@imnotharsh Where's Ericsson partnership in your summary? Bullish or bearish? Wondering why ERIC over NOK for deeper network integration , was trying to figure out. What's your thought on this? Seems like ERIC not gaining much CT attention yet compared to NOK. Could be an opportunity there?
@Augustine0091 Curious just for my own understanding and I've been learning as well. Isnt this the monthly chart for June? I thought June is like the strongest fire element month of the year with many of wood element?
Intel $INTC : The Trade Everyone Mocked Is Now Forcing a Rethink
For the better part of the last five years, owning Intel $INTC felt like defending a losing argument in a room full of people who had already moved on.
And frankly, the market had its reasons.
Execution missteps. Process delays. Margin compression. A capital-intensive foundry strategy that looked, at best, premature and, at worst, misguided. Meanwhile, NVIDIA $NVDA was printing money, and Taiwan Semiconductor $TSM was quietly reinforcing its dominance.
So the trade became simple: own the winners, ignore everything else.
That simplicity is now breaking down.
The AI Narrative Is Getting More Complicated And That’s Bullish for Intel
The first phase of AI investing was almost embarrassingly straightforward. You didn’t need nuance. You needed exposure to GPUs.
Buy NVIDIA. Maybe sprinkle in some Advanced Micro Devices. Ignore the rest.
But the second phase of AI is not going to reward that kind of linear thinking.
We’re now dealing with inference workloads, real-time applications, distributed systems, and increasingly complex infrastructure requirements. This is no longer just about raw compute it’s about coordination across the entire stack.
And that shift matters.
Because once AI moves from training models to actually deploying them at scale, the bottlenecks change. Memory, networking, power efficiency, and critically system orchestration start to matter more.
That’s where Intel begins to re-enter the conversation.
Not as the leader. Not as the obvious winner. But as a participant that is suddenly more relevant than the market was pricing in.
CPU Resurgence: From Commodity to Control Plane
For years, the CPU was treated like a background component in data centers. Necessary, but not strategic.
AI flipped that narrative initially in favor of GPUs.
Now it’s starting to rebalance.
In training environments, the architecture is relatively clean: one CPU can manage multiple accelerators. It’s efficient, predictable, and heavily GPU-centric.
Inference is different.
Once models are deployed into real-world applications, the system becomes far more dynamic. There’s constant interaction between compute, memory, storage, and networking layers. Latency matters. Coordination matters.
And suddenly, the CPU isn’t just “there” it’s orchestrating the entire workload.
This isn’t theoretical anymore.
Intel’s Data Center and AI segment posted $5.05 billion in Q1 2026 revenue, up 22% year-over-year. That’s not a turnaround narrative that’s demand showing up in numbers.
What’s happening beneath the surface is even more important.
The industry is quietly shifting from:
🔹Training-heavy, GPU-dominated architectures
to
🔹Inference-heavy, system-coordinated environments
And in that world, CPUs regain strategic importance.
If agentic AI scales the way many expect, workloads become even more fragmented and interactive. That pushes architectures closer to a one-to-one relationship between CPUs and accelerators.
That’s a very different demand curve than what the market modeled two years ago.
The “GPU-Only” Trade Is Cracking
This doesn’t mean NVIDIA loses.
It means the ecosystem expands.
The next phase of AI infrastructure is not a single-component story. It’s a systems story CPUs, GPUs, memory, packaging, and software working in tandem.
Intel doesn’t need to beat NVIDIA to win here.
It just needs to become necessary.
Partnership signals are already pointing in that direction. Intel’s deeper engagement with hyperscalers like Google $GOOGL suggests that large-scale AI deployments are placing increasing pressure on system-level design not just chip-level performance.
That’s a subtle shift, but it’s one the market tends to underappreciate until it shows up in sustained revenue growth.
Intel Foundry: Still Unproven, But No Longer Irrelevant
Let’s be clear: Intel Foundry is not fixed.
It’s still loss-making. External customer traction is limited. And execution risk remains high.
Treating it as a peer to TSMC today would be a mistake.
But dismissing it outright is becoming equally flawed.
The nature of AI hardware is changing. We’re moving toward larger, more complex chip designs that require advanced packaging, tighter integration, and secure supply chains.
This is no longer just about designing chips it’s about building them at scale, efficiently.
Intel’s progress on 18A and early signals on 14A yields matter more than they appear on the surface. Yield improvements translate directly into cost efficiency and customer confidence.
But the more immediate opportunity isn’t even leading-edge nodes.
It’s advanced packaging.
Technologies like EMIB and Foveros position Intel in a part of the value chain that is becoming increasingly critical as chip architectures move toward chiplets and heterogeneous designs.
Think of it less like manufacturing and more like system assembly at the silicon level.
This is where differentiation can happen.
Intel doesn’t need to dominate this market. It just needs to be a credible second source in a capacity-constrained environment.
That alone has value.
Earnings Tell a Different Story Than the Narrative
Intel’s Q1 2026 wasn’t just “better than expected.”
It forced a reassessment.
🔹Revenue: $13.6B (+7% YoY)
🔹Non-GAAP EPS: $0.29
🔹Gross margin: 41% (well above guidance)
But the most important detail wasn’t in the headline numbers.
It was supply.
Management indicated that demand exceeded supply, with over $ 1B in potential upside constrained by capacity.
That’s not recovery demand. That’s structural demand.
The Data Center and AI segment was particularly clean:
🔹$5.1B revenue
🔹+22% YoY growth
🔹ASIC business nearly doubling year-over-year
At the same time, the company is still burning cash, with negative free cash flow driven by heavy capex.
So no, this isn’t a flawless turnaround.
But it is the first time in years where the demand side of the story is no longer the problem.
$AMD CEO ON LONG-TERM EPS TARGET
“With the momentum we are seeing across the business and the expanding market opportunity, we see a clear path to exceed our long-term financial targets, including delivering more than $20 in EPS over the strategic time frame.”
Some color from the $AMD call that wasn't in the release.
-Server CPU TAM revision. From ~18% annually (FAD November baseline) to >35% annually, reaching >$120B by 2030. This is a structural reframe.
-Q2 server CPU revenue guide: >70% YoY.
-Q1 server CPU revenue: >50% YoY. Cloud and enterprise each up >50% YoY.
-Fourth consecutive quarter of record server CPU revenue.
-2H Gaming revenue to decline >20% vs 1H on memory and component costs. Concrete negative warning.
-2H Client outlook framed cautiously on memory/component pressure, but AMD still expects YoY growth and to outperform the market.
-Data Center segment operating margin: 28% vs 25% a year ago. Segment-level number.
-Embedded design wins up double-digit % YoY, with "billions of dollars" in new wins.
-ROCm cadence acceleration via "agent-based coding workflows."
-CFO-only Q2 guide details: Non-GAAP OpEx ~$3.3B, OIE +$60M gain, tax 13%, diluted share count ~1.66B.
Picking apart the Q&A...
@Augustine0091 Thanks for your tips! Appreciate it and looking forward for next one. I did mine and got some positive result. But I sold too early after seeing the talk and negotiation collapse on the weekend. Now I am just sidelined and kinda fomo not sure if I should re-enter. 😅🥲