Intel Nova Lake-S desktop chips target 74 TOPS neural processing for Copilot+

Intel is preparing Nova Lake-S desktop processors with an NPU rated at 74 TOPS, positioning the lineup to comfortably exceed Microsoft Copilot+ acceleration requirements and compete with AMD's upcoming Ryzen Artificial Intelligence 400 series.

Intel is preparing the Core Ultra Series 4 ‘Nova Lake-S’ as its first desktop processor family designed to deliver native acceleration for Microsoft Copilot+. The neural processing unit in ‘Nova Lake’ offers throughput of 74 TOPS, which directly targets the Copilot+ specification by far exceeding the 40 TOPS required by Copilot+. Intel already satisfies this bar in the mobile segment with its ‘Lunar Lake’ processors, but the current ‘Arrow Lake-S’ and upcoming ‘Arrow Lake-S Refresh’ desktop parts reach only 13 TOPS because they rely on an older generation NPU derived from ‘Meteor Lake’.

Product segmentation within the Nova Lake-S lineup is expected, and it is not guaranteed that every desktop model will provide the full 74 TOPS rating, as some chips could ship with lower NPU throughput. However, there is enough headroom in the design for Intel to configure the range so that even the least expensive retail configuration can deliver at least 40 TOPS in order to satisfy Copilot+ requirements. This approach would let Intel scale features and performance across the stack while still maintaining baseline compatibility with Microsoft’s on-device Artificial Intelligence experience.

Competitive pressure is emerging from AMD, which could claim the distinction of shipping the first socketed desktop processor with a Copilot+ capable NPU. AMD is positioned to introduce its Ryzen Artificial Intelligence 400 series ‘Gorgon Point’ APUs for Socket AM5 before Intel brings Nova Lake-S to market. The AMD launch could occur ahead of Intel’s rumored late-2026 debut for the Core Ultra Series 4 desktop processors, potentially giving AMD an advantage in time-to-market for Copilot+ ready desktop silicon.

55

Impact Score

HMRC signs £175m Quantexa deal for fraud detection

HM Revenue and Customs has signed a £175 million, 10-year agreement with Quantexa to unify fragmented data and strengthen fraud detection. The deployment is designed to automate routine work while keeping decisions transparent, auditable and subject to human approval.

Us supercomputers test new Artificial Intelligence chip suppliers

Sandia National Laboratories is evaluating chips from Israeli startup NextSilicon as major chipmakers shift their roadmaps toward Artificial Intelligence. The move reflects growing concern that mainstream processors are deprioritizing the scientific computing features government labs still need.

EU Artificial Intelligence Act amendments delay some deadlines and add new bans

A provisional Digital Omnibus on Artificial Intelligence would push back several EU Artificial Intelligence Act deadlines, refine how the law interacts with sector rules, and introduce new prohibited practices. The package also expands limited bias-testing allowances and strengthens centralized oversight for some high-impact systems.

Qwen 3.5 raises concerns about censorship embedded in model weights

A technical analysis of Alibaba Cloud’s Qwen 3.5 points to political censorship circuits embedded directly in the model’s learned weights. The findings highlight operational, compliance, and product risks for startups building on third-party Artificial Intelligence models.

Laptop prices rise as memory shortages hit PCs

Laptop prices are climbing as memory makers redirect production toward data center demand driven by Artificial Intelligence. The squeeze is spreading beyond RAM to graphics memory and SSDs, raising costs across the PC market.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.