Intel's comeback plan: edge AI, new chips, and a bet on robots
Las Vegas, NV - Intel is resetting its playbook. Backed by a historic investment from the Trump administration and a US government stake of roughly 10%, the company is pushing beyond PCs and into edge AI - fast.
The anchor of the plan is the newly announced Core Ultra Series 3. Intel says it will land in more than 200 new laptop designs this year, with a focus on better battery life and on-device AI performance for real workflows like coding assistants and smarter video calls.
Why this matters for executives
- On-device AI reduces latency, cloud costs, and data exposure - key for regulated industries and distributed teams.
- Edge compute expands the surface area for AI beyond the data center: laptops, kiosks, robots, and specialized devices.
- Intel is signaling faster decision cycles and customer feedback loops under CEO Lip-Bu Tan - a cultural shift to watch.
Intel's position: strong in PCs, pressure everywhere else
Intel still dominates PC chips with more than 71% market share in 2024, according to the International Data Corporation. But AMD is gaining, and Apple walked away from Intel silicon in 2020.
The company cut 15% of its staff last year. Shares lagged over the past five years, yet rallied roughly 84% in 2025 and are up about 98% year-over-year - a sign Wall Street is buying the turnaround story.
The Core Ultra Series 3: practical upgrades
- Battery life improvements for mobile-heavy teams.
- On-device AI acceleration for tasks like live meeting enhancement, coding agents, and creative workflows.
- Broad OEM adoption expected, giving IT buyers choice without niche lock-in.
"There's no one-size-fits-all for AI," said Jim Johnson, head of Intel's client computing group. "What a reporter needs may be different than what a gamer wants." That flexibility is the point.
Competition is pushing hard
AMD's latest laptop chips can run larger AI models locally, reducing cloud dependence. Qualcomm is moving deeper into PCs with a new laptop chip it says can deliver multi-day battery life and strong AI performance.
In data centers, Nvidia remains the core engine for AI services. It's also expanding into robotics, with new AI models and demos across healthcare and industrial use cases at CES.
Beyond PCs: the edge and robotics bet
Intel wants its chips in the "devices between PCs and the cloud," as Johnson put it. Think robots, smart endpoints, and AI-enabled equipment where latency, privacy, and cost control matter.
Early proof point: Oversonic Robotics plans to switch from Nvidia to Intel's Core Ultra 3 in its humanoid robots. Intel says the move cuts costs and speeds response because tasks don't need to round-trip to the cloud. Oversonic still trains models on Nvidia hardware - a common hybrid pattern emerging across the stack.
Still, demand for humanoid robots is uncertain. Gartner notes real deployments are scarce and limited by technical and physical constraints.
Signals from leadership and policy
CEO Lip-Bu Tan is pushing for faster feedback and execution. Johnson says Tan asked him to text directly if customers are unhappy - good, bad, problems, plans.
President Donald Trump praised Intel's US manufacturing push and the government's shareholder role on Truth Social. Policy tailwinds matter: they can de-risk capex-heavy bets and stabilize supply chains.
What to do now if you lead strategy, IT, or operations
- Run pilot comparisons: Intel Core Ultra 3 vs. AMD vs. Qualcomm for your top workflows (coding, video calls, content creation, analytics). Measure battery life, on-device inference speed, and TCO.
- Segment workloads: Decide what runs on device, at the edge, and in the cloud. Build a lightweight policy for privacy-sensitive AI tasks to default to local processing.
- Plan for heterogeneity: Expect a mixed fleet across vendors. Standardize on model runtimes and MLOps practices that work across chips.
- Explore robotics with constraints: If you test humanoids or mobile robots, start with narrow tasks, clear ROI windows, and local inference where possible.
- Negotiate with momentum: With Intel pushing hard and rivals responding, lock in pricing, refresh cycles, and support terms while competition is high.
Metrics to watch
- Actual OEM shipment volumes for Core Ultra 3 laptops (not just design wins).
- On-device AI benchmarks for tasks your teams use daily.
- Total cost per AI-enabled employee: device + cloud inference + support.
- Edge/robotics pilots that move from POC to production with measurable labor, quality, or safety gains.
- Government incentives and procurement signals that could shift vendor economics.
Bottom line
Intel is playing offense again. The bet is simple: win laptops with better battery and on-device AI, then extend into the edge where privacy, latency, and cost decide outcomes.
Rivals won't slow down. Your advantage comes from testing early, standardizing for flexibility, and funding what proves ROI in your environment.
Related sources
International Data Corporation - market share research and PC trends.
CES - event and product announcements shaping the device roadmap.
Upskill your teams
If you're planning on-device and edge AI rollouts, align training with roles. See curated options by job function: Complete AI Training - Courses by Job.
Your membership also unlocks:
