Alphabet's AI buildout is straining compute and cash, the Hormuz deal has not normalized oil, AAOI and VIAV moved optics from promise to execution, and TEAM/NET versus TTD showed software is splitting on real growth—not a blanket AI-disruption story.
Alphabet's cloud business and DeepMind are competing for the same scarce TPU capacity. A former Google researcher called the leadership departures a setback, while Alphabet returned to bond markets after free cash flow turned negative and buybacks stopped. Demand for AI services is not the problem; the decision risk is whether commercial cloud returns arrive quickly enough to fund the research race.
Iran and Oman moved a temporary Hormuz shipping route toward final drafting, but the market quickly found the limitation: reports said US- and Israel-linked vessels could still be restricted. Crude rose 4%, reversing the initial relief. The agreement may reduce some shipping friction without restoring neutral access, so inflation and rate expectations remain exposed to vessel rules and actual tanker flows.
AAOI's revenue and earnings beat, yet the cleaner signal was that demand may outpace production capacity through mid-2027 while margins absorb expansion costs. VIAV separately reported a fifth straight beat and guided well above consensus as optical testing complexity rises. The cycle is moving from policy and backlog stories to factory execution: capacity, yields and Q4 delivery now matter more than headline demand.
Atlassian's 28% revenue growth and Cloudflare's 36% gain came with stronger forward guidance, while Trade Desk sharply missed already reduced expectations. That split argues against a blanket 'AI kills software' thesis. Software spending is still available for products proving cloud adoption, customer growth and agent use; weak execution is being punished rather than hidden by the sector rebound.
Labubu explicitly bought both names into the first SpaceX unlock session; this was a fresh trade, not a view inferred from tone.
Northwise is accepting near-term spending pressure while MELI defends logistics, credit and market share against PDD, SE and Amazon.
The self-reported purchase leans on a trailing P/E near nine, but the post gives no operating catalyst that rules out a value trap.
The contributor reported selling MSFT at $494 and plans to rotate most of the exposure into META for a cleaner advertising-driven AI revenue loop.
Caesar clarified the shares were bought before earnings and kept afterward; no new post-earnings purchase was disclosed.
The Reddit contributor disclosed a three-week straddle, but the first unlock produced heavy volume without the expected range break.
Con recently de-risked after gains and would rather re-enter 5-10% higher if upcoming earnings clear the elevated bar.
The disclosed $13.97 sale followed the earnings miss and was execution-based risk reduction, not a rejection of power demand.
The public newsletter text states a 20% US-ETF allocation, but supplies no target, stop or supporting thesis.