I trust AI code that crashes more than AI code that works, and 66% of developers seem to agree
79% of engineering leaders noticed the same thing.
A recent survey of 300 senior engineering leaders reveals that despite AI coding tools speeding up code production, release cycles remain unchanged. Debugging takes up 42% of developers' time, with 79% of leaders believing AI-generated code is almost right but not perfect. This leads to 35% of AI-generated code reaching production without full understanding, causing 81% of incidents or outages in the past six months.
Trust in AI output is low, with only 6.6% of developers trusting it for important decisions. To mitigate issues, developers suggest having AI explain bugs before fixing them and writing failing tests first. The survey, conducted by Undo, a debugging tool company, primarily involved developers working on mission-critical systems in finance, networking, and chip design.
Written by urgent.news from XDA Developers's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
