Sid Jayakumar

Some data / stats on AI related products at Microsoft. Real revenue! Azure AI Services - $5b run rate up 900% YoY - 60k customers up 60% YoY - Responsible for ~8% of overall Azure growth this Q Developer Tools - GitHub at $2b run rate (It was ~$1b in Sept '22) - GitHub CoPilot: ~$300m ARR. They accounted for 40% of overall GitHub growth this year and is "larger than GitHub when Microsoft acquired it" - 77k orgs using CoPilot up 180% YoY - 480k orgs using AI powered capabilities in Power Platform. 48m MAUs up 40% YoY Future of Work - Daily users of Office Copilot doubled QoQ - Office CoPilot customers grew 60% YoY - Customers with >10k seats of Office CoPilot doubled QoQ - 1k customers used CoPilot for Security

Everyone who arrives from the 3rd world deletes the entire tax contribution of a higher rate taxpayer.

This was a vile terrorist attack that attacked Jews, because they are Jews. Antisemitism is a hatred that is rising, once again. Britain must defeat it, once again. To every Jewish person in this country: I promise that I will do everything in my power to guarantee you the security you deserve.

The amount of ambition in San Francisco is hilarious and amazing. Last night, Chris and I went to a random house party. I ask the first guy I talk to what he's working on: "I'm building a massive multi-player virtual reality world." "Wow, that's awesome. What's the end game?" "I want it to get so big, and to require so much power, that we need to build a Dyson sphere around the sun to capture enough electricity"

Vibe coding is another perfect example of how AI lowers the barrier to entry: Then: your ability to execute Now: your ability to invent ideas worth executing

to me, the greatest sadness of all is the banality of most people's lives i'll spend time with my non-elite friends, and it's clear that their lives are devoid of heroism and lore they lock in/clock in, but they fight no great battles, they dash no great foes they drop by a club or bar, but find no great loves there they travel the world, but it is by no means an adventure they return home, but they are not welcomed by a triumph – no, all they have are their succulents and the pile of dirty laundry they failed to do before they left

The idea that a LLM could build SpiderRock or Enfusion or RiskMetrics or any of the highly domain knowledge intensive and performant SaaS applications in finance is completely absurd - in any time frame much less a few hours

A founder hit me up after I posted this and showed me his new highly specific, highly domain-knowledge-driven app which has very achievable goals that align what I'd expect should be achievable by LLMs in finance (automation of rote/boring/junior work). Perfect.

I still haven't heard a good answer to this question, on or off the podcast. AI researchers often tell me, "Don't worry bout it, scale solves this." But what is the rebuttal to someone who argues that this indicates a fundamental limitation?