In partnership with
We track Product so you don't have to. Top Podcasts summarised, the latest AI tools, plus research and news in a 5 min digest.
Hey Product Fans!
Welcome to this week’s 🌮 Product Tapas.
New here? We're your shortcut to staying sharp. Essential stories, practical tools, real insights.
For the best reading experience, check our web or app version and sign up for future editions here.
What’s sizzling? 🧑🍳
Anthropic claims world's best coding model whilst slashing prices (convenient timing), OpenAI finished hardware prototypes for devices nobody asked for (because shopping research features weren't enough surface area), and Google's unifying ChromeOS with Android into "Aluminium OS" (admission that ChromeOS was a failed experiment or just efficiency drive?). .
On the menu:
🌍 Not Boring → Model wars choreography, visibility panic, platform admissions
⌚️ Productivity Tapas → AI file intelligence, prompt libraries, agent workspaces
🍔 Blog Bites → Agent-first commerce, product as PE lever, blockchain capital markets
🎙️ Pod Shots → Jason Knight helps you escape the feature factory and build strategic partnerships with sales
Let's go 🚀
📰 Not boring
🤖 The AI Model Wars - the Catwalk as I’ll now call it
Anthropic released Claude Opus 4.5 at $5/$25 per million tokens, claims world's best coding model
OpenAI launched free shopping research feature in ChatGPT
OpenAI finished first hardware prototypes, plans to reveal device within 2 years
Google launched Nano Banana Pro image generator powered by Gemini 3
Google Workspace rolling out Nano Banana Pro to enterprise accounts
Google Slides added "Beautify this slide" button with AI enhancement
OpenAI introduced ChatGPT group chats for multi-user AI collaboration
Gemini 3.0 benchmarks show outperforming in 8 of 12 categories
White House announced Genesis Mission—national AI platform integrating federal datasets, supercomputing and models to accelerate scientific discovery across manufacturing, biotech, nuclear and semiconductors
Replit launched Design Mode with AI-generated full UI designs from prompts
The model release cadence feels increasingly choreographed - Anthropic drops pricing, everyone claims "best ever", Google pushes enterprise rollouts. OpenAI pivoting to shopping research and hardware prototypes - they want to be everywhere: software, commerce, devices, voice. Replit's Design Mode is the real workflow shift - full UI from prompts changes rapid prototyping entirely. Genesis Mission positioning AI as infrastructure for scientific discovery shows governments treating this as critical national capability, not consumer feature.
💰 Money & Infrastructure
Adobe buying SEMrush is the clearest signal yet: nobody knows how to stay visible in the AI era, so you buy the company tracking visibility. Meta discussing Google TPUs is either genuine diversification or negotiating leverage against Nvidia. Either way, infrastructure decisions made now lock in advantages for years. The real question: does SEMrush's search visibility data even matter when ChatGPT answers questions without showing sources? Google needing to double capacity every six months shows demand isn't slowing - but at what point does the infrastructure spend stop making economic sense?
📱 Platform Shifts
Google unifying ChromeOS and Android into Aluminium OS desktop platform
Alibaba launched Qwen AI app with 10M+ downloads in first week
Android Quick Share now works with Apple AirDrop
Ocado robotic warehouse tech struggling as Kroger shuts 3 facilities, taking $2.6bn charge
WhatsApp rate-limiting bug let researchers scrape 3.5 billion phone numbers and profile data
Alibaba's 10M downloads in week one proves distribution wins again. Google merging ChromeOS into Android feels like quietly admitting defeat—or maybe just acknowledging phones won. Android and Apple finally playing nice on file sharing shows even fierce rivals bow to user frustration. WhatsApp leaking 2bn profiles through a rate-limiting bug whilst preaching privacy is peak 2025. And Ocado's $2.6bn warehouse write-down? A reminder that automating atoms remains brutally hard, no matter how good the bits get.
⚙️ Enterprise & Productivity
OpenAI adjusted data hosting rules for regional compliance
Figma launched Make features with AI-powered design automation (generate design variants, import npm packages, create developer-ready code)
CB Insights released comprehensive AI agent market map showing landscape exploded from 300 to thousands of players
OpenAI expanding data residency shows enterprise compliance concerns aren't going away - governments want data sovereignty, enterprises want regulatory cover. Figma's Make features show how AI's augmenting creative tools - generate variants, not replace designers. CB Insights mapping the agent market suggests structure's emerging: infrastructure layer vs application layer, agentic commerce vs agentic productivity.
