I have been building a bear case on AMD for weeks and I think the situation is worse than most people realize. Let me walk through it. Advanced Micro Devices is carrying debt levels that look manageable when rates are low and revenue is growing. But what happens when growth normalizes and refinancing costs are 200 basis points higher than the debt they are rolling off? The interest coverage ratio deteriorates fast. That is not a hypothetical. That is arithmetic. Then there is the competitive landscape. Two quarters ago a new entrant started taking share in the core market. Management dismissed it on the earnings call. The next quarter the share loss accelerated. Management blamed macro. Do you see the pattern? When management blames macro for company specific problems it is usually a sign the moat is narrower than they claimed. Inventory is also building. Days sales of inventory is up meaningfully year over year. That is either channel stuffing or demand softening. Neither is good. I have seen this combination before. High debt, rising competition, building inventory, and a management team in denial. It rarely ends well for shareholders. What data point would change my mind? Sustained organic growth reacceleration for two consecutive quarters. Not one. Two. One quarter can be a fluke or a benefit from easy comps. Two quarters is a trend. Until I see that I am not touching this.

lisa su said mi300x is supply constrained not demand constrained. that is literally the best problem a semiconductor company can have. the ai accelerator market is big enough for 2 players

@semiconductor_vet_88 sure but nvda has 90% market share in ai accelerators. amd is fighting for table scraps. the stock prices in amd taking 20% share and i dont see it happening

server cpu share at 25% and climbing every quarter. intel is on the back foot and has been for 3 years. this is a slow motion share gain story that nobody seems to appreciate

Client segment still weak. PC market not recovering as fast as people hoped. The AI story carries this stock but the valuation assumes everything goes right.

MI300X is shipping in volume now. Lisa Su said supply constrained not demand constrained. That's a great problem to have. The AI accelerator TAM is big enough for two players.

Server CPU share is at 25% now. Steady gains every quarter. Intel is on the back foot and AMD is executing. This is a multi-year share gain story.

I split my AI allocation 60/40 between NVDA and AMD. Feels like the right hedge. If AMD steals share I win, if NVDA stays dominant I still win. Can't go wrong either way probably.

AMD has a bear flag on the daily chart. Needs to break above 170 to invalidate. I'm long but nervous. Stop at 155 if it fails.

AMD MI300X is actually competitive with NVDA for inference workloads. Meta and Microsoft are already buying. Lisa Su is executing flawlessly. This is the value play in AI chips, not NVDA at 65x earnings.

AMD is a great company but the stock priced like NVDA's growth rate without the margins. P/E of 45 on 10% revenue growth doesn't compute for me. Need to see MI300 revenue numbers accelerate.

AMD bears keep pointing at the multiple like it's 2022 and rates are still spiking. Stock's sitting around $157 and yeah, it's not cheap - but the IV on near-term options is actually pretty compressed right now, which tells me the market isn't pricing in a big move either way. That's interesting to me. I've been watching the $160 calls for next cycle and the delta skew isn't screaming "crash incoming" the way the bear thesis would suggest. The data center GPU ramp is real, MI300X adoption isn't slowing, and Lisa Su doesn't miss execution targets very often. Bulls aren't perfect here either - the PC segment is still meh and Intel's irrelevance doesn't automatically hand AMD a win. But my honest read is bears are fighting the last war on this one. So with AMD holding around $157 despite the macro noise, what's the ONE thing bears keep getting wrong that would actually flip their thesis if they acknowledged it?

AMD is starting to feel like a crowded trade hiding behind AI hype, and I'm genuinely nervous about what happens when the enthusiasm meets reality. Yeah, the data center story is real. But the stock's still sitting around 170x forward earnings - that's not a bargain, that's a prayer. And with VIX creeping back up and macro headwinds from a stronger dollar pressuring semiconductor demand globally, I don't see the margin of safety here. Lisa Su is brilliant, but brilliant doesn't protect you when institutional money decides to rotate out of semis during a risk-off move. The NVDA comparison is exhausting too - AMD isn't NVDA, and pricing it like it's on the same trajectory feels dangerous right now. I'm not short, but I'm definitely not adding. Sitting on hands feels right. So here's what I actually want to know - with AMD still priced at ~170x forward earnings, what specific catalyst would have to materialize (or collapse) before you'd pull the trigger and either buy or bail today?

Honestly, I think AMD bulls are getting way too comfortable right now. Yeah the AI story is compelling, but this stock is still trading around 160x forward earnings - that's a multiple that prices in basically flawless execution for years. I hold broad tech exposure through index funds and even I can see that valuation is doing a lot of heavy lifting here. Nvidia ate their lunch in the data center, Intel is a mess but still competing on price, and AMD's MI300 ramp has been solid but not the runaway hit some expected. Meanwhile the PC and gaming segments are still soft. When I look at diversified semiconductor ETFs, AMD is actually dragging on a few of them lately while the index holds up fine. I'm not saying it's a bad company. It clearly isn't. But the market has it priced for perfection. With AMD still sitting near 160x forward earnings, what's the one catalyst bulls are counting on that they're quietly glossing over when it comes to actual free cash flow conversion?

Okay hot take: AMD is actually the more interesting AI chip play right now and I'm tired of pretending it isn't. Everyone's obsessed with NVDA (rightfully so, fine), but AMD's MI300X is genuinely eating into data center share and I don't think the market is fully pricing that in. Stock's been bouncing around the $155-160 range after getting absolutely crushed from its highs, and Reddit/X sentiment has flipped weirdly bearish on it lately - which honestly makes me MORE curious, not less. Their data center revenue hit $2.3B last quarter. That's not a meme number. That's real. Bears keep saying Lisa Su is overpromising on AI ramp timelines. Bulls say the NVDA supply crunch alone creates a AMD runway that's being ignored. So here's what I actually want to know from people who've done the work: with data center revenue at $2.3B and still growing, what's the bear case that ISN'T just "NVDA will always dominate" - because that feels lazy to me?