Mobile Phone and Seatbelt Technology screens out most camera photos

The Queensland Audit Office reports that in 2024 the MPST image-recognition AI cut the assessments needing human review at its vendor by 98.7%, to 2.7 million.

  • 1 an organization

    Reviewers at the vendor checked 2.7 million of the 208.4 million assessments that the AI made in 2024, a 98.7% cut in review work.

  • Not rated society

    The audit counts fines, and it does not measure whether the cameras reduced crashes or injuries.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Reported benefit.
Evidence

foundstudiedreportedclaimed

An independent audit report supports the benefit.

Compared with

Human review of every assessment that the AI made. Assumed without a measurement.

Without the system
208.4 million assessments performed by the AI in 2024
With it
2.7 million assessments reviewed by the external vendor
The system’s part
Not isolated. The 98.7% figure compares the assessments that the AI made with those that people still reviewed. The auditor does not compare the program with one run without AI or measure how many offenses the AI missed. The auditor describes the system as using machine learning and computer vision.
Limits
Not yet checked.
Costs and harms
Not yet checked.
Later
Not yet checked.

Download data Cite Suggest a correction

Benefit type
Repetitive Task Automation
System
Mobile Phone and Seatbelt Technology (MPST) image-recognition AI
Made by
Acusensus, the external vendor that owns and manages the AI (named as TMR's contractor in a ministerial answer, while the audit names no vendor)
Used by
Queensland Department of Transport and Main Roads, with the Queensland Revenue Office reviewing possible offenses
Where
Queensland, Australia
Stage
In routine use
Date
Kind
Aggregate result
Broad benefit
Identifying Objects, Actions, and Events, Evaluating Information to Determine Compliance with Standards
Collections
Scrutiny and findings
Last revised
31 October 2026

Sources

  1. Audit report, by authors with no tie to the system.

    Queensland Audit Office. Managing the ethical risks of artificial intelligence. Report 2: 2025–26, tabled 24 September 2025, chapter 5 (archived copy of the report page).

    In 2024, the AI system reduced the volume of assessments requiring human review at the external vendor by 98.7 per cent to 2.7 million.

  2. Other, by the organization that used it. Context or contrary evidence.

    Queensland Parliament. Question on Notice No. 1542, asked on 28 November 2023 and answered by the Minister for Transport and Main Roads.

    Since 2020, the Department of Transport and Main Roads (TMR) has established two separate Standing Offer Arrangements with Acusensus, one for the Mobile Phone and Seatbelt Safety Technology, and the other for the Transportable Road Safety Camera Solution.

Details

AI system

Workflow

Cameras photographed passing vehicles, and the AI assessed each photo for phone use and seatbelt offenses. Staff at the vendor and then at the Queensland Revenue Office reviewed each possible offense before a fine. Weekly audits and quarterly performance reviews fed back into the vendor's AI.

Follow-up

Conditions for replication

A person must check every possible offense before a fine. The contract must set performance indicators for the AI's accuracy.

Entry details

Sector

Traffic enforcement (road safety).

Purpose

Acusensus built the system to find possible mobile phone and seatbelt offenses in roadside camera photos for Queensland's transport department.

Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). Mobile Phone and Seatbelt Technology screens out most camera photos (AIBD-0677). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0677.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)

BibTeX Data Field definitions