# Simon Allen

Simon Allen writes and speaks about digital infrastructure, including the power, data centers, chips and supply chains needed to scale AI. They represent the Quantum Infrastructure Council and have moderated industry discussions.

- Profile: https://b2binfluencer.marketing/creators/simon-allen/
- LinkedIn: https://www.linkedin.com/in/simonallen123
- LinkedIn followers: 31,482
- Topics: artificial-intelligence, supply-chain, ai-infrastructure
- Based on 11 dated posts, latest 2026-09-17

## Summary

Simon Allen is most useful for explaining AI infrastructure constraints: power, data centers, chips, supply chains and skills.

- Best for: UK or European digital infrastructure operator, investor or advisory firm planning a programme on delivery constraints.
- Not for: A global enterprise-software vendor selling AI governance or automation tools to non-technical executives.
- Check first: Confirm audience data, post-level engagement and whether they offer a paid partnership format.

### Takeaways

- AI depends on a physical chain from electricity and substations through cooling, chips and networks.
- Energy, data center infrastructure, chips and advanced manufacturing account for roughly one in 45 workers.
- They say the AI infrastructure debate stops too early at the data center.

## Main ideas

- **AI depends on a physical chain, not just computing power**: Allen argues that discussion of AI infrastructure often stops at data centers. They point further upstream to chip manufacturing and its needs for electricity, water, chemicals, grid resilience and energy security. In another post, they map the chain from electricity and substations through data centers, cooling, chips and networks. Together, these examples show a habit of widening the frame from a visible facility to the systems that make it possible.
- **Delivery constraints deserve as much attention as AI demand**: Allen highlights the practical limits on infrastructure expansion: power equipment, chips, cooling, supply chains and skilled workers. They focus on how projects can proceed amid scarcity, including through standardisation, alternative sourcing and procurement. This puts execution alongside technology ambition.
- **Technology readiness should be grounded in local infrastructure realities**: Allen presents quantum readiness as a practical planning question, not only a distant technology prospect. They connect it to power, cooling, site design and long-term strategy, and favor examining the domestic ecosystem in its specific context. The UK post also links infrastructure discussion to planning reform, regional growth, investment, supply-chain delivery and skills.

## Work to read

- [AI's physical stack rests on roughly 1 in 45 workers](https://www.linkedin.com/posts/simonallen123_ai-datacenters-digitalinfrastructure-activity-7492519585821204481-FY2m)
- [The AI infrastructure debate stops too early at the data center](https://www.linkedin.com/posts/simonallen123_nvidias-silent-ai-colonialism-is-trapping-activity-7480200750149713920-HMFp)
- [Quantum readiness framed as a practical infrastructure question](https://www.linkedin.com/posts/simonallen123_ukinfrastructure-digitalinfrastructure-energytransition-activity-7432688984700624896-bpkW)
- [Moderating a panel on delivering projects amid shortages](https://www.linkedin.com/posts/simonallen123_capacityeurope-datacenters-digitalinfrastructure-activity-7502749220391645185-vMzi)

## Brand relationships

- **Quantum Infrastructure Council** (mention)
- **Platform Markets Group** (disclosed_partnership)
- **Capacity Europe** (mention)

## Partnerships

Simon Allen's posts show no paid partnership offer. They document industry-event participation: representing the Quantum Infrastructure Council (described as a partner of Platform Markets Group) at Platform UK and moderating a panel at Capacity Europe 2026. These are industry-forum appearances, not a published sponsorship format. No media kit, rate card, sponsored-content policy, disclosure practice, or official inquiry route appears in the posts. Buyers considering a paid placement, sponsored post, or event collaboration should ask Simon Allen directly about commercial terms.

## Alternatives

- Aleksandar Rodic: Pick Aleksandar Rodic when the firm's audience is investors, boards or policy-adjacent clients and the question is where AI compute will physically sit and who controls it. Each post takes a single development and runs it through the same fixed format: why it matters, the board-level implication, who gains and loses power, and what to watch next. The first covers Anthropic's first Australian data-centre lease, a planned 2.16-gigawatt campus used for inference from 2027, alongside Nvidia's planned 2 GW of Australian capacity against roughly 1.6 GW today, and turns it into questions about jurisdiction, grid connections, approvals and who pays for new grid capacity. The second does the same for European energy security, tying gas storage at about 67% full, lost Qatari and UAE volumes and Asian spot prices near $40/mmBtu to rearmament, industrial competitiveness and AI build-out. That repeatable structure suits a firm wanting a recurring written briefing or a session on sovereign compute siting, and the Europe post suits a client base of European industrials or utilities.
- Michael Freedman: Pick Michael Freedman when the audience is technical — infrastructure and platform engineers, controls engineers, systems integrators, data teams — and the firm wants the operational argument for why power, not chips, now sets the compute envelope. The longer post traces the shift from power treated as an optimisation variable, through late-2000s work on energy-proportional computing and power-aware systems, to power as the binding constraint. It supports the argument with a market signal (a proposed $67B NextEra–Dominion deal read as a direct response to AI electricity demand) and a firsthand example: hyperscalers refusing server expansion not for lack of hardware or demand but because of regional power capacity limits. It ends on the operational control loop, dense and liquid-cooled designs, and why power, cooling, water and equipment state have moved from dashboard metrics to system architecture. The second post shows a talk format for controls engineers and integrators on combining industrial control with Postgres that scales to physical-world data volume. Use this for a technical deep-dive or a session about the operational data layer.
- Eric Victor Gitter: Pick Eric Victor Gitter when the firm's target audience is the IT channel or mid-market technology providers — MSPs, systems integrators, resellers — or when the firm wants the spend-composition view of the AI build-out. The first post splits the projected $6.7 trillion data-centre build to 2030 into silicon and compute against power (15–25%), cooling (10–15%) and facilities and real estate (5–10%). It flags utilities facing a $1.3 trillion capital call, names direct-to-chip and immersion cooling and modular design as the bottleneck-solving categories, and argues the opening is for mid-sized industrial and energy suppliers and the channel that serves them. The second covers supplier concentration — the Stargate memory deals securing around 900,000 wafers per month, Micron's consumer-brand exit, over 90% of memory supply focusing on a few AI customers — and turns it into positioning advice for channel businesses, including where hyperscalers will not own the ground, such as sovereign AI and edge deployments. That combination suits a channel-facing session or a market-structure briefing.

---
AI-assisted summary of public posts. Audience demographics are not available. Engagement appears only where we checked recent posts. Method: https://b2binfluencer.marketing/methodology/