Amazon's $1 Billion Bet Reveals the Real Reason Enterprise AI Keeps Failing

Amazon Web Services committed a billion dollars to build a Forward Deployed Engineering organization, embedding thousands of engineers directly inside customer companies to build and deploy AI systems in 45 day cycles. Microsoft followed with a 2.5 billion dollar commitment and 6,000 engineers of its own. A new PYMNTS Intelligence report found that 71 percent of executives at billion dollar companies say organizational readiness, not the technology, is the primary barrier to AI performance. A separate Workplace Intelligence survey of 2,400 executives found that while 97 percent report some benefit from AI, only 29 percent see significant organizational ROI. Soren Kaplan breaks down what's actually happening inside these engagements. The job title says engineer, but the daily work is change management: building trust with the people whose behavior has to change, working inside a company's real approval processes and power structures, and proving results fast enough to satisfy a board. Amazon's team used this approach to cut Lyft's driver support resolution time by 87 percent. The lesson holds whether or not a company has a billion dollars to spend on outside engineers. Soren lays out three moves any leader can make now. Map the three or four people whose daily behavior needs to change before deploying any AI tool. Assign a named internal champion who owns adoption and removes barriers in the first 90 days. Track whether target workflows actually change within 30 days of deployment, instead of counting seat activations. Subscribe and listen on: YouTube: https://www.youtube.com/playlist?list=PLU3e5k8HJeOGEa7AGA97kvVjtpi6uqG0t Spotify: https://open.spotify.com/show/033iseLQje3RnPGqUlh2I8?si=cb19276a3c68459d Apple Podcasts: https://podcasts.apple.com/us/podcast/leapfrogging-the-headlines-with-soren-kaplan/id1896880297