人工智能市场力量.pdf

1、 2022 Agrawal,Gans,GoldfarbMarket Power in Artificial IntelligenceCentre for Competition EconomicsOctober 2024Joshua Gans,University of TorontoTechnologies usually improve competition-transportation-energy-information technology-the InternetTechnologies usually improve competition-transportation-ene

2、rgy-information technology-the InternetBut THIS TIME ITS DIFFERENTOutlineOutlineCalibration:What is AI?OutlineCalibration:What is AI?Algorithmic CollusionOutlineCalibration:What is AI?Algorithmic CollusionMarket Power in Training DataOutlineCalibration:What is AI?Algorithmic CollusionMarket Power in

3、 Training DataMarket Power in Input DataOutlineCalibration:What is AI?Algorithmic CollusionMarket Power in Training DataMarket Power in Input DataMarket Power in PredictionOutlineCalibration:What is AI?Algorithmic CollusionMarket Power in Training DataMarket Power in Input DataMarket Power in Predic

4、tionDiverse Business ModelsOutlineCalibration:What is AI?Algorithmic CollusionMarket Power in Training DataMarket Power in Input DataMarket Power in PredictionDiverse Business ModelsDiverse GovernanceWhat is AI?8Source:https:/www.ais.uni-bonn.de/deep_learning/Artificial Intelligence$19502018Artifici

5、al Intelligence$19502018Artificial Intelligence$19502018Cost of PredictionPREDICTION:Using information that you do have to generate information that you dont have Market Power in Training DataFirms incentives to sell training data are constrained by fears of creative destructionFirms incentives to s

6、ell training data are constrained by fears of creative destructionMarket Power in Input DataLiterature on Sharing Information Under OligopolyLiterature on Sharing Information Under OligopolySharing input data does not change market structure but the intensity of competition within a marketLiterature