I want to quantify how recent advances in artificial intelligence are influencing forecasts of French GDP. I already have access to long-run historical GDP time-series and would like you to work only with that dataset for now. The task is strictly quantitative: build and compare solid économétriques models that incorporate an IA variable (or proxy) and measure its marginal effect on growth projections. You will be free to choose among standard time-series techniques—VAR, ARDL, error-correction, whatever proves statistically sound—provided the approach is fully documented and reproducible in R, Python, or Stata. Along the way, please explain every transformation, stationarity check, and specification test so that the methodology can be defended in an academic setting. Deliverables • Cleaned and commented dataset (CSV) • Script or notebook with all code and step-by-step comments • Forecast outputs plus a concise technical note (≈5 pages) summarising model selection, diagnostics, and the estimated AI impact on future GDP growth. I will review your code for reproducibility and the note for clarity and rigour before signing off.