10 #Bitcoin ETFs outflows 3,062 $BTC(-$302.76M) and 9 #Ethereum ETFs outflows 16,114 $ETH(-$53.01M) yesterday.

IBM released its Q2 financial results, reporting a mere 1% year-over-year revenue increase, a downward revision to its full-year revenue outlook, and that plummeting mainframe sales weighed on its performance. The tech giant’s Q2 revenue reached $17.2 billion, up roughly 1% YoY, while adjusted earnings per share came in at $2.93—both below market consensus. Its full-year revenue growth forecast was cut from "over 5%" to a range of 4% to 5%. Revenue from its Z-series mainframes plunged 42% YoY, dragging down its infrastructure segment by 7%. Meanwhile, its software business remained relatively robust, with revenue rising 5% YoY. Following an earnings warning issued a week prior, IBM’s shares climbed around 3% in after-hours trading.

Elon Musk’s SpaceX AI plans to build at least one gigawatt-scale AI data center in Texas, with a scale matching or even exceeding its Memphis facility, further intensifying the competition in AI infrastructure. The company’s AI capital expenditure rose to $7.7 billion in the first quarter, tripling year-over-year. While expanding computing power, it also faces massive capital burn and energy supply pressures.

According to The Wall Street Journal, AMD has signed a multi-billion-dollar AI server partnership agreement with Anthropic. Under the terms of the deal, Anthropic will purchase up to 2 gigawatts of AMD’s latest-generation Instinct MI450 chips for its data centers starting in the first half of 2027, while leasing some computing power from other major cloud service providers or new cloud vendors. Upon reaching certain computing power deployment milestones, AMD will invest up to $5 billion in Anthropic, marking its first investment in an AI company. AMD’s CEO stated, “We have long aimed to be a key infrastructure partner for Anthropic, and the two companies’ engineering teams have been collaborating for some time.”

According to a Wall Street Journal report, OpenAI is expanding its data center scale and increasing its spending budget in this field. Sources revealed that OpenAI has raised its projected 2030 computing expenditure from approximately $600 billion earlier this year to around $750 billion. This growth reflects OpenAI’s new agreements with cloud computing service providers to rapidly obtain the massive computing power required for developing and operating its AI models. OpenAI announced Wednesday it will invest $20 billion to launch the "Camellia" data center project in Effingham County, Georgia. Sachin Katti, OpenAI’s vice president of compute strategy, said the company has signed a contract with Georgia Power to secure 3.2 gigawatts of power supply between 2028 and 2032. The project’s total cost is expected to exceed $30 billion, potentially making it one of OpenAI’s largest data centers to date. Per people with knowledge of the matter, OpenAI has hired Brent Mayo—one of the lead architects behind Elon Musk’s data center construction projects—to help ensure the data center project progresses as scheduled. Meanwhile, OpenAI has officially launched its enterprise AI agent platform Presence, which deeply integrates AI models with internal corporate data, policies, and business processes, offering automated solutions for scenarios such as customer service, sales, and IT support, moving beyond solely selling model access rights.

According to Bitget market data, South Korea's KOSPI index rose to touch 6,900 points, gaining 2.44% intraday, with SK Hynix up 4% and Samsung Electronics rising 3.2%.

Alphabet reported stronger-than-expected Q2 2026 results, with earnings per share (EPS) of $9.11 and revenue of $119.8 billion, beating Wall Street consensus estimates of $6.23 EPS and $116.51 billion in revenue respectively. However, investors’ concerns over the company’s sharply increased capital expenditure plan and recent margin pressure pushed Alphabet’s share price down 4.24% to $327.4 in after-hours trading, per market data from BIT (bit.com). Market analysts attribute the post-session sell-off to Alphabet raising its full-year 2026 capital expenditure forecast to $19.5–$20.5 billion, up from the prior projection of $18–$19 billion. Management stated the hike reflects plans to accelerate capacity delivery to meet demand, with roughly 60% of the expenditure allocated to servers and 40% to data centers and networking equipment. The company noted demand for AI infrastructure remains robust, though supply constraints are still limiting the pace of capacity delivery. Management also added that third-party capacity usage in Q3 may exert some margin pressure, while the integration of Wiz will create short-term headwinds in 2026. Looking ahead, Alphabet expects sustained strong demand for its Search, YouTube, and Cloud businesses, with AI features driving higher usage and improved monetization. The company also noted that most revenue from TPU system sales is expected to be recognized in 2027, not 2026.