Adrian Allen
Part two of two
In part one of this two-part mini-series of articles (ESG has not gone away), I set out why AI enthusiasm and Net Zero delivery are increasingly competing for the same finite resources - capital, grid capacity, skilled labour and boardroom attention - rather than complementing each other, and why a June 2026 change to the MEES regulations means most owners are now working from an outdated timeline. This second part looks at where AI genuinely helps, where it's being oversold, and what a sensible sequencing of the two looks like.
Closing the expertise gap
None of this is straightforward to execute in-house, and that is worth saying plainly rather than glossing over. A defensible EPC B pathway, a genuinely useful ESG strategy, and a procurement process that lands at the price the feasibility study assumed are three quite different skill sets, and most owners and asset managers do not carry all three internally, nor should they need to.
This is the gap that advisers, such as Remit Consulting, are built to close. We bridge the gap between ESG experts and the real estate market, pairing genuine ESG expertise with the practical realities of running property, from strategy design and policy review at the front end, through to reporting, ongoing monitoring, and the procurement and tender process. That translation work is precisely what most owners are currently missing, and precisely where the current AI enthusiasm is least likely to substitute for genuine advisory expertise.
Where AI genuinely helps, and where it's being oversold
None of this is an argument against using AI in the built environment. Some of the sharpest emerging use cases in commercial real estate ESG delivery are directly relevant to Net Zero delivery, not adjacent to it.
AI-enhanced building management systems are already demonstrably reducing energy costs and carbon footprints by optimising HVAC and lighting against real-time and forecast weather data: a genuinely mature, low-risk application with clear payback.
ESG reporting automation, particularly for GRESB-style disclosure, is one of the least glamorous but highest-value applications, freeing asset managers from data-wrangling to focus on the retrofit decisions that actually move the needle.
Predictive maintenance shifts plant management from reactive expense to proactive strategy, extending asset life and reducing the embodied-carbon cost of premature replacement.
Peer-reviewed research on data centres themselves shows that usage-based building controls can cut energy consumption by 20 to 30% simply by aligning cooling output to actual demand rather than static schedules, a lesson equally applicable to commercial office plant.
The distinction worth drawing is this: AI is a powerful instrument for executing a Net Zero strategy more efficiently. It is not itself a strategy. A BMS that optimises a building's existing plant will not get a 1985-vintage secondary office from EPC E to EPC B; only capital investment in fabric, glazing, heating electrification and genuinely improved payback economics on-site generation will do that. Since 2016, solar installation costs have fallen roughly 75%, and panel costs around 90%, with efficiency up 67% and payback periods now frequently under four years at IRRs of 10 to 18%. That value-add case stands on its own economic merits, with or without an AI narrative attached.
The uncomfortable question for asset managers
There's a reason AI dominates the current conversation, and Net Zero retrofit doesn't: AI adoption is visible, fast and cheap to talk about- a pilot project, a co-pilot licence, a case study - while fabric-first retrofit is capital-intensive, slow (12 to 18 months from assessment to completion is standard for commercial retrofit), and politically unglamorous. It is far easier for a business plan to claim to be "AI-enabled" than to commit the capital a genuine EPC B pathway requires. AI adoption could become a form of ESG theatre: a way for owners to appear forward-looking while deferring the harder, more expensive decisions that actually reduce a building's carbon footprint and regulatory exposure.
The counterargument deserves airtime too. Capital is finite, and if AI genuinely delivers efficiency gains that free up management time and reduce operating costs, that saved capacity can, in principle, be redirected towards retrofit planning rather than away from it. Whether that redirection happens in a given portfolio is an execution question, not a technology question. It depends entirely on whether the asset manager treats AI tooling to fund and prioritise fabric investment, or as a substitute headline for it.
The practical takeaway
For owners, asset managers and occupiers navigating 2026, the sensible position is not "AI or Net Zero" but a clear-eyed sequencing.
Commission, or refresh, an EPC B feasibility study now, regardless of the extended 2031 deadline. Supply chains for heat pumps, solar PV and skilled retrofit contractors are already constrained, and a late scramble in 2029 to 2030 will be more expensive and less certain than early planning.
Treat the sub-1,000 sqm carve-out with caution, not comfort. It covers roughly 85% of UK commercial property, but the smaller capital outlay typically required to upgrade smaller assets makes voluntary early action more affordable, not less urgent. The market is already pricing sustainability ahead of the regulatory floor regardless of size.
Deploy AI where the payback is genuinely proven: BMS optimisation, ESG reporting, predictive maintenance, and be honest that these are efficiency tools, not substitutes for capital-intensive fabric works.
Revisit solar economics on every asset with roof or car park capacity. The payback case has fundamentally shifted since 2016 and stands independently of any AI narrative.
Watch the regulatory detail, not just the headline. The gap between "EPC B by 2030," the assumption most of the market is still working to, and "EPC B by 2031, buildings over 1,000 sqm only," the actual June 2026 position, is exactly the kind of distinction that separates a defensible business plan from a stale one.
Bring in specialist bridging expertise where it's missing internally. Advisers such as Remit Consulting, who work across strategy design, policy review, reporting, monitoring and procurement, exist precisely to connect ESG intent with commercial delivery, reducing the risk of a feasibility study or strategy document that never translates into contracted works.
The built environment's Net Zero challenge was never going to be solved by a single technology trend, however compelling. It will be solved by owners who keep doing the unglamorous work - fabric upgrades, on-site generation, EPC pathway planning - while using AI as a tool to do that work more efficiently, rather than as a reason to stop talking about it