Tesla needs artificial integrity over artificial intelligence.
The difference between artificially intelligent-led and integrity-led machines is simple: the former are designed so that we could. The latter because we should. This distinction underscores the growing need for artificial integrity.
A recent incident involving a Tesla Model Y, where a valley resident found himself trapped inside his car, illustrates the difference between intelligence-led and integrity-led machine design.
One day, the power to Diane's car unexpectedly shuts off, trapping her inside without opening the doors or windows. Even the glove box was locked with the owner's manual.
Desperate, Diane called a neighbor who couldn't open the car from the outside. Eventually, through the Tesla app, he learned via text about a hidden, unmarked manual latch under the armrest, which finally freed him.
Diane's case of being stuck in her Tesla Model Y due to a dead battery is not isolated and essentially highlights the limits of artificial integrity in Tesla's design.
Diane's car situation is characterized by an AI-First mode, which means a heavy reliance on AI and electronic systems, human agency through manual intervention capabilities, without sufficient consideration of human factors.
It failed to effectively prevent and deal with the power outage and ensure Diane's safety. It also failed to consider the psychological state of the person trapped in the vehicle and the associated stress of feeling powerless and not having control over the machine.
This scenario places an unnecessary burden on the driver's knowledge of the car's emergency features, which can prove fatal in high-stress situations.
A scenario reflecting artificial integrity would be to design the car to move from AI-first mode to fusion mode, thereby moving from a heavy reliance on AI to a synergistic approach where AI systems and Human intelligence works seamlessly.
This includes designing AI systems that not only manage technical aspects but also support human situations and responses, providing clear, accessible support and failsafes that consider human psychology and stress responses. do
Let's explore what could have been the scenarios where Diane's Tesla might have exhibited artificial integrity in its work from AI-first to fusion mode.
A low-level maturity scenario of synthetic integrity
Diane's car shuts down unexpectedly, but this time, the AI system detects an electrical fault and tries to help. A notification pops up on his phone, instructing him how to use the manual door release. The notification includes a clear diagram showing the location of the hidden latch.
In this scenario, the AI system represents a fundamental effort to transition from AI-first mode to fusion mode by providing real-time support through notifications. However, the system's ability to consider human factors and proactively support the situation is still limited. The AI tries to fill the gap by guiding Diane through emergency procedures, but the heavy reliance on external devices and the lack of intuitive design elements for manual overrides highlight the system's immature integration of synthetic integrity.
Synthetic integrity medium maturity scenario
As Diane's Tesla shuts off, the AI system detects a power loss and attempts to switch to a non-electronic backup manual mode. The AI system immediately displays a message on the central screen and on his phone, explaining the situation and guiding him to the manual release latch on the door. The latch, designed for easy access and visibility, glows softly to attract attention.
This scenario shows an improved attempt by the AI system to transition from AI-first mode to fusion mode. AI plays an active role in detecting power faults and providing quick guidance, thereby integrating human-AI interaction. However, the system's reliance on active electronic components and a backup mechanism highlights its limitations.
A high-level maturity scenario of synthetic integrity
As soon as Diane's Tesla is turned off, the AI system immediately detects a power loss and activates a secondary, non-electronic manual override for the doors and windows. The AI also performs regular system checks and sends alerts to inform Dian in advance of the state of the main and auxiliary batteries. Upon detection of an electrical fault, the doors and windows switch seamlessly to manual mode. Diane receives a detailed explanation on her phone of what happened, with step-by-step instructions displayed on the car's main screen and via an audible alert system. Additionally, a detailed explanation is sent to a trusted person (be it family, friend, etc.) designated by Diane to notify in advance of any problems encountered while using the vehicle. Additionally, AI raises alerts to Tesla's roadside assistance team and national emergency services if early warnings are not heeded.
This scenario shows a strong potential to move from an AI-first approach to a fusion mode, where AI and human intelligence work synergistically to ensure safety and reliability. AI systems play an active role in monitoring and communication while incorporating human-based design and redundancy to reduce the risk of failures and improve the overall user experience. This scenario overcomes some potential limitations through advanced AI functions, such as active monitoring, automatic failsafes, and multi-channel communication, which provide a comprehensive approach to driver safety and confidence.
We can think of other scenarios to ensure that Diane is driving a car that is not only led by artificial intelligence but also led by integrity, which would make her use her car. It is regrettable but prevents us from facing life-changing situations.
Artificial Integrity on Artificial Intelligence represents this paradigm shift.