Five AI funding rounds closed or were announced on March 17, 2026, reaching a reported combined total of over $1.4 billion. The deals span five distinct market positions, though four of them share a common thesis: enterprises need more than models, and investors are now paying for the layer underneath.
Reuters and TechCrunch confirmed Quince raised $500M in a Series E round led by ICONIQ at a post-money valuation of $10.1 billion. Quince is an AI-powered direct-to-consumer apparel brand, the one deal in this cycle that isn’t an enterprise infrastructure play.
The other four rounds tell a different story. Nexthop AI raised a reported $500M Series B led by Andreessen Horowitz, per the company, at a reported $4.2 billion valuation, targeting AI networking infrastructure. Lightspeed Venture Partners is also reported as a participant, though lead investor attribution has not been independently confirmed from a T2 source. Kai emerged from stealth with a reported $125M Series A led by Evolution Equity Partners, according to reports, for an agentic AI cybersecurity platform designed to bridge IT and OT security, a sign that autonomous AI agents are already creating audit surface area that traditional security tools weren’t built to handle. Oro Labs raised a reported $100M in a reported Series C co-led by Brighton Park Capital and Goldman Sachs Alternatives for agentic procurement orchestration. Axiom raised a reported $200M in a reported Series A at a reported $1.6 billion valuation, according to multiple technology outlets, for formal verification of AI-generated code safety.
The pattern is hard to miss. Of the five rounds, four are directly solving problems that arise when enterprises try to move AI from pilot to production: network performance, agent security, procurement workflows, and code trustworthiness. Quince is the outlier. The rest are infrastructure.