The Idea Factory
Ideas used to be cheap. Execution was not. The gap between having an idea, building a product, and turning that product into a business was wide. The value sat in the ability to close that gap.
AI compresses it.
What once required months of engineering effort can now be prototyped in days, sometimes hours. The constraint is no longer building the product. It is deciding what should exist and how it becomes a business.
This raises new questions:
- Will founders continue to ideate and build?
- Will venture firms begin doing more of the same?
- Or will both operate in parallel as producers of new companies?
We expect the latter.
Capital Allocation
AI will reshape how capital is deployed.
If agents can write and ship products quickly, the capital required to start a company declines. Large engineering teams are less necessary in the early stages. The bottleneck shifts away from development and toward judgment, distribution, and strategy.
At the same time, certain forms of capital become more important. Compute infrastructure and hardware still require significant investment. GPUs, data centers, and the systems that power AI remain capital intensive.
Capital allocation begins to fall into three buckets:
- Financial capital — infrastructure and compute
- Human capital — ideation, productization, and company building
- Compute — the systems that turn ideas into working products
For early founders the mix shifts toward ideation and compute, with less reliance on large upfront funding.
Future Founders
The question is no longer how to build a product. It is how to build a company around it.
The priorities shift:
- Turning a product into a durable business
- Strategy and distribution
- Sales and partnerships
- Capital expenditure where infrastructure is required
In many cases the product will arrive first and the company will follow.
Talent Networks
Company structure will also evolve.
Lean teams become the norm. Operational overhead declines. Engineering productivity increases. What rises in importance are skills around distribution, partnerships, and sales.
Technical leverage increases. Organizational size decreases.
Ephemerality
Not every product will need to become a long-term company.
Some may exist briefly, capture attention or revenue quickly, and disappear. That would have been impossible under the traditional venture model where companies required millions of dollars to get off the ground.
Now a product that lasts six months can still be economically meaningful.
Anagram
These shifts point toward a model that looks increasingly like what Anagram was built around.
Traditional financial capital will matter less on its own. Building alongside founders matters more. The line between investor and operator continues to blur.
Anagram was designed around the builder–investor model. As the landscape evolves, we expect to lean further into that approach. That means more customized support for founders, deeper involvement in company formation, and new structures that combine compute, capital, and product development.
The tools have changed. The model for creating companies is changing with them.