AI
CEO Brian Chesky sees the AI future, even if he's concerned about proprietary models.
Call centre interactions are "ripe for automation".
Leaders need to look out for "model decay and data drift: Your model loses relevance over time" as well as "runaway training costs – GPU bills can kill ROI if not governed..."
Is "services as software" really a $1.5 trillion opportunity or just more "AI overhype"?
Ready to hand your card over to an AI agent?
$72 billion in annual infrastructure spend is not enough, suggests Meta.
Open source issues, excessive model access, and no authentication layers putting AI projects at risk.
New models often worse for hallucinating, OpenAI data shows.