Google AI Launches Agentic Capabilities for Search
Today's curated insights on AI product management from X/Twitter across 60+ expert sources and YouTube channels.
Google AI Launches Agentic Capabilities for Search
From X
AI Product Launches & Updates
Natural language stock screening for Indian equities: Arav Srinivas @AravSrinivas announced Perplexity Finance, enabling simple AI-driven stock screening across Indian markets.
Agentic capabilities in Search: Google AI @GoogleAI launched agentic capabilities in AI Mode for Search, streamlining complex queries like dinner reservations into a single workflow.
Government-grade AI platform: Logan Kilpatrick @OfficialLoganK unveiled Gemini for Government, combining Google Cloud, state-of-the-art models, and agentic solutions for under $0.50 per agency.
AI Tools & Applications
AI at work one-pager: Aakash Gupta @aakashg0 highlighted Anthropic’s concise one-pager on using AI at work, providing a go-to resource for PMs integrating AI into daily workflows.
Notion integration: v0 @v0 announced automatic embedding of v0 prototypes in Notion, enabling side-by-side writeup and prototype sharing.
Custom link-in-bio template: Bolt @boltdotnew released a free Bolt template for fully customizable link-in-bio pages, eliminating subscription fees and platform revenue cuts.
Product Management Insights & Strategies
Interview system design blueprint: Aakash Gupta @aakashg0 shared a system design blueprint that covers 95% of common PM interview scenarios.
Prototyping perspective: Dharmesh Shah @dharmesh noted that a prototype is worth a thousand pictures and a product is worth a thousand prototypes.
Executive influence coaching: Shreyas Doshi @shreyas published a video on influencing executives on contentious issues, using AI-anonymized coaching dialogue with a top-tier startup PM team.
AI Industry Developments & News
AI inference energy impact: Jeff Dean @JeffDean published a technical paper showing Gemini inference uses just 0.24 Wh per text prompt, underlining AI’s environmental footprint.
Nuclear safeguards classifier: Anthropic @AnthropicAI partnered with NNSA to build a classifier that detects nuclear weapons queries while preserving legitimate research access.
GPT-5 Pro scientific breakthroughs: Kevin Weil @kevinweil reported GPT-5 Pro performing novel mathematics and theoretical physics, signaling major advances in large-language models.
From YouTube
I Spent 72 Hours Studying How Hormozi made $83m in 5 days (Ads, Emails, Pages, Everything)
Greg Isenberg • August 21, 2025
Greg Isenberg and Jonathan Courtney dissect Alex Hormozi's record-shattering five-day, $83 million book launch, exposing his phased webinar email sequences, ad creative strategy, long-form sales pages, referral program, and physical mailers.
Key Takeaways:
- Hormozi's webinar funnel ran in four phases—about 6 pre-registration emails, 4 registered reminders, 4 live-stream alerts, and 6–10 post-event follow-ups—using hook-story-offer direct-response copy.
- He activated an affiliate contest awarding prizes to the top 10 and top 50 referrers, tracking live referrals and announcing winners on day three to amplify registrations.
- To boost attendance, Hormozi spent roughly $450K per day on marketing and mailed physical letters at approximately $0.50 each as unexpected touchpoints.
How Intercom rose from the ashes by betting everything on AI | Eoghan McCabe (founder and CEO)
Lennys Podcast • August 21, 2025
Eoghan McCabe recounts how he salvaged Intercom by slashing 40% of its workforce, overhauling its values, and investing $100 million to pivot to an AI-first model with Fin—an agent built just six weeks after GPT-3.5 that’s driven 300%+ growth and is poised to hit $100 million ARR in under three quarters.
Key Takeaways:
- Intercom endured five straight quarters of net new ARR decline, prompting McCabe’s “wartime” return to cut 40% of staff, rewrite company values, and back an AI-first strategy with a $100 million investment.
- Six weeks after GPT-3.5’s release, Intercom’s ML team delivered Fin’s beta; Fin grew from $1 million to $12 million ARR in its first year and now posts mid-eight-figure ARR at over 300% annual growth.
- Intercom wrote off $50 million in legacy ARR to replace complex pricing with a simple outcome-based model, charging $0.99 per ticket resolved by Fin—roughly one-twentieth of industry rates—to boost resolution rates and customer loyalty.