A consortium led by Mubbadrah Investment has unveiled plans to build a 3 billion dollar artificial intelligence data centre in Oman, with planned capacity of up to 400 megawatts. The project was announced at a signing ceremony in London during the state visit to the United Kingdom of Sultan Haitham bin Tarik, and it is intended for hyperscale operators running large-scale AI training and inference workloads, with room to expand beyond the initial build.

Artificial intelligence has delivered "zero financial payoff" to corporate America outside the technology sector, according to Torsten Slok, partner and chief economist at Apollo Global Management. In a Daily Spark note published on October 24, titled "No Signs of AI Payoff Outside the Tech Sector," Slok argued that despite massive investments and widespread hype, there is still no evidence that revenue growth or profit margins are improving for most large companies.

How do you run a model that is far too large for any single chip? Cerebras, the wafer-scale chipmaker challenging the GPU orthodoxy, has published a detailed technical post answering that question with unusual specificity: spread the model across 59 of its WSE-3T wafers, keep every layer whole and on-chip, and let only the intermediate results travel between them.

Google has quietly rewritten one of the clearest promises in its privacy policy. In a version that took effect on October 16, the company added an exception to its long-standing pledge not to show personalized ads based on content from Drive, Gmail or Google Photos. The pledge still stands, but only until a user chooses to share that content with another Google service that supports personalized ads.

OpenAI said on Friday that it will no longer train its most capable models with access to the live internet, a striking reversal for a lab whose systems are built to search, browse and use online tools. Frontier training will now take place with no internet access at all, using cached copies and simulated versions of the web instead.

Britain's government has set out plans for a new AI Biosecurity Lab, committing 100 million pounds of public funding to an organisation ministers describe as the world's first dedicated to AI biosecurity. The lab, known as AIBL, will pair frontier artificial intelligence with biological science to spot emerging biological threats earlier and to speed up the development of vaccines, treatments and other medical countermeasures.

Daytona Platforms, a three-year-old startup building cloud infrastructure designed specifically for AI agents, has raised a 60 million dollar Series B round at a 1 billion dollar valuation. FirstMark Growth led the round, which the company announced this week, and the new money brings Daytona's total funding to 91 million dollars — a rapid climb for a company selling one of the least glamorous and most necessary pieces of the AI stack: somewhere for agents to actually run.

Google is putting far more weight behind an unusual answer to the AI power crunch: paying households and businesses to use less electricity at the moments the grid needs relief most. On Thursday, Google and distributed energy platform Voltus announced a tenfold expansion of their "Bring Your Own Capacity" agreement in the PJM region, adding 1,070 megawatts of flexible capacity to the 100 megawatts the two companies first committed in June.

Warp Sciences emerged from stealth on Thursday with 215 million dollars in funding and a plan to rebuild, in the United States, a piece of the drug discovery business that largely moved overseas: making and testing compounds. The round, led by Lightspeed, values the startup at 865 million dollars including the new money, according to the company's announcement.

Meta delivered a record quarter on Wednesday, and investors spent Thursday worrying about what it will cost to keep going. Shares in the Facebook and Instagram owner fell sharply, shedding more than 100 billion dollars in market value and dragging the Nasdaq down 2.6 percent, its biggest one-day fall since June. Nvidia slid 4.4 percent as the sell-off spread. The trigger was not weak business. It was the scale of Meta's artificial intelligence spending, and a new plan to help pay for it.