Key Concepts
Age assurance technology, age estimation, facial scanning, teen social media ban, accuracy of age verification, alternative age verification methods, government-backed ID, false positives, false negatives, privacy concerns, MyMahi, robustness of the system.
Age Estimation Technology and the Teen Social Media Ban
The Australian government is planning to implement a ban on social media for individuals under 16, scheduled to begin in December. To enforce this ban, the government has been trialing age assurance technology in schools for the past 4 months, involving over 1,000 students. These trials primarily focus on age estimation tools that use biological traits, particularly facial features, to guess a person's age.
Facial Scanning: The technology analyzes facial features like ear and nose size, and the distance between the eyes, to estimate age. However, the accuracy of these tools is inconsistent, with age estimations varying significantly for the same individual.
Examples of Inaccuracy:
- One person's age was estimated as 15, then 19 in quick succession.
- Another participant received age estimations ranging from 19 to 37.
- One participant was estimated to be 32, which was considered "insulting" as it was double their actual age.
Accuracy Concerns and Trial Results
While some age estimations are close to the actual age, the inconsistency raises concerns about the reliability of the technology. Preliminary results from the trial indicate that 85% of the age estimations were within 18 months of the student's reported age. Andrew Hammond, who is running the trial, initially hoped for a 12-month margin of error.
Consequences of Misjudgment: An 18-month margin of error could lead to "false positives" (incorrectly identifying someone as underage) or "false negatives" (incorrectly identifying someone as overage), resulting in unfair lockouts or allowing underage individuals to bypass the ban.
Expert Opinions: Experts suggest that the accuracy needs to be much higher, approaching 95% or more, to ensure robustness and public trust in the system.
Quote: "I think if the policy is relying on age estimation through these kinds of facial software technologies I don't really think that it should be going ahead at this point" - An expert expressing concern about the reliance on facial scanning technology.
Alternative Age Verification Methods
Facial scanning is not the only method being considered. Other options include:
- Voice and Hand Movement Analysis: Similar accuracy problems to facial scanning, with fewer companies offering these technologies and less testing.
- Age Inference: Using data like credit card information, .gov email addresses, or mortgages to infer age. This is less helpful for younger teenagers.
- Online Activity Analysis: Guessing age based on social media activity, friends, and interests, which is also unreliable.
- Government-Backed ID: Passports or driver's licenses are the strongest proof of age, but social media companies are not allowed to mandate them, and many young people don't possess them.
Amendment to the Bill: A last-minute amendment requires platforms to offer alternative age verification methods.
Quote: "We do know generally that young people are going to be less likely to have a government-issued ID that would satisfy some form of age verification so it's definitely more likely that for your typical you know 14 15 16y old they're going to have to rely on those alternative modes of confirming their age" - Highlighting the reliance on alternative methods for younger teens.
Gaming the System and Industry Response
Under 16s may attempt to bypass the ban by manipulating facial estimation tools, for example, by wearing glasses, using makeup, or changing their hairstyle.
Industry Perspective: Age verification companies acknowledge the accuracy limitations but argue that some tools are more accurate than others, especially for teenagers. They suggest a tiered approach similar to bottle shops, where age estimation is the first step, and those estimated to be younger than a certain age (e.g., 21 or 25) must provide stronger proof.
Third-Party Involvement: Schools could potentially get involved using platforms like MyMahi, which rely on school-provided data for age verification.
Quote: "You would never rely on age estimation for people who are literally at the age of 16 because it was never going to be good enough for that" - An industry representative acknowledging the limitations of age estimation for those close to the age threshold.
Government Stance and Public Expectations
The government acknowledges that the system won't be perfect and has avoided promising immediate solutions.
Quote: "We don't argue that the changes that we will be legislating uh will fix everything immediately" - Government acknowledging the limitations of the legislation.
Despite the limitations, the government believes the ban is a positive step for young people, limiting their exposure to potentially dangerous online environments.
Preliminary Findings: The official preliminary findings from the trial suggest that age assurance can be done effectively, and there are no significant technological barriers. However, the question remains whether the current level of accuracy is sufficient.
Synthesis/Conclusion
The Australian government's plan to ban social media for individuals under 16 relies heavily on age assurance technology, particularly facial scanning. While trials are underway, the accuracy of these tools remains a significant concern, with inconsistent results and a potentially large margin of error. Alternative age verification methods are being explored, but each has its own limitations. The success of the ban will depend on addressing these accuracy issues, implementing robust alternative methods, and managing public expectations. The government is expected to announce further details on the implementation of the ban in the coming months.
AI summaries can miss context or contain errors. Check important details against the original video.