Some thoughts -

We have 4 of the hyperscalers going into earnings today $GOOG, $AMZN, $MSFT $META - market will probably be watching the hyperscalers for their capex guidance. Any slowing down will be bearish for AI tickers while the market reaction to upside surprises can be a good gauge of strength. As a reminder, post $TSMC blow out earnings, it actually suffered a drop.
Expectations going into earnings
$META - the largest announced step up in capex, 2026 guidance of $115-135b from 2025's 69B. If $META delivers anything close to 30B for Q1, that will be an indication for urgency
$AMZN - 2026 guidance of $200b - will be interesting to see if the Q1 numbers come in close to the avg run rate of 50b
$GOOG - 2026 guidance of 175-185b capex,
$MSFT - 2026 guide of 110-120b, quarter ending dec'25 was already 37.5b - the commercial backlog was at 625B , of which 45% is tied to OAI.
the most interesting for me is $MSFT - since the Oct'25 earnings release, $MSFT was rerated down harshly by the market for very good reasons. However, going into Apr, we had 2 big changes
1. OAI restructuring agreement - financial derisking from the revenue sharing
2. Voluntary separation program - signalling to the market that management is aware of the hit on EPS via depreciation from the ramp up in capex.
Beyond hyperscaler capex - if we continue to hear the line of capacity constraints - then it is good to bear in mind what these constraints are - namely, compute, power and increasingly memory.
$BE call today indicated that $ORCL project jupiter is pivoting to fuel cells over a backlash on emissions - immediate note is that this will reduce the backlog for gas turbines as DCs work around for solutions from grid issues.
In a world of persistent energy shortages and high energy prices, it is increasingly likely that nuclear will be held up as the ONLY viable solution to power the world's electrical needs in a post AI world. IMO - Nuclear will become a core theme on bottle necks for the market
Good luck!