On the second day of COP31, the Climate Action Tracker released a chilling assessment: more than half of the climate commitments announced by the world's largest companies are deemed largely insufficient to align their emissions trajectories with the 1.5 °C target. This finding, echoed by several independent observers, underscores that public pledges do not always translate into concrete action on the ground, information unconfirmed at this stage. Faced with a climate emergency that becomes clearer each year, the gaps between reduction targets and actual results threaten to undermine the collective efforts needed to limit warming. Analysts are therefore calling for a strategic turnaround, at both government and private sector levels.
Details of the announcement: who, what, precise figures
The report presented at the summit targeted over a hundred multinationals operating in the energy, fashion, agri-food, and technology sectors. Among them, several oil giants have declared carbon neutrality targets for 2050, while fashion leaders pledged to reduce their scope 1 and 2 emissions by 30 % by 2030. No exact figures were confirmed in the press release, but analysts note a « major gap between commitments and actual trajectories », information unconfirmed at this stage.
Evaluators used predictive models based on neural networks and satellite data to compare public declarations with actual emissions records. By cross-referencing data from Copernicus, ECMWF, and greenhouse gas tracking databases, they identified discrepancies of several tens of percent, particularly in sectors where scope 3 emissions are the most difficult to quantify. This approach, which incorporates machine learning to reduce forecast uncertainty, shows that commitments are often underestimated.
To close this gap, the report recommends three major levers: first, impose reporting requirements based on verifiable and audited standards, notably by leveraging Copernicus open data platforms. Second, integrate performance clauses into energy purchase contracts, so that targets are linked to measurable indicators. Third, encourage the use of AI tools like GraphCast or Pangu-Weather to model emissions reduction scenarios and anticipate non-compliance risks, thereby reducing the forecast uncertainty associated with commitments.
