Wall Street Ratings Roundup Wednesday: AMD Price Target Raised to $670, Arm Target Lifted to $475
0xBroomberg
Multiple Wall Street firms raised targets on core AI chip names Wednesday — UBS lifted AMD to $670 and TD Cowen lifted Arm to $475 — yet Seaport held its sell on Nvidia, highlighting a sharp split between CPU optimism and financing-constraint fears.
AMD's target jumps $200 in one move — what does UBS see?
UBS kept AMD at buy and raised its target from $470 to $670, a gain of over 40%.
The thesis: standalone CPU rack demand — servers that run CPUs alone, without GPU pairing — is gaining market acceptance. This means → UBS views AMD not as "Nvidia's sidekick" but as a company with its own growth story on the CPU side.
In plain terms = the market used to focus entirely on GPUs; now it recognizes that pure-CPU servers are deploying at scale too — and that is AMD's home turf.
Arm's target nearly doubles — what makes TD Cowen so aggressive?
TD Cowen kept Arm at buy and raised its target from $265 to $475, nearly a 2× move.
The bull case: Arm's fiscal-2031 target of $15 billion in AGI CPU revenue is credible, TD Cowen says. But the GPU-to-CPU pairing ratio and per-core pricing remain key variables.
TD Cowen is also more bullish than consensus on Arm's IP licensing — the business of licensing chip blueprints to other manufacturers. This means → the bet is not just on Arm selling its own chips, but on Arm's designs being adopted by a widening circle of makers.
Why does Nvidia still carry a sell rating?
Seaport maintained its sell on Nvidia, citing rising financing burdens both on and off the balance sheet.
Broadcom has also joined the revolving-credit facility wave — a type of arrangement where companies repeatedly draw short-term loans. This reflects an industry-wide financing constraint, not a problem unique to Nvidia.
In plain terms = everyone is burning cash to expand capacity, but where the money comes from — and at what interest cost — is becoming harder to ignore.
Tesla and Reddit — what caught analysts' attention?
RBC kept Tesla at outperform, forecasting Q2 deliveries of roughly 405,000 units — slightly above the Visible Alpha consensus of 401,000. The data is expected within two weeks.
Needham kept Reddit at buy, calling its competitive position "rare" — able to "make its own weather" and largely immune to the AI-agent wave. This means → the analyst sees Reddit's community-content ecosystem as a moat AI cannot easily replicate.
Among newly covered names, which calls stand out?
Goldman Sachs initiated Twilio at buy with a 12-month target of $300, citing margin upside.
Stifel initiated BlackBerry at buy, arguing the market still misreads the company as an auto supplier when it has pivoted to a physical-AI software layer with deep ties to Nvidia, Qualcomm, and AMD. This means → Stifel is betting the perception gap will close.
BTIG initiated Take-Two Interactive at buy, target $290. Citizens initiated Bitdeer and MARA Holdings, both at outperform — these Bitcoin miners are redirecting compute capacity toward high-performance computing services for hyperscale cloud customers.
What signal does this wave of ratings send?
The round centers on the AI compute chain and tech growth names. The sharp target hikes on AMD and Arm reflect institutional optimism about a CPU demand recovery.
Yet the maintained sell on Nvidia warns that financing-constraint risk has not faded. This reflects a split within Wall Street — bullish on the demand side, cautious about funding pressure on the supply side.
In plain terms = most firms are saying "chips will sell better," but at least one is reminding the market "don't forget the risk that the cash runs out." Earnings season ahead will test which side is right.
Content is for reference only, not financial advice.