Developers Report AI Coding Tools Addictive Yet Untrusted, Costs Threaten Budgets

Developers report “I can’t stop” when using AI coding assistants, with an 80% poll response saying the tools feel more addictive than helpful, indicating widespread daily reliance despite mixed utility.

Gartner predicts that the rapid adoption of AI coding services could drain a project’s budget in days, estimating that total spend may soon equal a developer’s monthly salary and ultimately match full developer pay.

AI Coding Tools are now used by almost every developer on a daily basis, yet a separate survey finds that fewer than 10% fully trust the generated code, prompting concerns over hidden bugs and security gaps.

Companies integrating these assistants must now allocate extra funds for license fees and schedule dedicated code‑review cycles, because the combined cost pressure and low trust levels are expected to extend development timelines and strain startup cash flows.