Open Systems Win

In the AI era, many traditional moats may become weaker.

---

Knowledge becomes searchable.
Code becomes generatable.
Content becomes reproducible.

---

Which means systems built mainly around restriction or rigid control may slowly struggle.

---

I think the stronger approach is different:

Build systems that can participate in the new ecosystem.

---

Don’t spend too much energy trying to separate:

human users
from AI agents.

---

Because increasingly, both will exist together.

---

(Disclaimer: I saw this from social media discussions, so details may evolve.)

Xero partnering with Anthropic and enabling MCP-style access into its systems is an interesting example of this shift.

Instead of treating AI agents as something to block, the system starts allowing trusted agents to interact with accounting workflows through structured interfaces.

---

That changes the question completely.

Not:

“How do we stop AI from interacting with our systems?”

But:

“How do we allow it safely and usefully?”

---

The future may belong less to closed systems defending boundaries,

and more to systems that become part of a larger ecosystem.