Did Google Just Fall Behind Again?, iPhone Fold Cometh, Anthropic Files To Go Public

Episode Summary

Executive Summary: MG Siegler and Alex Kantrowitz argued that Google’s AI effort is strategically behind OpenAI and Anthropic, especially in coding and agentic “super app” workflows, while Apple appears to be heading into a transitional WWDC focused on fixing Siri rather than debuting major hardware. They also assessed Meta’s messy branding and business shift, and ended with breaking news: Anthropic confidentially filed to go public, escalating pressure on OpenAI’s own IPO timing.

Main Topics: Google’s AI strategy and the I/O problem (Priority: 5/5): The hosts argued Google I/O felt underwhelming because Gemini 3.5 Pro was not ready, leaving Google to showcase Flash instead of its flagship model. They see Google as strong on search integration but behind in coding and agentic product experiences. The rise of AI “super apps” and agentic computing (Priority: 5/5): They debated whether ChatGPT, Claude, Codex, and similar tools are evolving into super apps that combine chat, coding, browser control, and task execution. The key idea is that these tools may become the interface for most digital work. Google’s structural disadvantage versus OpenAI and Anthropic (Priority: 5/5): MG argued Google’s size, legacy product silos, and web-centric culture make it harder to reorient around AI-native workflows. By contrast, smaller labs can move faster and integrate products more tightly. Apple WWDC, Siri, and the iPhone fold (Priority: 4/5): They previewed WWDC as likely a software-first, transitional event centered on Siri improvements rather than foldable hardware or glasses. The iPhone fold could arrive later, with a design that may emphasize typing and a new form factor rather than just a bigger display. Meta’s branding, morale, and business model pressure (Priority: 4/5): The conversation described Meta as confused on product naming and increasingly reliant on monetization experiments like subscriptions. They suggested the company may need to diversify beyond ads and could even become a cloud provider due to huge CapEx spending. Anthropic’s confidential IPO filing and OpenAI pressure (Priority: 5/5): The final breaking news was that Anthropic confidentially filed to go public, potentially forcing OpenAI to clarify or accelerate its own IPO plans. The move matters because Anthropic’s growth and profitability narrative could compare favorably to OpenAI’s.

Key Arguments: Google’s I/O was weak because its flagship Gemini 3.5 Pro was not ready, undermining the expected developer-conference cadence. Google is not competitive enough yet in AI coding products like Claude Code and Codex, which are becoming the gateway to agentic workflows. The real AI battleground is not just model quality but who controls the interface to browsing, email, and task execution. AI products are evolving into “super apps” in a Western, AI-native sense: one tool that can chat, code, browse, and act on behalf of the user. Google’s internal structure makes it harder to move fast: product silos are used to being pushed from a central AI team, but the new paradigm requires products to converge around AI. Apple likely cannot ship a major foldable or glasses product at WWDC because the event will be dedicated to catching up on Siri and AI features. Meta’s many subscription tiers and confusing branding suggest a late-stage monetization push rather than a clear product vision. Anthropic’s IPO filing increases pressure on OpenAI because public investors will compare their revenues, profitability paths, and product momentum directly.

Data Points: Google I/O timing: June 1 / summer rollout language - Google announced many AI features as coming later, especially Gemini 3.5 Pro, rather than shipping them at I/O. Gemini 3.5 Flash: Released at I/O - Google showcased Flash as its available model, while the stronger Pro version was still pending. Gemini 3.5 Pro availability: “next month” / sometime in June - The hosts discussed Google saying the flagship model would come after I/O. Anthropic ARR: “closing in on 50 billion in ARR” - MG referenced market chatter suggesting Anthropic’s business has scaled rapidly. OpenAI MAUs: 900 million - MG said OpenAI would likely point to consumer usage as a differentiator. Gemini MAUs: 900 million - MG noted Sundar Pichai announced Gemini also reached 900 million monthly active users. Meta subscription prices: $2.99, $3.99, $7.99, $14.99, $19.99, $49.99 - The transcript listed multiple Meta/WhatsApp/Facebook/Instagram subscription tiers as an example of confusing monetization. Ulta Beauty deployment: 1,300 stores - Mentioned in the intro as part of ServiceNow enterprise AI examples. Enterprise AI project failure rate: 80% to 95% - Used in the ad read to explain why AI projects need better implementation support. Anthropic fundraising: $65 billion - Mentioned as the amount Anthropic just raised before its confidential IPO filing. OpenAI fundraising: $122 billion - Mentioned as recent capital raised by OpenAI. Potential cloud lease cost: $15 billion per year - MG cited a report that OpenAI may be paying huge sums to lease compute capacity.

Pivotal Quotes: "“Google looks sort of silly for not having their model ready to go.”" — MG Siegler: On Google I/O underdelivering because Gemini 3.5 Pro was not launched in time. "“The agents and the chatbots are going to merge.”" — Alex Kantrowitz: On the thesis that AI tools will become the main interface for computing and web interaction. "“We’re not quite in Microsoftian territory, but we’re close. Maybe Meta should just stick with ads.”" — Alex Kantrowitz: On Meta’s product confusion, subscription sprawl, and pressure to find a clearer business model.

Implications: AI competition is shifting from model demos to control of workflows, interfaces, and distribution. Google risks falling behind if it cannot unify products fast enough, Apple may need a strong Siri reset, and Anthropic’s IPO move could reshape the AI investment narrative and pressure OpenAI.