You have 1 article left to read this month before you need to register a free LeadDev.com account.
Estimated reading time: 2 minutes
Key takeaways:
- The share of developers saving 10 hours a week with AI fell from 68% to 25%, as code review eats the gains.
- Finding information and meetings still cause the most developer friction.
- A clear AI strategy makes reduced friction twice as likely for developers.
Ask a developer whether AI makes them faster and you’ll get a quick yes. Ask whether their team ships faster and the answer gets a lot more complicated. Atlassian’s 2026 State of Developer Experience report puts numbers on that gap.
The survey of 1,155 developers and 405 engineering leaders found that 87% of developers say AI tools help them write and ship code faster. Yet 72% report the same or more friction across the software development lifecycle (SDLC) than before AI arrived.
Your inbox, upgraded.
Receive weekly engineering insights to level up your leadership approach.
AI-generated code is piling up in code review
The number that jumps out is the drop in time saved. In 2025, 68% of developers said AI saved them at least 10 hours a week. This year, only 25% say the same.
That’s a steep fall in 12 months, and the report points to a familiar culprit: 65% of developers now regularly spend time reviewing, debugging, and correcting AI-generated code.
None of this will surprise anyone who has opened a 900-line pull request with the description “refactor.” For most teams, writing code was never the slowest part of delivery. Speed it up and the queue moves to the next stage, where merge conflicts, flaky integration tests, and a reviewer already juggling four other PRs are waiting.
The old friction hasn’t gone anywhere
The top sources of friction look almost identical to last year. Finding information leads at 51%, followed by meetings (50%), context switching (45%), tech debt (44%), and unclear requirements (40%).
AI-coding tools barely touch most of that list. A model can write a working endpoint in seconds, but in most setups it has no idea which of three conflicting wiki pages describes the current auth flow, or why product changed the spec on Tuesday afternoon.
The report notes that most AI deployments still focus narrowly on code generation rather than on how work flows across a team.
“Faster code is not the same as better software delivered sooner,” says Andrew Boyagi, customer CTO at Atlassian.
More like this
Usage numbers make a poor scorecard
The report’s sharpest finding is about leadership: only 17% of developers and 23% of engineering leaders say their organization has a clearly communicated and implemented AI strategy.
Meanwhile, 60% of developers say their organization treats AI usage as a productivity signal, while only 42% of leaders believe they have the right metric to separate real gains from activity.
If your dashboard counts prompts and accepted suggestions, it’s measuring how busy the tools are. It tells you very little about whether customers get working software any sooner.
New York • September 8 & 9, 2027
Loved LDX3 New York? Pre-sale tickets for 2027 are now available.
The good news is in the same data. Developers at organizations with a clear AI strategy are twice as likely to report less friction than before AI.
