The agents are inside the building — now someone has to watch them

You can date a technology's arrival by what the money starts funding. Capital has rotated from building agents to supervising them, which tells you agents are already in production doing things that matter.

Zenity raised $125 million — led by Norwest, with SoftBank, Hitachi and LG joining — to secure and govern AI agents operating inside corporate systems. Nobody writes that cheque for a hypothetical risk. They write it because the agents are already there.

Read the cap table, not the pitch

The strategic investors are the tell. Hitachi and LG are industrial operators with enormous internal software estates. Their money says the customer for agent supervision is the same enterprise that deployed the agents six months ago and has since realised it cannot fully describe what they are permitted to do.

Meanwhile the access is widening

And the surface keeps growing. HP is scaling agents across customer experience and operations, while MCP servers start exposing governed enterprise master data to any agent authorised to ask. Master data is the spine of a company's factual world. Putting a standard socket on it is enormously useful and exactly the kind of thing that needs a bouncer.

The sequence is better than last time

Payment rails are arriving on the same schedule — record rounds for agent-led purchasing infrastructure and enterprise agent security. Compare this with the last platform shift, where security arrived years after adoption and spent a decade catching up. The agent economy is growing its immune system roughly in step with its muscles.

That is not caution. That is what it looks like when an industry has learned something.

Fortune — SoftBank, Hitachi, LG back Zenity's $125 million round → · Crescendo AI — Latest AI startup funding news and VC deals →