Then she answered.
“No.”
“She would probably try to make you explain why.”
“I don’t know why.”
“That would irritate her.”
“Then she’d accept it.”
The boy nodded.
“Good.”
He returned to his headphones.
The daughter watched the car complete the straight.
Family legacy had finally become light enough that a child could ignore it.
That felt like success.
The company faced its next major ethical crisis fifteen years after the former CEO stepped down.
By then vehicles were increasingly autonomous.
Remote intervention systems had become standard in commercial fleets.
Not crude kill switches.
Complex supervisory controls.
Emergency steering.
Speed restriction.
Automated safe-stop functions.
Cybersecurity had improved dramatically.
So had the sophistication of attacks.
A junior software engineer at the company discovered something troubling during routine code review.
A new fleet platform contained an emergency command path that could force any connected vehicle into a controlled stop.
The feature was legal.
Documented internally.
Designed for stolen trucks and dangerous autonomous malfunction.
But the engineer noticed one undocumented function.
An administrator with sufficiently high privileges could suppress the driver notification normally displayed before remote intervention.
The code existed for law-enforcement emergency scenarios.
It required two approvals.
In theory.
In practice, one legacy compatibility service could generate both tokens if configured incorrectly.
The engineer filed an internal concern.
Her manager replied:
“Known technical debt.”
She filed again.
Cybersecurity office replied:
“Low probability.”
She requested escalation.
No response for nine days.
On the tenth, while waiting for a meeting, she visited the company museum.
Project Twelve was away at the garage that month.
But the $5 million hypercar sat in the remote-control ethics exhibit.
The display showed the hidden relay.
The wiring splice.
The timeline of the desert incident.
One line caught her attention:
Useful systems carry abuse surfaces. The relevant question is not merely whether a function works, but who may invoke it, how invocation is logged, and whether affected people can know.
The quote came from the single father.
The engineer stood there for a long time.
Then photographed it.
She returned upstairs.
Opened her concern ticket.
Attached the image.
Wrote:
We have already had this failure culturally once.
That got attention.
Not because history magically solved the problem.
Because the company’s current governance required any report explicitly linking modern control architecture to the historic remote-stop case to be reviewed by an independent safety committee.
That rule had been created decades earlier.
The committee examined.
The junior engineer was right.
The compatibility service created a possible single-point privilege bypass.
No evidence anyone had exploited it.
Yet.
The company delayed release.
Fixed.
Published advisory.
Lost approximately ninety million dollars in launch timing and contract penalties.
Several investors complained.
The current CEO appeared before analysts.
One asked:
“You delayed a profitable product over a theoretical security scenario?”
The CEO answered:
“Security incidents are theoretical until they happen.”
Another asked:
“Was this decision influenced by the company’s historic scandal?”