Tesla FSD Moves Toward 'Personalization'! Musk: Will Learn Driver Habits to Solve HOV Lane Misjudgment

Tesla CEO Elon Musk recently confirmed that the company's Full Self-Driving (FSD) system is transitioning from the past 'one logic fits all owners' general model to a truly personalized driving system. In the future, the neural network behind FSD will directly learn each owner's driving habits and remember user preferences, solving long-standing criticized issues such as the 'forced exit from High-Occupancy Vehicle (HOV) lanes'.

A Tesla owner posted on X, complaining that FSD frequently exits clear HOV lanes at inappropriate times, merging into congested traffic and reducing driving efficiency.

In response, Musk stated that the system will begin recording owners' manual intervention behaviors and adjust based on individual preferences.

Human drivers make different choices in various scenarios. For example, when parking, some prefer backing in, while others prefer pulling in front-first. FSD will now learn these differentiated behaviors and match the driving model closest to the owner's style.

This update also continues Tesla's recent simplification of HOV lane settings. If the system can learn from owners' repeated takeover actions, it will effectively reduce similar misjudgments.

Additionally, Tesla's recently launched 'Takeover Feedback Menu' requires owners to report the reason for each intervention. This data is helping the engineering team clarify priority correction items. Currently, parking remains one of the main reasons for FSD takeovers, and the personalized feature is expected to first improve this experience.

The market expects that FSD v15, featuring an approximately 10-fold increase in parameter scale, will be released by the end of this year or early next year, further enhancing autonomous driving capabilities.