🌍 Everything Else
iOS 27 quality improvements prioritised over new features for 2026 release
YouTube is working on a feature that will fix the messy home feed
Microsoft's AI chatbot Copilot leaves WhatsApp on January 15 (will anyone notice??)
Pocket Casts now lets you create playlists of favourite podcast episodes
Phictly launched social app bringing people together over books and TV shows
Mixup released Mad Libs-style app for creating AI images from photos, text, and doodles
Ben Evans has released his latest presentation exploring macro and strategic trends in the tech industry. It's a great read as always
Apple choosing polish over features: iOS maturity or AI not ready for prime time? YouTube promising (again) to fix the home feed and Microsoft yanking Copilot from WhatsApp - platform cleanup begins. Meanwhile the weird edges keep shipping: podcast playlists (finally), literary social networks, Mad Libs meets AI image gen. The gap between foundation model drama and consumer apps keeps widening - one moves at breakneck speed, the other iterates on features we should've had years ago.
Learn AI in 5 minutes a day
This is the easiest way for a busy person wanting to learn AI in as little time as possible:
Sign up for The Rundown AI newsletter
They send you 5-minute email updates on the latest AI news and how to use it
You learn how to become 2x more productive by leveraging AI
⌚ Productivity Tapas: Time-Saving Tools & Workflow Automation
Poly: store, browse, research, and organise your files with AI. Intelligent browsing system to search inside videos, documents, photos. Available on web and desktop
Snippets AI: Seems very niche but aiming to be the single store of all your top prompts - get instant access to proven prompts, save, adapt, and reuse them across top AI models.
Felo: another AI agent collaborative workspace aiming to create an intelligent canvas where you can work together with multiple agents and people simultaneously
Remember. Product Tapas subscribers get our complete toolkit - 460+ personally tailored, time-saving tools for PMs and founders. Your shortcut to efficiency and what's hot in product management 🔥
Check the link here to access.
🍔 Blog Bites - Essential Reads for Product Teams
Product Design: When Your Customer Is an Algorithm, Not a Human
Bandan Singh explores how AI agents are fundamentally reshaping ecommerce design, requiring product teams to optimise not just for human delight but for machine reasoning. As agents are projected to handle 25% of ecommerce by 2030 ($500 billion annually), the traditional focus on beautiful interfaces must expand to include structured, machine-readable product data. Read the full article here.
💡 "The agent doesn't experience a moment of 'Oh, I love this design!' Instead, it systematically filters against parameters: price ceiling, specific attributes, availability, shipping time, and seller reliability."
Key Takeaways
• The Dual Commerce Reality: You're now serving two channels—human browsers who experience your brand, and AI agents who only see structured data. Both need distinct strategies.
From Visual Delight to Data Transparency: Agents don't appreciate aesthetics—they evaluate certifications, inventory accuracy, and machine-readable policies. Discoverability is now about data architecture, not search rankings.
Machine-Readable Data as Competitive Moat: Structured data is essential infrastructure. One study: semantic schema drove 192% more add-to-carts and 278% more transactions.
Recommendations Shift from Crowd to Context: Agents solve for specific intent, not "what similar users liked"—potentially picking a 300-review product over a 15,000-review bestseller. Winner-take-all dynamics are breaking down.
Designer Role Expansion: Designers must now own structured data and conversation design alongside visuals—auditing whether AI can extract product value from current data structures.
Strategy: Why Product Management Is PE's Most Undervalued Growth Lever
Craig Unsworth explores how product management has evolved from a cost centre to a critical value creation engine in private equity portfolios. He demystifies the discipline through practical frameworks that connect customer needs, business goals, and technology capabilities to measurable financial outcomes. Read the full article here.
💡 "Product is the operational bridge between customer experience and financial performance."
Key Takeaways:
• The Product Triad: PMs balance customer needs, business goals, and tech capabilities—acting as the "capital allocation function" for your roadmap. They manage a portfolio of bets that move metrics like revenue, retention, and efficiency.
•Triple Growth Impact: Product influences all three PE growth levers—acquisition (better onboarding), retention (workflow automation), and monetisation (upsell paths). A hypothetical £1m product investment generated £3.8m EBITDA impact—£38m in enterprise value at 10x.
•Systematic Customer Centricity: Strong product orgs listen systematically through analytics and interviews, connect insights to clear priorities, and measure outcomes not outputs. This reveals what customers actually do, not just what they say.
•Maturity Markers: Five traits define strong product organisations—clear PM ownership, evidence-led prioritisation, cross-functional squads, regular delivery cadence, and outcome metrics over feature counts. Product is a decision-making system, not just a department.
