EY built an AI router to slash costs by dynamically routing tasks to the cheapest viable model. The system targets runaway AI spend, a growing pain point as firms like EY burn cash on overpowered LLMs for mundane tasks. The move follows PwC’s similar cost cutting tools, underscoring the Big Four’s shift from AI hype to fiscal discipline. Europe’s high cloud costs and sparse local AI infrastructure compound the pressure.

This is Europe admitting it cannot afford its own AI ambitions. The continent’s reliance on US cloud giants (AWS, Azure, Google Cloud) means every query carries a premium, with margins leaked across the Atlantic. EY’s router is a stopgap but reveals a deeper problem: Europe’s lack of scale in AI infrastructure forces firms to optimize for cost, not performance. The second order effect is a cap on innovation. If even deep pocketed consultancies baulk at AI spend, startups and SMEs will retreat to safer, cheaper tools, ceding ground to US and Chinese rivals.

Winners: US cloud providers (AWS, Azure, Google Cloud) retain pricing power. Losers: European firms (SAP, Siemens) face higher AI costs, stifling adoption. EY wins short term savings but loses long term leverage. Mistral AI and other European LLM players gain if firms pivot to local, cheaper models.

Watch for EU policy intervention to subsidize local AI infrastructure or mandate cost transparency from cloud providers. Expect more Big Four firms to follow EY’s lead, accelerating a race to the bottom on model pricing. Monitor whether European startups pivot to open source or on premise solutions to avoid cloud lock in.