Formula 1® drivers and teams have voiced concerns about the unpredictable nature of the new self-learning power units introduced for the 2026 season. These advanced engines use adaptive algorithms to optimize energy deployment based on driver inputs and track conditions, but their behavior has sometimes caught drivers off guard, affecting both qualifying and race performance.
During the 2026 Belgian Grand Prix weekend at Spa-Francorchamps, several drivers including Oscar Piastri and George Russell reported unexpected straight-line speed deficits that they could not easily explain. The complexity of the power units’ self-learning systems means that small changes in driver input, environmental factors such as wind and grip, and other external parameters can influence power delivery in ways that are difficult to predict or control. This has led to a sense of unpredictability on track, particularly at circuits sensitive to energy management.
The power units, developed by manufacturers supplying teams like McLaren, Mercedes, Red Bull Racing, and Ferrari, incorporate algorithms that adapt power output lap-by-lap and even within a lap. An example of this challenge was seen when Red Bull’s junior driver Isack Hadjar attempted to tow Max Verstappen during qualifying, an action that confused the engine’s software and highlighted the sensitivity of the system to unexpected inputs. McLaren’s team principal Andrea Stella described the factors influencing power unit behavior as “random,” pointing to the difficulty in accurately modelling such sensitive parameters.
Drivers have expressed that these unpredictable power unit characteristics can influence qualifying grids and race pace, sometimes overshadowing driver skill. Teams are actively working to improve their simulation tools to better understand and manage the self-learning systems, but a full grasp of the power units’ behavior remains a work in progress. The current regulations embed these adaptive features, and while planned changes for 2028 aim to adjust the power deployment split, they are not expected to completely resolve the unpredictability.
The challenges faced with the 2026 power units underline the increasing complexity of Formula 1® technology, where software-driven systems play a significant role in performance. As teams continue to refine their approaches, drivers and engineers alike are adapting to the new dynamics introduced by these self-learning engines.
At this stage, the situation remains fluid, with teams and drivers gaining experience with the systems over multiple race weekends. The ongoing efforts to decode and manage the power units will be a key technical focus for the remainder of the season.