•Emerging Trends: AI-augmented PM accelerates discovery at scale, PLG drives acquisition through self-serve, customer success integration tightens churn feedback loops, and outcome-based roadmaps shift focus from "build X" to "achieve Y impact."
Fintech: How Blockchain Is Rebuilding Capital Markets From the Ground Up
Mike Cagney explores his journey from SoFi's record-breaking $1 billion raise to building Figure Technologies, the blockchain-based fintech that's processed over $50 billion in transactions and is now heading for IPO. The serial entrepreneur reveals how displacing "trust with truth" through blockchain technology is fundamentally reshaping lending, trading, and capital allocation. Read the full article here.
💡 "Blockchain's ability to displace trust with truth creates a paradigm that allows you to overhaul everything in markets—from trading to financing."
Key Takeaways:
• The Three Pillars of Blockchain Value: Figure's model delivers transactional efficiency (100+ bps cost reduction), liquidity (real-time loan trading via on-chain assets), and financing (direct lending without custodian intermediaries). Result: largest non-bank HELOC lender, processing $100m+ daily.
• Strategic Product Selection: Cagney chose HELOCs not to build a lending company, but because banks refused to move first on blockchain. He needed a greenfield category to prove the tech—transforming a 45-day process into five minutes without degrading credit performance.
• Culture Before Capital: Unlike SoFi's meritocratic approach that created friction, Figure's founders spent a month defining cultural norms before raising money or building product. Cagney learned an aligned organisation outperforms one with superstars at odds.
• Democratised Prime's Vision: The platform brings institutional-grade prime brokerage to everyday investors—borrowing against digital assets and controlling stock lending directly to capture premiums currently flowing to brokerages. Figure is positioning for Web 3.0's "Magnificent Seven."
• B2B2C as Growth Accelerator: Partnering with 170+ origination partners like Guaranteed Rate delivered market acceptance, product feedback, and distribution scale. This model, combined with removing capital market risk from partners, drove explosive growth beyond what D2C alone could achieve.
🎙 Pod Shots - Bitesized Podcast Summaries
Remember, we've built an ever-growing library of our top podcast summaries. Whether you need a quick refresher, want to preview an episode, or need to get up to speed fast - we've got you covered.
Check it out here
🤖 🤝 How to Escape the Feature Factory and Build Strategic Partnerships with Sales
This one hits close to home. I've lived through the product-sales dysfunction at multiple companies, and it's exhausting. The constant escalations, the "strategic client" requests, the feeling of being stuck in a feature factory with no clear strategy.
💡 "Sales is responsible for this year's revenue. Product management is responsible for next year's revenue. But here's the reality: every single sales rep will tell you they need product's help to hit THIS quarter's numbers."
Jason Knight is absolutely spot-on with this breakdown. What I love about this episode is how practical it is - he doesn't just diagnose the problem, he gives you actual frameworks and tactics you can use tomorrow. The decision matrix for handling sales requests alone is worth the read.
Jason Knight, Hustle Badger, Product Management, Sales Alignment, B2B SaaS
🎥 Watch the full episode here:
📆 Published: 10th November 2025 🕒
Estimated Reading Time: 5 mins. Time saved: 43 mins! 🔥
Key insights from the full article:
⏰ Time horizon conflict — Sales has quarterly targets; product builds for next year. The tension is structural, not personal—understanding this changes everything
🏭 Feature factory symptoms — When product lacks vision, sales requests fill the vacuum. Escaping requires clear strategy, objectives, and market evidence
💰 Revenue debt compounds — Short-term, misaligned deals create long-term strategic paralysis. Like technical debt, you pay interest for years
🧭 The decision matrix — Evaluate requests on two dimensions: alignment to objectives + effort required. Hard no for low-alignment/high-effort, regardless of "strategic" clients
🚀 Pipeline visibility prevents crises — Get ahead of the wave by participating in deal reviews. Don't wait for escalations to understand what's being promised
📊 Track outcomes, not output — When you build for a deal, did it actually close? Create feedback loops to understand if your efforts have the intended impact
☕ Start with coffee — Build human relationships before fixing processes. Take sales reps out individually to understand what's really motivating them
🎯 Frame in financial terms — Leadership understands CLTV, cycle time, and retention. Translate product strategy into metrics executives care about
🤝 Everyone wants a great product — Sales teams aren't the enemy. They're operating under different constraints. Product's job is creating conditions for better collaboration
That’s a wrap.
As always, the journey doesn't end here!
Please share and let us know what you would like to see more or less of so we can continue to improve your Product Tapas. 🚀👋
Alastair 🍽️.