The Twenty Minute VCAnthropic's Raise & What It Means for Potential IPO? Mag7: Google & Amazon Up, Meta & Microsoft Down
At a glance
WHAT IT’S REALLY ABOUT
Mag7 earnings reveal AI capex winners, SaaS rebound, private-market frenzy
- Alphabet is framed as the earnings-season standout as Search resilience and surging Cloud backlog/capex signal that incumbents are “leaning in” rather than being disrupted by LLMs.
- Microsoft and Meta are contrasted on AI spend: Microsoft’s growth is portrayed as heavily AI-dependent while Meta’s larger capex is punished because the ROI path is harder to model in spreadsheets.
- Palantir’s blowout quarter is interpreted as evidence that enterprises want large, CEO-level AI transformation programs and will pay $10M+ checks to credible, implementation-heavy vendors.
- Public SaaS shows signs of life (Atlassian, Twilio) but the hosts argue “SaaSpocalypse” isn’t broadly over; winners either monetize AI into the base and/or add net-new customers, especially near infrastructure.
- Private markets look overheated yet rationalized by AI capex math: Anthropic raising huge sums reduces IPO urgency, while Sierra’s premium valuation bets on labor replacement and category expansion beyond customer support software.
IDEAS WORTH REMEMBERING
5 ideasThe Mag7 are entering an unusually aggressive, incumbent-led investment regime.
The discussion characterizes the quarter as “the most aggressive” because the largest firms are simultaneously accelerating growth and expanding capex fast enough to consume much of free cash flow, aiming to prevent disruption rather than respond defensively.
Google’s win is about ‘everything working at once,’ not just AI.
They highlight Search/ads staying strong despite LLM fears, while Cloud backlog nearly doubling signals sustained demand; the main constraint becomes allocation of GPUs/compute across internal and customer needs.
Hyperscalers’ AI revenue is still largely ‘selling picks and shovels’ to LLM labs.
A core critique is that Big Tech growth is boosted by hosting/training spend and reselling tokens, while foundational-model IP value may accrue disproportionately to OpenAI/Anthropic—creating uncertainty about who ultimately captures the profit pool.
Microsoft’s valuation sensitivity to AI is higher than many assume.
Rory argues that excluding AI-related growth, Microsoft’s broader business looks flat, implying that if the AI bet disappoints, growth expectations—and therefore the equity multiple—could reset meaningfully even if the company remains profitable.
Meta is penalized because its AI capex is harder to tie to near-term, attributable revenue.
Unlike Google/Amazon, Meta’s justification is framed as qualitative (future experiences, chatbot-native social) or indirect (ad lift) and therefore resists the Wall Street spreadsheet approach to ROI, increasing the discount rate investors apply.
WORDS WORTH SAVING
5 quotesThis is leaning in like you've never seen leaning in before. This is the top of the distribution pulling away.
— Rory O’Driscoll
Without the AI initiative, Microsoft the corporation is flat revenue.
— Rory O’Driscoll
Anyone on LinkedIn that talks about their team, fire them. "My team this." They're all so precious about their team. Lead from the effing front with AI.
— Jason Lemkin
Two of them full-time, two full-time human equivalents, $254 a month.
— Jason Lemkin
There is no such thing as too much cash on your balance sheet. There is no such thing.
— Rory O’Driscoll
High quality AI-generated summary created from speaker-labeled transcript.
