Articles
- Your Agent's First Reply Is a Retention Decision
Default prompts, agent tone, and eval scores predict AI feature retention. Here's what the data shows and what to fix on your team first.
- Attention is the product. So is walking away from it.
Tinder made a TikTok dating show while a $59 gadget locks your phone. Here's what the engagement versus attention split means for your product team.
- Your AI Product Doesn't Behave Like Your Old One. Your Metrics Should Change Too.
AI product OKRs should measure what users do, not model accuracy. Here's how to write behavior-based key results and version-control your PM workflow.
- Your AI Budget Just Became a Line Item.
Token rationing is here. Here's how design and product leaders should govern AI inference cost before finance does it for you.
- Your A/B Test Wins Are Lying to You
AI floods your team with variants while your experimentation discipline slips. Here's what breaks, and what to check before you ship the next winner.
- Cheap agents just got here. Test them on your own work before you ship.
Anthropic's Sonnet 5 makes agentic AI cheap, and eval shops like Arena are booming. Here's how to benchmark models on your tasks before betting a feature.
- Your CSAT Score Is Lying to You, and Voice Input Is About to Change Your Team's Workflow
CSAT breaks down as AI handles more support conversations. Here's how to measure customer experience at scale and why voice input is worth a real look.
- Your AI feature works. Nobody can tell if it's lying.
New human-AI design principles give product teams real guardrails for the AI features they already shipped. Here's what changed and what to do Monday.
- The people who lasted didn't chase the algorithm, they chased the thing that still felt fun
What VidCon 2026's longest-running creators reveal about durability, and why product leaders selling to creators should care about burnout more than reach.
- Config 2026 Put a Number on Your Design System Debt. It Didn't Hand You the Budget.
Figma's Config 2026 tools make design system debt visible to everyone. Here's what changed and what to bring to your next planning review.
- Ford Hired the Gray Beards Back. Here's What That Tells You.
Ford rehired veteran engineers after AI missed its quality bar, and new research shows calling agents coworkers makes your team worse. Here's what to do.
- The Most Defensible Design Move Is Building Less
Adaptive reuse, grown color, and plant-waste materials all point to one sustainable design move: build and consume less. Here is what it means for your team.
- Your Research Data Has Strangers In It
Survey bots, smart incentives, and baselines are the three levers that keep your user research credible when AI noise is everywhere. Here's what to do.
- AI Stopped Being a Feature. It Became Infrastructure.
AI is moving into the plumbing of health, public health, and web data. Here's what that shift means for your product roadmap and what to ask your team.
- Your Next Design Review Is Now a Liability Review
Regulators, lawsuits, and surveillance failures are turning privacy and safety into hard design constraints. Here is what changed and what to do Monday.
- Autonomy Just Grew Up. Now It Has to Make Money.
Robotaxis and humanoid robots are moving from demo to product. Here is what the SPAC, the Zoox redesign, and the brake-pedal rule mean for your team.
- Your Job Just Moved From Making the Thing to Judging the Thing
AI now does the making. The design skill that survives is critique: defining what good looks like and judging output. Here's how to staff and run reviews for it.
- Your Agents Will Leak. The Question Is Whether You Built for It.
AI agent security just moved from research footnote to product requirement. Here's what DeepMind, ServiceNow, and others found, and what to do about it.
- Your AI Ships Fast. It Can't Tell You What's Worth Shipping.
AI output is fluent but taste-free. Here's what that means for trust, differentiation, and how your design team makes calls in 2026.
- Your AI Has a Focus Problem, and Now You Can Prove It
New research gives design and product leaders real language for the AI trust talk: attention limits, correlated errors, and a productivity backfire.