The new resource categorizes over 280 companies that specialize in building software primarily through AI-assisted methods. As of June 2026, researchers have begun evaluating these firms based on their ability to deliver secure, maintainable code, moving beyond the industry hype surrounding AI velocity. This shift comes as projections from Gartner suggest that 60% of all new code will be AI-generated by the end of the year.

Evaluation criteria prioritize engineering discipline over self-reported capabilities. Goodfirms analysts conduct direct interviews with past clients to confirm project outcomes and verify technical rigor. This vetting process is critical, as recent data from the Veracode 2025 GenAI Code Security Report highlights that AI-generated code frequently introduces significant security flaws, including those in the OWASP Top 10. By highlighting agencies that harden and test AI-produced code, the platform seeks to mitigate the risks inherent in automated development.