The AI race is shifting from bigger models to cheaper, smarter systems. AI companies are moving beyond a simple race over who has the biggest or newest model to systems that choose which model to use for each task.
Perplexity CEO Aravind Srinivas states that the real product is becoming the system that selects which model to use.
Benchmark’s Peter Fenton says open-weight models could soon handle most AI usage, putting pressure on the economics of the biggest model providers.
For the past two years, the artificial intelligence race has been easy to score: bigger models, better benchmarks, and whichever company could claim the lead.
As companies move from testing AI to using it in real products and workflows, it's no longer about tapping the best model, but accessing the one that's the best fit for a specific job.
Srinivas notes that the model alone is no longer the product, but the system that decides which model to use, when to use it, and what outside tools or company data sources are necessary.
Additionally, open-weight models, which can be downloaded, tuned, and run by companies themselves, are becoming more capable.
Fenton believes that over 90% of the tokens created will come from open-weight models over the next 18 to 24 months.
Srinivas argues that the U.S. should support open models because they make AI more affordable.
