Skip to main content
Promotional banner for a CUI buildout guide
IncidentSeverity M: moderate, 2/5Confirmed: a company, official or court statement

ACT mobile phone detection camera wrongly flagged a driver's card holder as a phone; court threw out $641 fine

On November 17, 2025, an ACT mobile device detection camera on Athllon Drive in Greenway captured a Canberra AI specialist driving with a card holder made from an old phone case on his lap. He was then issued a $641 mobile phone infringement with three demerit points. Access Canberra rejected his appeal. At an ACT Magistrates Court hearing on October 19, 2026, a government expert witness said she could see a screen in the photographs. The magistrate dismissed the charge and acquitted him. The case renewed political concern about camera accuracy, with MLA Peter Cain noting almost 58,000 mobile phone fines and no independent audit of the system.

What the AI did

An automated mobile phone detection camera in the ACT wrongly flagged a card holder made from an old phone case, resting on a driver's lap, as a mobile phone. This led to a $641 fine and three demerit points that a court later threw out.

First reported November 10, 2026 · Added to the register November 15, 2026 · 1 source

Deployer
ACT government / Access Canberra
Affected
Carl Jackson
Country
Australia (ACT)
When it happened
November 17, 2025 to October 19, 2026
First reported
November 10, 2026

What this means for you

Could this affect you?

Yes

If you drive in the ACT or use automated detection to issue penalties. Drivers in the ACT whose images are checked by the automated camera system can be wrongly fined. A local MLA said almost 58,000 mobile phone fines have been issued with no independent audit of the system, and the government had not disclosed the system's error rate when the driver in this case asked for it.

What to check

  • Publish the error rates of any automated detection system used to issue penalties.
  • Keep human review that checks the actual evidence before a penalty is issued.
  • Commission independent audits of the system's accuracy.

Areas of your AI programme this touches

Timeline

  1. November 17, 2025Started
  2. October 19, 2026Ended
  3. November 10, 2026First reported
Every fact and its source (8)
  1. Court hearing date2026-10-19
    “the hearing finally taking place on October 19, 2026”[1]
  2. Date camera captured driver2025-11-17
    “On the evening of November 17, 2025, Mr Jackson”[1]
  3. Scale of fines without auditAlmost 58,000 fines, over $36 million, no independent audit
    “Despite almost 58,000 mobile phone fines and over $36 million in penalties, there has been no independent audit of the traffic camera system.”[1]
  4. 2026 court matters372 lodged, 37 proven, six dismissed
    “372 mobile phone offence matters had been lodged with the courts this year as of October 22, with 37 proven and six dismissed.”[1]
  5. Item mistaken for a phoneCard holder made from old phone case
    “Resting on his lap was a card holder crafted from an old mobile phone case.”[1]
  6. Penalty issued$641 fine and three demerit points
    “the $641 fine and copping three demerit points”[1]
  7. Court outcomeCharge dismissed, driver acquitted
    “the magistrate dismissed the charge and acquitted Mr Jackson on the spot”[1]
  8. Undisclosed error rateGovernment did not disclose system error rate
    “Crucially, I asked for the system's error rate, a figure they still haven't disclosed.”[1]

Sources

  1. ACT traffic cameras flawed after AI expert beats $641 fine | Bay Post-Moruya Examiner
    batemansbaypost.com.au · November 10, 2026

How this record is classified. Severity M (2/5): harm to individuals, or a policy breach with limited reach. OECD level: incident, an event in which an AI system led to actual harm. Evidence: Confirmed, meaning a company, official or court statement.

The facts in At a glance and in Every fact and its source are quoted from the sources listed. The summary, What the AI did and What this means for you are written from those sources and checked against them automatically before publication. Records reflect what has been disclosed, not everything that has happened. Incidents are found through news feeds and the GDELT Project.

Promotional banner for the Penetration Report Template Kit