AI's new North Star: Intelligence per dollar (www.businessinsider.com)

The AI industry is shifting its focus from simply having the most intelligent model to maximizing the efficiency of AI spending, introducing a new metric: intelligence per dollar. With many models reaching a baseline of effectiveness for various business applications, companies like Amazon are prioritizing cost-effectiveness in their AI deployments. For instance, internal documents reveal that Amazon is directing requests through a less potent version of its Alexa AI to avoid unnecessary expenses associated with more advanced models, reserving their use for tasks that genuinely require higher performance. This trend toward efficiency means that companies are now assessing AI capabilities not only on their raw performance but also through a lens of operational cost. Factors like reliability in completing tasks, pricing models, potential for information reuse, and the overall work required are critical to consider. As Kylan Gibbs of Inworld describes, the aim is to optimize AI models for affordability and speed, not just intelligence. Consequently, while high-performing models may dominate headlines, it is those that offer practical, cost-effective solutions that are poised to thrive in the competitive AI market.