EU signs Orange Business for TESTA-EIRIS backbone as energy costs squeeze industry

The deal, announced by Orange Business and carried by Light Reading on 29 September, makes the operator the network partner for TESTA-EIRIS (Enhanced Infrastructure for Reliable Interconnectivity and Security), the EU's replacement for its older TESTA network. The company will supply a Layer 2 and Layer 3 backbone connecting at least 12 points of presence across five European regions, with 99.999% availability and a secure-by-design architecture.

It is infrastructure news with a sovereignty angle. Orange Business says the platform will connect EU institutions, bodies, agencies and national public administrations across member states, working with the European Commission's Directorate-General for Digital Services. Its Evolution Platform, which the company describes as an AI-first connectivity platform, will replace dedicated hardware with software-based virtual services. None of that changes the price of a kilowatt hour. But it does show where Brussels is willing to spend on digital plumbing even while the wider industrial base is under energy strain.

The price shock is the backdrop

Europe's energy problem has not gone away in the last week of September. In a piece published on 29 September, The Conversation noted that US crude at the main American hub went from US$66 a barrel in late February 2026, before the US and Israel attacked Iran, to $101 a barrel on 13 April, after the Strait of Hormuz was effectively closed. Those are American benchmark numbers, but the authors, an energy economist and an international trade economist, make the point that the oil market is global: price moves travel everywhere, and so do the products made from oil.

That matters for manufacturers in a way that is easy to understate. Crude is not just fuel. It is feedstock for plastics, a cost input across almost every supply chain, and a raw material for fertiliser. When supply is interrupted without much warning, the authors write, prices can rise sharply in a short period. The bulk of the extra money heads towards the source of the oil, meaning oil companies, with shareholders taking much of the gain through dividends, buybacks and debt reduction.

European industry is on the paying side of that equation. The dossier's recent headlines, which we are not using as citable facts, point to the same pressure from several directions: diesel prices at record highs in the UK, gas prices rising as QatarEnergy extends force majeure on LNG supplies, and INEOS idling Europe's last world-scale acetyls plants.

Electrification cuts demand, but not overnight

The structural answer to high fossil fuel prices is electrification, and the numbers behind that case are better than the slogans. In a 29 September post, Hannah Ritchie works through a model by Oxford professor Nick Eyre showing that global final energy demand falls from 416 to 247 exajoules in a post-transition system where suitable sectors are electrified and the rest run on hydrogen. Electricity demand rises from 110 to 189 EJ at the same time. The world uses less total energy, because electric systems waste less of it.

The sector detail explains why this is slow. Electric vehicles convert around 80% of energy into motion, against roughly 20% for petrol cars, so post-transition demand for cars and vans drops to about a quarter of today's. Buildings fall sharply as gas boilers give way to heat pumps. Heavy goods vehicles are only half electrified in the model, with hydrogen covering the rest. Buses are 80% electrified; short-haul aviation is electric, medium and long haul is not; marine transport is just 10% electrified. High-temperature industrial processes cannot be electrified at all, which is the bit that keeps European factories exposed to gas and oil prices.

Ritchie is careful about the limits. The model assumes no efficiency gains beyond electrification and hydrogen, and it does not account for energy growth as countries develop. It is a comparison of today's energy use, not a forecast.

Supply and demand, but not only

If the transition is the long answer, this year's jobs data is the short one. On 28 September, Electrek reported that the US lost 36,949 clean energy jobs in 2025, the first annual decline since the pandemic, according to findings from E2 based on US Department of Energy employment data. Clean energy employment fell to 3.52 million, erasing nearly 40% of the gains made in 2024. Losses reached 35 states, with California shedding nearly 21,000 jobs, the most of any state, while Florida gained about 3,800. Energy efficiency, renewables and EVs all lost jobs; battery storage, grid work and biofuels posted small gains.

E2 ties the downturn to the Trump administration and the Republican-controlled Congress rolling back federal support for clean energy and EVs, and to the resulting project cancellations. Its tracker recorded 142 clean energy manufacturing, generation and storage projects cancelled or downsized in 2025. The employment figures do not establish how many jobs were lost to any particular policy change, which is a caveat worth keeping in view. The wider US energy industry shed an estimated 86,000 jobs, with clean energy accounting for roughly 43% of that decline, but it still employs more than 3.5 million people, against 958,000 in oil and gas, 125,000 in coal and 70,000 in nuclear. E2's full Clean Jobs America 2026 report is expected in October.

Where the next electricity bill comes from

Demand for electricity is also being pushed by computing. On 29 September, CBC News reported that Newfoundland and Labrador's energy minister, Lloyd Parrott, told the provincial legislature during a special session on the new Churchill Falls agreement in mid-September that the government could consider selling excess Churchill Falls power to a data centre company. "We said our door is open for business and we're looking forward to doing business with the world," Parrott said at the time.

A department spokesperson, Brodie Thomas, later confirmed by email that the provincial government has been approached by companies about potential data centre development, adding that commercially sensitive discussions mean it will not comment on specific proponents or projects. Premier Tony Wakeham and Barry Perry, the lead negotiator on the Churchill Falls deal, also floated data centres as a possible use for the power. Paris Marx, author of Hyperscale, told CBC that AI products are computationally intensive and that tech companies found they needed far more computation to run them. Meta and Bell have announced billions of dollars for large data centres in Alberta, Ontario and Saskatchewan this year, according to the same report.

That is the shape of the problem in one province: cheap power, available land, and a debate about whether the electricity should go to a data centre or somewhere else. Scale it up and it is the European argument too, with the difference that Europe is importing the fossil fuel it is trying to stop burning.

The suppliers are repricing efficiency

Chipmakers are making the same bet from the other side. On 29 September, Efficient Computer announced a $97 million Series B round led by TQ Ventures, with CEO and co-founder Brandon Lucia writing that the company builds energy-efficient processors. His argument is that CMOS process scaling has slowed and that most performance gains now come at a cost in efficiency, through large memory structures, interconnects and front-ends that try to extract parallelism from sequential code.

Lucia also takes aim at fixed-function AI accelerators, calling them a devil's bargain: efficiency and speed for today's AI in exchange for programmability, with the risk that the silicon goes cold when algorithms change. He invokes Amdahl's Law to make the point that the unaccelerated part of a workload sets the ceiling on system efficiency. It is a vendor's pitch, and should be read as one, but the underlying constraint, that energy is the limit on computing, is the same one now shaping where data centres get built.

None of this week's news reverses the price shock. The Orange contract is a multi-year infrastructure decision. The E2 numbers are a 2025 scorecard. The Churchill Falls discussion is a proposal. What connects them is the same calculation European manufacturers have been running since the spring: what energy costs, where it comes from, and who gets to use it.