Community Buzz Ep. 23: Google's $500B Revenue and GPT-Rosalind Using Agents for Drug Discovery

Lots of new buzz in the community today—Twitter and Reddit have been totally dominated by two things this week.

One is Google's earnings report that just maxed out the AI narrative, and the other is OpenAI getting thousands of agents to run drug discovery for scientists.

You're not wrong—the community mood this week is weird. It's part celebration, part defiance.

Let's cut to the chase: what people are really arguing about this week isn't the tech—it's money and trust.

Then let's start with the money. Li Bo, you first—why did Google's earnings report blow up like that?

Pichai's tweet: Q3 wrapped up, annualized revenue run rate hit $500 billion, Alphabet revenue up 25%.

But that's not the point. The point is the numbers he casually dropped—once you hear them, you'll get why the community exploded.

Go ahead, I'm taking notes.

Search queries hit their highest year-over-year growth in five years—even commercial queries are up. Cloud revenue grew 103%.

And first-party model APIs are processing nearly 32 billion tokens per minute, up 40% quarter-over-quarter.

32 billion tokens per minute? I can't quite wrap my head around that scale—what's that equivalent to?

Think of it this way: Google's own model API has become a massive high-traffic infrastructure, not just a lab demo.

He also named Argon and Carbon as leading the frontier, plus Genie 4, the latest world model.

I saw people on Reddit straight-up mocking it, saying search results keep getting worse while revenue keeps climbing.

That's the biggest controversy this week—the gap between beautiful earnings and product experience.

Right, that post's comment section got pretty heated. The opposition says it's classic AI summaries polluting search results.

The other side fires back directly: query volume is at a five-year high, users vote with their feet—just because you think it's bad doesn't mean everyone does.

The neutral camp is more interesting—they say it's actually two different groups: one searching for information, one searching for answers. The experience isn't the same thing at all.

I buy that. When I worked on products, I noticed the same thing—users complain, but they still use it. Switching costs are too high.

So the conclusion is: Google isn't winning on tech reputation this round—it's winning on distribution and scale, and the moat is still widening.

Okay, second item. I watched that OpenAI GPT-Rosalind post twice.

That one's legit hardcore—not benchmarks, but running scientific hypotheses on real mouse brain data.

The DCTN1 gene is linked to Perry syndrome. When knocked out, cholesterol synthesis-related gene expression goes up.

And then GPT-Rosalind's explanation was that cholesterol might not be reaching where neurons need it?

Exactly—it translated a gene expression phenomenon into a testable mechanistic hypothesis. That leap is key.

The most explosive part is the scale: thousands of agents in parallel, linking expression results from millions of brain cells with literature in 32 minutes.

32 minutes? A PhD student would be at it until graduation.

That's why some on Twitter are saying this is the first time agents have done something structurally meaningful in research—not just writing summaries.

But the calm folks on Reddit point out that a mechanistic hypothesis is just the starting point—validation still needs wet lab work.

Exactly—the opposition worries this news gets read as 'AI discovers drugs,' but that's miles off. It just narrows down the candidates.

But one neutral take I really agree with: the value of agents isn't making conclusions for scientists—it's saving them screening time.

So this and the Google story at the start are actually the same trend: AI moving from demos to production systems.

Right—one is scaling traffic and revenue, the other is scaling the research process. The common thread is competing on systems, not point solutions.

So how would you sum up the community mood this week? Excitement or anxiety?

I'd say excitement mixed with defiance—people acknowledge the scale while desperately looking for cracks in it.

Agreed. And the fiercer the argument, the more they care—nobody argues this long about something unimportant.