You’re definitely not alone that “Step 5 bottleneck” is where most AI-assisted teams hit reality. Right now, most teams aren’t fully automating reviews yet. The common pattern I’m seeing is a hybrid approach, not purely human or purely automated. What others are doing AI generates code → Automated checks (linting, tests, security, architecture rules) → Targeted human review (not full manual review) 👉 The key shift: humans review intent + architecture, not every line.

I think most companies are still in the “AI-flavored features” phase, not truly AI-native yet. Adding a chatbot or a quick automation is fast and looks good in demos, but building AI into the core workflow is much harder. It requires redesigning processes, handling reliability, and actually trusting AI to do meaningful work not just assist. That said, the shift is definitely happening. The companies that are winning are the ones focusing less on “adding AI” and more on solving real problems like saving time and reducing manual effort. Right now it feels like: 70–80% = AI as an add-on 20–30% = moving toward AI-native But over the next few years, that balance will flip. The real winners will be those who treat AI as infrastructure, not a feature.