Linear's new AI feature could replace product decision makers
Plus: How to use MCP in your design workflows, Apple's AI comeback, Vibe coding is now mandatory at Shopify
This week’s briefing is supported by the new AI presentation tool Chronicle.
It’s an impressive new way to create designer-grade presentations at work. You can craft presentations that embed widgets from product tools like Miro, Figma, Notion, Google Sheets and others to bring your presentation to life.
Hi product people 👋,
As the dust settles on the mixed feedback after the launch of GPT-5, does this mean we’re entering an AI trough of disillusionment? Get my round-up on what people are saying about it here.
Plus, a new feature from Linear that could outsource most product decisions, how AI agents are impacting product strategy at Pinterest, Airbnb and Replit and Shopify’s COO announces some bold changes to the interview process for new product hires.
Happy Friday and have a great weekend!
Rich
Watch on YouTube | Follow on Substack Notes
Key reads and resources for product teams
New from the Department of Product Substack this week:
Deep - AI Agents in Practice
How are some of the world’s top companies using AI Agents? this Deep dive is designed to be your ultimate guide to getting up to speed with where we’re at in the real world quickly. We’ll take a closer look at how AI agents are being used internally at companies like Uber and McKinsey, new AI Agent capabilities, as well as new AI Agent features and products. 20+ real-world examples of AI agents in action, including internal use cases from Uber, McKinsey, Figma, Walmart and more. (Department of Product)
Process - How to use MCP servers in your design workflows
Paired with MCP servers, design systems become a productivity coefficient for AI-powered workflows, ensuring that AI agents produce output that’s relevant and on brand (Figma blog)
AI Fluency Course from Anthropic - a free course on developing AI fluency in 1 hour. developed by Anthropic in partnership with educators (Anthropic)
How AI can help tackle collective decision making (Harvard Business Review)
The one slide you can use to share SaaS product updates - Equals co-founder put together this example of a one-pager that captures all of the essential data points a tech leader wants to see to quickly get up to speed. (
)
How AI agents are impacting product strategy at Pinterest, Airbnb and Replit
Tech CEOs are increasingly having to defend themselves from questions about how AI might disrupt their businesses.
This week, Pinterest’s CEO says that agentic shopping where AI agents shop on behalf of users is still a long way off:
“I think this notion of an agent just going and buying all the things for you without you doing anything …I think that’s going to be a very, very long cycle for that to play out, both in terms of how the users think about it, where the users are going to be ready to just let something go run off and do everything for them, save for maybe some very utilitarian journeys”.
And Airbnb’s CEO Brian Chesky said he doesn’t think AI agents will turn his business into a commodity and that it is open to integrating with AI agents.
Replit’s CEO says the company is having to evolve its monetisation strategy to accommodate AI agents. It is shifting from flat-fee pricing toward “effort based pricing” pegged to how much compute power it took for an agent to complete a task because running advanced agents is getting more expensive, especially for complex tasks.
New product features and innovation this week
Linear has launched a new feature called “Product Intelligence”. It will assess new pieces of work, triage them and make suggestions about which team should pick up the work. In other words, it does a big chunk of what many product managers might do. Some teams may prefer to lean on AI to defer prioritisation decisions because it’s notoriously difficult for many companies to actually make meaningful decisions. Very curious to see how well this actually works - and what the unintended consequences might be.
In other news…
It’s been a week since the launch of GPT-5. It’s pretty clear by now that the launch was met with a rather mixed response. But as part of the release, OpenAI invited a bunch of developers to take part in a hackathon and showcase what they were able to build with GPT-5. Here’s a selection of products that were built at the event.
OpenAI also unveiled a new set of “Connectors” this week including support for Gmail, Google Calendar and Contacts. Once enabled, these Connectors let ChatGPT automatically reference information from these services in your conversations.
Perplexity is doing an excellent job of keeping itself in the headlines. This week, it released a new update which adds video models to its Pro and Max plans. When combined with the Comet browser, these video models can perform full end to end tasks. In this example shared by one of Perplexity’s product managers, Tyler Tate, he shares how you could use the new video models to complete a task like posting a property listing online. It takes a static image of a property, creates a flyover video of it and then shares it to LinkedIn. Some engineers are even using Perplexity to build prototypes.
And this week, Perplexity also made a $34.5 billion offer for Google Chrome - more than it has raised in funding to date. Google may be forced to sell Chrome as part of its ongoing antitrust case. According to The Information, Perplexity came close to acquiring Brave for $1 billion earlier this year while OpenAI has held discussions about acquiring The Browser Company.
Apple plots a comeback?
Apple is reported to be delaying the rollout of the new AI-powered Siri until 2027. Analysts have welcomed the move if it means avoiding the debacle of Apple Intelligence, but there’s also been reports this week that Apple is set to launch a completely new piece of hardware at the same time. Bloomberg suggests that this new robot resembles an iPad mounted on a motorized arm. The arm allows the display to swivel, reposition, and extend about half a foot in any direction meaning it can turn toward people who are speaking or summon its attention, mimicking a human head’s movement. Some people familiar with the product call it the “Pixar Lamp,” referencing the animated lamp from Pixar.
Tools you can use
Chronicle - craft stunning decks in minutes. Chronicle pairs expert storytelling with AI to elevate your next presentation. It also integrates with tools like Notion, Slack, Figma and others - with professional looking results.*
Profound - increase your product’s visibility in AI tools. This week, they raised $35 million in new funding.
Sheet0 - transforms web pages into structured spreadsheets.
Sixteen - a distraction blocking app that connects to tools like Linear to reward you once you’ve finished completing a task.
Shopify’s COO on the future of product management interviews
Get up to speed with vibe coding skills here:
📈 Product data and trends to stay informed
Google, Nvidia, Meta and Apple are some of the best companies for retaining engineering talent. Anthropic, OpenAI, and Meta are growing engineering teams 2-3x faster than they’re losing them, while Tesla, Bloomberg, and Walmart are losing talent faster than they can hire.
The report from SignalFire shows that the top reasons for retaining engineering talent include strong technical engineering leadership, career growth beyond the management track and a healthy pace of work and real psychological safety.
53% of engineers say that AI can already code better than humans according to a new survey of over 800 engineers.
Most employees are comfortable working with AI agents but far fewer are comfortable being managed by them. New research from Workday found that 30% of workers would be comfortable being managed by an AI Agent. To be honest, while this is being framed by Workday as a low percentage, I’m actually surprised it’s that high (!).
Other stats from the report worth knowing about for product teams:
82% of organizations are expanding their use of AI agents.
Fewer than 1% report no plans to use AI agents.
31% are in early production, 14% are looking to scale, and 37% are rolling out agents after pilot testing.
Downloads for AI companion apps have surged to 220 million globally, up 88% year-over-year, and revenue per download has more than doubled since 2024. Character.AI told Wired magazine that it's generating revenue at a run rate of $30M+ and has 20M MAUs who spend, on average, 75 minutes a day chatting with a bot.
Paid subscribers get the full DoP Substack including: The Knowledge Series for sharpening your tech skills, AI tutorials for putting AI into practice at work and DoP Deep dive reports to learn from the world’s top tech companies.
Apologies for the incorrect link on the Linear article - this has now been updated! https://linear.app/changelog/2025-08-14-product-intelligence-technology-preview