AI-powered coaching reduces cellphone use behind the wheel by 9% in South African study

A recent study conducted by Discovery Insure in collaboration with Johns Hopkins University and British consultancy Affective Mobility has shown that AI-powered coaching can significantly improve driving behaviour. The research involved 5,000 motorists covering 15-million kilometres over 1.7-million journeys. The study found that AI-powered coaching reduced cellphone use behind the wheel by 9% and improved other driving habits, such as reducing harsh acceleration and braking by 5% each.

The AI-powered coaching used in the study combines machine learning with generative AI to identify driving patterns and select appropriate goals and interventions for each driver. According to Discovery Insure chief analytics officer Michael Rodewijk, the technology performs the work of a personal driving coach, providing personalised goals and feedback to drivers. The system generated 23,000 personalised goals, 390,000 coaching messages, and 160,000 notifications and voice alerts over four months.

The study drew on the Insight-Motivation-Action framework developed by behavioural scientist Dr Neale Kinnear of Affective Mobility. This framework provides a structure for creating personalised messages that include insight, motivation, and action. According to Rodewijk, each individual message is unique and tailored to the driver's specific needs and weaknesses. This approach allows for a high degree of personalisation, which is essential for changing driving behaviour.

The ability to identify individual drivers' weaknesses and tailor coaching accordingly is a major advance in applying behavioural science at scale. The researchers found that drivers with identical driving scores but different weaknesses required different coaching approaches. For example, a distracted driver who uses their phone frequently requires a different approach than a driver with braking problems.

Discovery Insure has also explored how habits outside the vehicle affect driving risk. In July, it released research linking four years of sleep records with driving information from more than 10,000 motorists. Accumulated sleep deprivation proved up to five times more predictive of accident risk than traditional insurance factors considered individually. Drivers with less than an hour of sleep debt over three consecutive nights had up to 36% lower accident risk than those with five hours or more.

Discovery Insure has introduced Vitality Drive Sleep Points to encourage healthier sleeping through additional rewards. The company has also launched an Advanced Driving Instructor, which uses large language models to create individual goals based on each customer's driving habits. According to Discovery Insure CEO Robert Attwell, the objective is to encourage positive behaviour change without punishing customers for poor sleep or driving habits.

While the study established improvements in driving habits, equivalent reductions in actual crashes remain to be demonstrated. However, Johns Hopkins associate professor Jonathan Ehsani notes that crashes are predictable and can be preventable. The study's findings demonstrate that personalised AI interventions can persuade drivers to change habits that increase their risk of accidents, and insurance companies can now influence individual conduct on a scale that was previously impossible.

Key points

• AI-powered coaching reduced cellphone use behind the wheel by 9% and improved other driving habits, such as reducing harsh acceleration and braking by 5% each.
• The technology combines machine learning with generative AI to identify driving patterns and select appropriate goals and interventions for each driver.
• Accumulated sleep deprivation proved up to five times more predictive of accident risk than traditional insurance factors considered individually.