DORA 2025

DORA 2025 measures a positive correlation between AI adoption and delivery throughput, and a negative one with delivery stability when the foundations do not follow. Alongside that measurement sits a claim: AI acts as an "amplifier" of an organization's existing strengths and dysfunctions. The pair travels together, and both reappear as premises in arguments about how the software lifecycle should be redrawn rather than as findings to be discussed on their own.

Paula Hingel builds "How AI Changes the SDLC: A Six-Stage Guide" (8 June 2026) on the report: process maturity counts for more than the tool, and the throughput/stability split is the symptom of an adoption carried out without redrawing review boundaries. Her six stages are an attempt to give the negative half an address. Deployment gets explicit rollback policy, testing gets circular validation named as the core risk.

«SFEIR» places DORA 2025 next to the ADLC and Google's new-SDLC whitepaper as evidence of industrial convergence in "Un SDLC piloté par l'IA : le cycle SFEIR à 11 phases" (16 June 2026), paired with its own diagnosis of 2025: organizations deployed AI tools without transforming their operating model, "tout change… et rien ne change". The amplifier framing carries a corollary none of the notes settles. If AI amplifies what an organization already has, then the measured loss of stability is a reading of those organizations themselves, and the report offers no instrument for telling a weak foundation from a fast one.