Get the newsletter
Intel · AI / ML · Filed Mar 28, 2025 · USPTO published Oct 1, 2026 · Oct 1 release

Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks

Running an AI model on-device is a power hog, and most chips don't adjust mid-task. Intel's new patent describes a processor that fires up all its cores to get started, then hands off to a smaller cluster to finish the job.

A processor's internal architecture, including multiple cores, cache levels, and a power management circuit for AI tasks. Drawing from patent filing US 2026/0300178 A1.
A processor's internal architecture, including multiple cores, cache levels, and a power management circuit for AI tasks.All 40 drawings ↓
Publication number
US 2026/0300178 A1(US20260300178A1)
Applicant
Intel Corporation
Filed · Published
Mar 28, 2025 · Oct 1, 2026
Inventors
Sujit Mahto, Jayesh Gaur, Adithya Ranganathan, Shankar Balachandran, Sudhanshu Shukla, Sreenivas Subramoney, Eran Shifer
Status when we published
On hold at the patent office
Document
20 claims
US classification
711/141

How Intel's mode-switching chip handles AI word by word

When you ask an AI assistant a question, the chip inside your device has to generate an answer one word (or word fragment) at a time. That first word takes the most effort: the chip needs to look up a lot of information at once. But each word after that is a simpler operation, mostly reading from what it just figured out.

Intel's patent describes a chip that takes advantage of that difference. It uses all of its processor cores working together to produce that first word, then automatically switches to a smaller group of cores for every word that follows. The rest of the chip powers down or steps aside, burning far less energy.

The switch happens without you doing anything. A built-in power-management circuit watches what the AI model is doing and reconfigures the chip's internal connections on the fly. The goal is to get the same answer with less battery drain, which matters a lot for phones, laptops, and anything else running AI locally.

How the processor shifts cores and memory links mid-inference

The patent centers on a specific observation about how large language models (the AI systems behind chatbots and on-device assistants) actually work. Generating a response happens in two distinct phases.

The first phase, often called the prefill stage, is compute-heavy: the chip reads your entire prompt and sets up a compressed representation of it. This benefits from many processor cores working in parallel. The second phase, token generation (producing each word or fragment of the answer), is less compute-bound but reads a large amount of cached data repeatedly, which taxes memory bandwidth more than raw processing power.

Intel's design exploits this split by building a chip with two configurable operating modes:

  • Mode 1 (full-chip): All processor cores, the interconnect (the internal highway linking cores to memory), and all cache-coherence circuits (the hardware that keeps every core's local memory in sync) are active to produce the first output token.
  • Mode 2 (reduced-cluster): A subset of cores, a narrower interconnect path, and fewer coherence circuits take over for all subsequent tokens.

The switch is managed by a power management circuit baked into the chip itself. Critically, the cache-coherence circuits, hardware that normally prevents cores from reading stale data from each other, are partially disbanded when those extra cores go idle. Fewer active nodes means less coordination overhead and less wasted power keeping dormant hardware in sync.

What this means for AI on laptops and low-power devices

For you, the most direct payoff is battery life. Running an AI model locally, whether that's a writing assistant, a real-time translation tool, or a coding helper, currently drains a laptop or phone faster than almost any other task. A chip that automatically scales back to a leaner configuration after the hardest part of the job is done could meaningfully extend how long you can use those features unplugged.

Intel has been filing around on-device AI efficiency since at least 2024, and this patent fits that pattern. The design also has implications for data-center inference, where thousands of requests are processed simultaneously and energy costs are enormous. Shaving power from the token-generation phase, which is where most of the time in a long AI response is actually spent, is where real efficiency gains live.

Intel's 44th filing we've tracked in the AI chip wars since May adds to a run that includes faster memory lookups and blocking wasted data fetches.

Source: US 2026/0300178 A1 at the USPTO (PDF). Full patent text and figures are in the official publication.

Get our take in your Top Stories

Liked this breakdown? Add Patentlyze as a preferred source on Google, and our plain-English take shows up more often in your Top Stories the next time Intel patent news breaks.

There are more where this came from

We read the new patents from Big Tech, carmakers and more and send you the ones worth knowing. Plain English, free, every week.

The drawings · 40 sheets

40 drawing sheets from US 2026/0300178 A1 · click any drawing to enlarge

Processor with cores and caches., drawing sheet 1 from US 2026/0300178 A1
Sheet 1 Processor with cores and caches.
System with multiple processor cores., drawing sheet 2 from US 2026/0300178 A1
Sheet 2 System with multiple processor cores.
Software prefetch request flow., drawing sheet 3 from US 2026/0300178 A1
Sheet 3 Software prefetch request flow.
Table of memory requests., drawing sheet 4 from US 2026/0300178 A1
Sheet 4 Table of memory requests.
Tensor processing prefetch code., drawing sheet 5 from US 2026/0300178 A1
Sheet 5 Tensor processing prefetch code.
Page level prefetch instruction., drawing sheet 6 from US 2026/0300178 A1
Sheet 6 Page level prefetch instruction.
Instruction fetch and execution., drawing sheet 7 from US 2026/0300178 A1
Sheet 7 Instruction fetch and execution.
Instruction set architecture translation., drawing sheet 8 from US 2026/0300178 A1
Sheet 8 Instruction set architecture translation.
Show the other 32 sheets
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 9
Sheet 9
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 10
Sheet 10
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 11
Sheet 11
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 12
Sheet 12
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 13
Sheet 13
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 14
Sheet 14
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 15
Sheet 15
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 16
Sheet 16
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 17
Sheet 17
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 18
Sheet 18
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 19
Sheet 19
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 20
Sheet 20
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 21
Sheet 21
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 22
Sheet 22
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 23
Sheet 23
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 24
Sheet 24
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 25
Sheet 25
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 26
Sheet 26
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 27
Sheet 27
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 28
Sheet 28
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 29
Sheet 29
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 30
Sheet 30
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 31
Sheet 31
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 32
Sheet 32
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 33
Sheet 33
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 34
Sheet 34
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 35
Sheet 35
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 36
Sheet 36
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 37
Sheet 37
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 38
Sheet 38
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 39
Sheet 39
Intel Patents a Chip That Shuts Off Its Own Processing Units During AI Tasks patent drawing, sheet 40
Sheet 40
Patent filing page

More in AI training & infrastructure

All 324 AI training & infrastructure patents →

Watchlists this filing belongs to

All watchlists →
Reader comments

No comments yet

Start the discussion

Real name or a handle, either is fine. Comments are read by a person before they appear, so allow a little time. Keep it about the filing.