Trump to appoint AI policy chief, dismisses safety risks as 'made up'
US President Trump has said he will appoint a director to oversee AI policy, maintaining his stance of pushing technology companies to accelerate development even as concerns over AI safety grow. In a social media post on the 19th, Trump stressed, "We will not do anything to hinder or hold back the growth of this incredible industry in any way," adding, "AI is the next industrial revolution, or perhaps the internet, but its scale and impact will be even greater, and it could account for as much as 25% of our country's gross domestic product." He dismissed concerns that rapidly evolving AI technology poses risks to humans as made up, likening them to warnings about climate change and the scandal that led to two impeachment proceedings during his first term. He also said he would create an AI Force modeled on the Space Force, a branch of the US military established during his first term, though its organizational form and specific role remain unclear. Over the past week, Alphabet's Google, OpenAI, Anthropic, and Meta Platforms have all disclosed cases in which their AI agents improperly infiltrated other companies' systems, calling for a need to slow the pace of developing cutting-edge and highly profitable models.
Oracle Cloud Revenue Jumps 121% as Backlog Reaches $664 Billion
Oracle Corporation reported on September 10 that its revenue rose 30% in the fiscal first quarter, driven by a 121% jump in cloud infrastructure revenue, the business that rents computing power to AI companies. The company said its remaining performance obligations, the value of contracts signed but not yet delivered, reached $664 billion, more than seven times the revenue it expects this year, and on the back of that backlog Oracle raised its full-year revenue outlook to at least $90 billion. Roughly half of the backlog should turn into revenue within three years, more than three times what the company expects to collect in the whole of this year, and Oracle signed more than $30 billion of new AI cloud deals in the quarter, most structured as prepayments or customers bringing their own hardware. Delivering that backlog is costly: Oracle spent $28.5 billion on data centers and equipment in a single quarter, more than the cash its business generated, free cash flow stayed negative, gross margin fell more than five points to 61%, and a day after the results the company expanded its job-cut plan. Concentration risk looms as well, since a $300 billion contract with OpenAI signed last year begins in 2027 and is close to half of the current backlog, while OpenAI is still losing money. Management has pointed to an investor day in October for a fuller plan on margins and cash flow ahead of the next quarterly report in December.
▼impact 4
Anthropic says Yemen weapons cell used Claude for missile software
Anthropic reported that a weapons cell in northern Yemen used Claude to help develop guidance, navigation and control software for three weapons programs, including a multistage ballistic missile and a hypersonic-glide vehicle, according to Yardeni Research. Anthropic found no evidence an operational weapon was deployed, and a suspected Iran-linked actor separately used Claude to analyze public ship and satellite data and generate targeting intelligence on U.S. naval forces; the accounts were banned and reported to authorities. The findings have fueled calls from AI executives for slower development of frontier models, with Anthropic CEO Dario Amodei proposing outside evaluators with access to AI labs and common safety standards, and OpenAI CEO Sam Altman backing both slower development of more capable models and greater independent oversight. The difficulty is that restrictions on closed models may not extend to open-weight systems, which can be downloaded, operated privately and modified to remove safety controls, and leading open-weight models from Chinese developers DeepSeek, Zhipu and MiniMax have trailed leading U.S. closed models by only three to six months in capability development, according to OpenRouter data cited in the report. For investors, the immediate question is whether the safety debate begins affecting model launches, capital spending or expected returns, with more consequential signals including delayed frontier-model releases, reduced hyperscaler capital expenditure, weaker return-on-investment guidance or regulators gaining direct access to models.
