{"version":"1.0","type":"rich","provider_name":"Acast","provider_url":"https://acast.com","height":250,"width":700,"html":"<iframe src=\"https://embed.acast.com/$/69ab3b7c7036d739021982df/6a11f9bea9d3d2ec14569072?\" frameBorder=\"0\" width=\"700\" height=\"250\"></iframe>","title":"Why Big Tech Now Runs an AI Factory","description":"<p>What's really happening inside the AI supply chain that powers every model you use?</p><p><br></p><p>The common story is that AI is a software business with a fancy backend. The reality is more complicated, and it changes how you should buy, budget, and contract for AI.</p><p><br></p><p>In this podcast, I share the inside scoop on why your AI vendor contract is now a supply contract in everything but name:</p><p><br></p><p> • Why \"capacity constrained\" points to memory and packaging, not GPUs</p><p> • How hyperscaler CapEx reshapes every vendor agreement you sign</p><p> • What questions belong in your next AI investment review</p><p> • Where a single supply chain delay stops you from shipping AI</p><p><br></p><p>For operators and CFOs, the takeaway is sober: cheaper tokens are real and serving costs keep falling, but the industrial base underneath your AI strategy still demands supply assurance, utilization discipline, and contracts that account for allocation risk.</p><p><br></p><p><br></p><p>Subscribe for daily AI strategy and news.</p><p>For deeper playbooks and analysis: https://natesnewsletter.substack.com/</p>","author_name":"Nate B. Jones"}