As AI systems scale, energy demand has become a central concern.

Two contrasting approaches are often discussed:

→ expanding terrestrial data centre infrastructure

→ pursuing extreme future energy concepts such as a Dyson Swarm

At first glance, these appear to represent different solutions to the same problem.

They do not.

System A: Terrestrial Data Centres

Characteristics:

→ high energy demand

→ significant water usage for cooling

→ dependence on local grid infrastructure

→ direct coupling to land, ecosystems, and communities

Constraints:

→ energy availability

→ water systems

→ environmental impact

→ social and regulatory limits

Failure Mode:

Systems continue to scale despite increasing pressure on the conditions they depend on.

Result:

→ grid strain

→ water depletion

→ ecological degradation

→ eventual constraint enforcement (delays, shutdowns, collapse)

Water Systems — Operational Reality

A common claim is that data centres “only circulate water” and therefore do not meaningfully consume it. This is incomplete.

While many cooling systems reuse water internally, heat removal at scale requires continuous dissipation.

In most large data centres, this occurs through evaporative processes.

As a result:

→ water is continually lost to the atmosphere
→ and must be replaced from external water systems

At scale, this becomes significant.

Water use is not defined by internal circulation,
but by the ongoing requirement for external replenishment.

HABITS Clarification

From an admissibility perspective, the relevant question is not:

→ whether water is reused within the system

It is:

→whether the system can operate without exceeding the capacity of the water systems it depends on

System B: Dyson Swarm (Theoretical)

Characteristics:

→ large-scale solar energy capture beyond Earth

→ reduced dependence on terrestrial energy generation

→ extreme infrastructure and coordination complexity

→ reliance on orbital systems and transmission layers

Constraints:

→ material extraction and manufacturing

→ launch capacity and maintenance systems

→ energy transmission and distribution

→ continued dependence on Earth-based systems

Failure Mode:

Assumption that increased energy supply removes systemic limits.

Result:

→ constraints displaced rather than removed

→ increased system complexity

→ continued exposure to non-energy limits (water, materials, ecosystems)

HABITS Evaluation

These systems differ in implementation. They do not differ in principle.

Both are evaluated against the same condition:

Do they remain within the conditions required for their continued existence?

Key Insight

Neither system inherently satisfies this condition.

Data centres:

→ exceed limits through direct pressure

Dyson Swarm:

→ attempts to bypass limits through expansion

In both cases:

→ admissibility is not resolved

Critical Distinction

Increasing capacity does not ensure viability.

A system that depends on conditions it degrades cannot persist, regardless of scale.

HABITS Principle

Before any system is allowed to scale, it must demonstrate:

→ alignment with energy systems

→ compatibility with water systems

→ stability within material and ecological limits

If these conditions are not met:

→ the system is not optimised

→ it is inadmissible

Conclusion

The question is not:

“How do we generate more energy?”

It is:

“Are we allowing systems to exist that cannot remain within the conditions they depend on?”

Until that is resolved:

→ scaling continues

→ constraints accumulate

→ failure is deferred, not avoided

Admissibility is not about limiting progress. It determines whether progress can persist