AVGO Wednesday: Does It Still Feel Inevitable?

Four price targets I keep seeing pop up:

• Evercore ISI: Buy – $582
• Citigroup: Strong Buy – $500
• JPMorgan: Overweight – $500
• Morgan Stanley: Buy – $470

Broadcom reports Wednesday, June 3, 2026 after the close (5:00pm ET).

I’m going to say the quiet part out loud: the market’s already leaning toward “good quarter” as the default.

So the bar isn’t “beat by a little.” The bar is: do you leave that call feeling like the next 12–18 months are getting clearer, or like they’re getting fuzzier?

Here’s the thing. Broadcom has positioned itself as the company that can help hyperscalers do two hard things at once: build custom AI compute, and move all that data around without the whole system choking. Custom accelerators plus networking isn’t a cute bundle. It’s the bundle.

And then there’s the line everyone quotes back at them: AI chip revenue above $100 billion in 2027.

What’s interesting is that it’s not even the number that makes people nervous. It’s the confidence. When a CEO says “line of sight,” investors hear: customers are committed, the roadmap is real, and capacity planning is mostly handled. When that wording changes, even slightly, it changes how people model the whole thing. Fast.

Slight tangent, but it matters: custom silicon is as much a finance decision as it is an engineering decision. Power, cost per inference, control of supply… those are board-level conversations now. That’s why the opportunity can be sticky. That’s also why it can get paused without warning if budgets tighten or priorities shift.

Last quarter’s anchors were strong:

• Q1 FY2026 revenue: $19.3B (about +29% year over year)
• Q1 AI semiconductor revenue: $8.4B (about +106% year over year)
• Q2 FY2026 revenue guidance: about $22.0B (about +47% year over year)

But I don’t think the “anchors” are where the decision gets made this time. It’s the little stuff: how they talk about ramp timing, whether programs are overlapping smoothly, and whether AI networking demand still feels like a companion wave (not an afterthought).

Also: concentration risk. People nod past it because the customers are high quality. Still, a small number of buyers can move a lot of dollars with one internal email. If spend shifts by a quarter, the market tends to punish first and ask questions later.

So Wednesday night, I’m listening for three things:

1) Do they repeat the 2027 view in plain language, or do they hedge it with qualifiers?
2) Does “custom” sound like it’s broadening, or just deepening with the same few names?
3) Does networking still sound like demand is being pulled by real deployments, not lab builds?

At first glance, this is just another earnings call. It isn’t. It’s a confidence test.

Take a closer look Wednesday night. If management sounds almost boring – specific, consistent, a little repetitive – that’s usually when the longer-term story is actually working. If it gets vague, the questions get louder.