Posts

AI news flash: Big Tech’s AI infrastructure spending is making investors nervous. Alphabet, Amazon, Meta, and Microsoft reported over the past two weeks that their combined capex this year tops $740 billion, which Wall Street reads as spending on AI infrastructure like data centers and chips. Early in the AI boom, the market mostly rewarded big spending like this, but in Q4 the focus has shifted to negative free cash flow, rising debt, and whether these investments will pay off…

AI news flash: This TechCrunch piece is a lighthearted Halloween-eve feature listing a few last-minute costume ideas for tech nerds, each labeled with a different level of effort. The first idea is based on ‘recursive self-improvement.’ AI leaders like Dario Amodei and Sam Altman worry AI is moving too fast, and if models can keep improving themselves without human involvement, any effort to control the pace of the frontier falls apart…

AI news flash: Blackstone predicts that Anthropic could receive a ‘high junk’ rating from credit rating agencies after its IPO. The original article makes clear this is only Blackstone’s prediction, not a formal decision already made by the agencies. The key point of the prediction is financing cost. A junk rating means higher credit risk, so if Anthrop…

AI news flash: Intel’s stock plunged about 28% in a single day on Friday, one of its worst single-day performances ever, after the company gave a Q4 forecast below market expectations. Intel expects Q4 revenue of $15 billion to $16 billion, versus analyst estimates of about $17 billion, and gross margin will drop from 43% to 37%. With heavy investment spending, it also doesn’t expect positive cash flow. The drop wiped roughly $12 billion off the value of the US government’s Intel stake…

AI news flash: Nvidia is in talks with AI cloud provider Nscale to guarantee billions of dollars in debt financing for a West Virginia data center leased by Anthropic, first reported by The Information and later corroborated by Bloomberg. According to Nscale’s IPO filing, Anthropic signed a $44.6 billion contract for the first building alone…

AI news flash: Ollie, an AI assistant startup that pitched itself on respecting user privacy, is shutting down. Co-founder and CEO Bill Lennon announced on X on Friday that the app will close on November 6, about a week after the announcement. Lennon said a small team built a really good personal assistant, and he thanked the team, users, and investors for their support.

AI news flash: Murata Manufacturing announced it will invest 1 trillion yen (about $6.3 billion) over the next five years to meet growing demand for its advanced electronic components, which are mainly used in AI servers and related devices. Judging from the headline, the money will mostly go into expanding multilayer ceramic capacitor (MLCC) capacity, with production sites in Japan, the Philippines, China, and Thailand. The report notes that data-center-related MLCC demand is climbing fast, but some in the market worry about whether the AI boom will last, which gives this large investment a degree of demand risk. Because the original summary is very short, details like how the money is split by country, how much new capacity is coming, or when it will be finished aren’t provided, so see the original link for more.

AI news flash: According to Nikkei Asia, the combined operating profit and net profit of the world’s top five memory chip suppliers last quarter surpassed those of the top five US tech companies for the first time. It’s the first time in history that memory chipmakers’ profits add up to more than the American tech giants’. The main driver is ever-growing AI investment lifting chip demand, and demand is still growing faster than supply, which has memory makers posting strong financial results in this AI wave …

AI news flash: the Japanese government has asked US AI developers Anthropic and OpenAI to give Japan broader access to their advanced models used for finding security weaknesses, and it will encourage operators of critical infrastructure to adopt them. According to Nikkei Asia, the move comes as cyberattacks against Japan keep rising, and Tokyo wants to strengthen the country’s cyber defenses. Based on the original summary, here is what can be confirmed…

AI news flash: OpenAI announced stronger violence prevention for ChatGPT, focusing on three areas: broader safety protections, earlier intervention, and a new commitment to public safety reports. On the protection side, OpenAI expanded ChatGPT’s safeguards around violence-related content so the system can recognize and respond more effectively when a conversation shows potential risk of harm, instead of waiting until things get worse.