures to exceed $175 billion this year.

first_img Anthropic: Small models may self-replicate on IoT devices

On October 12, 2026, William Zhou and Daniel Freeman, frontier red team researchers at the artificial intelligence company Anthropic, published a study. The article states that small models may have the ability to self-replicate on commercial IoT devices such as smart ovens and generic Android TV boxes within months. These replicas do not have a single shutdown point, and the doubling time may be measured in days, potentially adapting to unfamiliar devices on their own.Evaluation shows that a 9B model completed inference on 8 out of 9 hardware types. Models of 30B and below have been able to complete many single steps of replication, but errors occur when combined end-to-end; in simplified evaluations, GLM-5.2 and stronger models can achieve end-to-end replication. The team also distributed Gemma4-26B-A4B across 15 generic Android TV boxes, each costing about $55, operating at a speed of approximately 0.39 tokens/second, with each box holding two layers of the model.End-to-end testing was conducted on 8 GCP nodes behind a firewall, propagating through an unauthorized ADB port and setting up Llama 3.1-8B inference. Among the tested open-source models, only GLM-5.2 completed replication fully, transmitting to all 8 nodes in 4 out of 5 trials. The article divides autonomous replication and adaptation into four steps: discovering the target, obtaining execution, establishing inference, and continuing propagation from the new host, stating that there are currently no complete uncontrolled replicas.

IMF: The tokenized market is still small and fragmented, requiring clear laws and regulations

On October 8, the IMF published a blog based on Chapter 3 of the Global Financial Stability Report for October 2026, indicating that the tokenized market is growing rapidly but remains very small and highly fragmented. The average daily trading volume of tokenized repurchase agreements is about $300 billion to $350 billion, while other tokenized assets amount to about $65 billion, compared to an average daily volume of about $13 trillion in the U.S. repurchase market, and a total global capital market asset size of $300 trillion.The report points out that investors have shown interest in certain tokenized features, with more than half of the trades occurring outside traditional market hours, and about 80% of tokenized stock trades being less than 1 share. However, the tokenized market has lower liquidity and higher volatility than traditional markets, and cross-platform fragmentation weakens network effects. The IMF believes that legal certainty, regulatory clarity, interoperability, and secure settlement assets are four mutually reinforcing constraints.The report also warns that scaling up could amplify traditional financial risks such as sell-offs, liquidity squeezes, and contagion. Policymakers should adopt a technology-neutral approach, clarify the legal rights of tokenized assets, ensure consistent regulation for similar activities, support interoperability between tokenized platforms and the traditional financial system, and continuously monitor interconnectedness, leverage, and liquidity risks.

Huobi HTX Chief Analyst Cloud: The Bitcoin pullback is a normal retracement, and the differentiation of altcoins may continue until this week's data is released

The 10-year U.S. Treasury yield returned to above 5.2% overnight, reaching a high not seen since 2007, putting pressure on global risk assets. Bitcoin has fallen back to around $83,000, with altcoins experiencing even larger declines. In this regard, Cloud, the chief analyst at Huobi HTX, believes that this drop is more akin to a normal pullback after a rebound and does not constitute the starting point of a trend reversal. The high U.S. Treasury yields mainly suppress valuations, with limited impact on the overall liquidity. The funding structure and holding costs of the crypto market itself have also not been disrupted. This round of decline seems more like a clearing of positions before key data is released.It is important to be cautious about the divergence between Bitcoin and altcoins. During a phase of marginal liquidity tightening, funds tend to concentrate on the most certain top assets; altcoins lack incremental funds to support them, and with heavier leverage, their pullbacks are therefore amplified. This divergence is likely to continue before the release of this week's PCE and non-farm data. If the data falls short of expectations, altcoins will have greater elasticity and thus higher risks; if the data exceeds expectations, Bitcoin's relative strength will be more pronounced. Whether Bitcoin can stabilize at key support levels will be the main signal to determine if this round of pullback has ended.Note: The content of this article does not constitute investment advice, nor does it constitute any offer, solicitation, or recommendation of investment products.
app_icon
ChainCatcher Building the Web3 world with innovations.