How to act concretely on industry's carbon footprint? At Tech For Industry Show 2026 (Paris Expo Porte de Versailles) — on a day when it was 41 °C in Paris — an ENGIE expert offered a broad, no-nonsense overview of the stakes, the available levers and their maturity, aimed at manufacturers.

The starting point is stark: CO₂ works like a budget. Once emitted, it stays in the atmosphere for centuries (about 50% remains after 1,000 years). So we know precisely how much can still be emitted for each warming pathway. To meet the Paris Agreement's 1.5 °C target, only about four years of emissions at the current rate would remain — a target already nearly out of reach. Staying "well below 2 °C," however, corresponds to about 25 years of emissions at the current rate: feasible by cutting human emissions threefold by the end of the century. Otherwise, current trends point to +2.7 to +3 °C, with consequences very hard to manage even with good adaptation.

For the speaker, "the context condemns us to optimism": the energy transition is progressing. Solar PV is exploding, wind is growing fast, storage and electric vehicles are taking off — sometimes from climate ambition (Europe), often from economic pragmatism (renewables are frequently the cheapest, most resilient option). China, the world's largest EV market, is said to have reached its emissions peak and to be plateauing, while Europe declines and India still rises. But reaching net zero will also require negative-carbon technologies (removing CO₂ from the atmosphere) from the 2030s — without serving as an alibi: 90% of the effort must go to reducing emissions.

The speaker structures action into several levers. First, working on demand: energy efficiency (improving yield) and sufficiency (reducing use) — two complementary approaches. Then electrifying uses wherever possible, since it's easier to substitute an electric source with a low-emission one (renewables, or nuclear, slower to develop). This requires managing grid flexibility to avoid blackouts as renewables rise. Finally, for hard-to-electrify uses (planes, ships, some processes), decarbonized molecules will be needed: biomethane (biomass) or green hydrogen and derivatives (electricity). All this requires heavy investment in infrastructure, notably electricity grids.

In European industry (~4,000 TWh of total energy consumption), half is heat (~1,860 TWh) — and it's electrified at under 10%. Today, heat is mostly produced from molecules, most often fossil (sometimes biomass). The potential to electrify this heat is estimated by the Fraunhofer institute at 1,300-1,600 TWh — considerable.

Three barriers. First, technical complexity: processes, products, constraints and temperatures differ. Below 200 °C, heat pumps work fairly easily; above (high temperatures), it's less certain despite promises. Then an economic problem: fossils can stay competitive against electricity, and above all manufacturers lack visibility (carbon price, energy prices, evolving schemes) to invest with measured risk. Finally, the rise of AI-linked data centers: of the 30 GW of connection rights granted in France in 2025, half reportedly concern data centers — formidable competitors for manufacturers' grid access.

Heat pumps are rising in temperature ever more reliably, and a promising path is hybridization: rather than replacing 100% of a boiler with a heat pump, you install one on 40-50% of capacity (cheaper). The benefit: an extra value lever, arbitrage. In tight electricity systems, curtailing consumption at specific moments, in a controlled way, is remunerated via ancillary services — additional revenue that can tip a business plan, with a payback of around 3-4 years. "Every day counts": a decision cutting a heat activity's footprint by 30-40%, economically "no regret," should be taken without hesitation.

Some heavy industries (steel, chemicals, fertilizers) could switch to electricity via processes that exist technically but aren't yet industrialized enough to be economical. Example: ammonia (fertilizers), whose hydrogen is the essential component — switching to green hydrogen would radically change the footprint (today's "grey" hydrogen weighs ~930 million tonnes of CO₂). But at scale, this won't happen in the next ten years in Europe, despite political encouragement.

For some industries, changing energy doesn't solve everything. In cement, two-thirds of the footprint comes from calcination (a chemical reaction), whatever the energy. The solution of choice then becomes carbon capture — all the more effective as CO₂ concentration is high (cost is inversely proportional to concentration). The CO₂ must then be transported and stored in geologically stable conditions for 1,000 years — possible, but costly.

The chemical industry (~3.9 gigatonnes of CO₂ in 2022) has a third of emissions linked to energy and two-thirds in the products (carbon is the base molecule of plastics). The stake isn't only to decarbonize, but to defossilize the carbon in products, via three renewable-carbon sources: recycling (limited — ~14% effective plastics recyclability in Europe, since recycled products often cost less than collection/sorting); biomass (a real but limited resource, hence arbitrage between boilers, chemical feedstock, biomethane…); and CO₂ capture from the natural environment (DAC from air, at ~€500-600/tonne today, or capture from water) — not yet economical before ten years. A positive example is cited (BASF), and ENGIE already supplies biomethane to chemists, but in modest volumes against the overall footprint.

Industry isn't easy to decarbonize, but levers, technologies and means allow action without waiting. The great value is to decide fast rather than wait for a perfect solution or a phenomenal price drop. The complexity is as much economic (isolating all variables to decide reasonably amid uncertainty) as technical. ENGIE pushes pragmatic, imperfect but quickly deployable, limited-risk solutions — letting other options mature. In 5 to 10 years, new economical solutions will emerge, and those who have already made part of the journey will be best placed to seize them.

What is a carbon budget? It's the total amount of CO₂ that can still be emitted to stay under a given warming threshold. As CO₂ persists for centuries in the atmosphere, this budget is consumed and shrinks fast: for 1.5 °C, only the equivalent of a few years of emissions at the current rate would remain.

Why is electrifying industrial heat difficult? For three reasons: technical complexity (above 200 °C, heat pumps are less mature), economics (fossils sometimes stay competitive and manufacturers lack visibility to invest), and data centers competing for grid access.

What is energy hybridization? Rather than replacing 100% of a boiler, you install a heat pump on part of the capacity (40-50%). This cuts cost and opens an arbitrage lever: curtailing electricity consumption at certain moments is remunerated via ancillary services.

How to defossilize the chemical industry? By replacing the fossil carbon in products with renewable carbon from three sources: recycling (limited), biomass (a restricted resource requiring trade-offs) and CO₂ capture (DAC, not yet economical before ten years).

This conference was recorded at Tech For Industry Show 2026, the B2B trade show for industrial transformation (Smart Manufacturing, AI, Supply Chain) at Paris Expo Porte de Versailles. Watch the full keynote video above, and join us on May 19–20, 2027, Hall 7, for the next edition.