The Weekly Cybers #81
It’s becoming predictable: social media age restrictions in the news again, AI and next week’s productivity roundtable, new cybercrime statistics, and much more.
15 August 2025
Welcome to a week where everything’s the same
Sadly, digital policy news is settling into a pattern, but I guess that’s what we’re here for.
The two bigger stories in Australia are the social media age restrictions and the regulation of AI, and that’s probably going to remain the case for at least the next couple of weeks.
Still, we have some new if slightly dated cybercrime figures, there’s the usual smattering of AI news here and abroad, and there’s a surprising recommendation from the UK government.
Read on!
Google versus eSafety emails revealed, and other social media age restrictions news
It looks like we’re going to see quite a bit of pushing and shoving over Australia’s social media age restrictions, which come into force in December. This week I have three stories.
I haven’t gone through it yet, but the Senate demanded that the comms minister cough up documents relating to the legislative rules which determine which platforms are induced in the system. And here they are. There’s a LOT of redactions.
The Guardian has, though, and they focus on the question of whether Google should be in or out.
The eSafety Commissioner, Julie Inman Grant, had warned the minister that famous influencers would wage a “ground war” on the government, they report.
“We all know that this is about monetisation of the millions of under 16’s on these platforms who HAVE accounts today – but may then not at commencement,” she wrote.
“This will have an impact on attracting certain advertisers that may be targeting the 8-15 demographic,” and that this in turn would hit the influencers’ income.
Where do these online harms come from?
Meanwhile, Crikey reports that there’s been a crucial shift of emphasis. Yes, this is all about protecting teens from online harms. Originally, though, it was about harms such as “cyberbullying, body image issues, eating disorders and addiction to scrolling”.
Now it’s “addictive behaviours caused by persuasive or manipulative design features, social isolation, sleep interference, poor mental and physical health (including unhealthy social comparisons and negative body image), low life satisfaction and exposure to inappropriate and harmful content”.
As Crikey’s Cam Wilson observes, “Apparently, the main problem with social media platforms isn’t the content they host, but the technology itself... If the major concern is the technology itself, then it’s less about Meta, TikTok or YouTube doing a better job at moderating their platforms, and more about the fundamental design of their services.”
EFA challenges age assurance tech trial’s “success”
Digital rights lobby group Electronic Frontiers Australia (EFA) has challenged both the methodology and the validity of some conclusions of the age issuance technology trial, which claimed in its interim findings that the technology “can be private, robust and effective”.
These findings were “strong on hype and rhetoric and difficult to reconcile with the evidence,” according to EFA chair John Pane, who was also a member of the trial’s Stakeholder Advisory Board.
In particular, he says the trial “failed to address the immaturity of some technologies”; the assessment of vendors’ privacy practices “appears to be based on a low bar, such as the presence of a privacy policy”; and there was “concerning evidence” that some vendors were proactively creating “regulatory backdoors” to retain personal data, possibly including biometric data, for potential anticipated requests from law enforcement, coroners, or regulators — even though there was no legal obligation to do so.
The full statement has yet to appear on EFA’s website, but it has been posted on Bluesky by Cam Wilson.
The tech trial’s final report was presented to the government on 1 August, but not yet made public.
AI in the news before productivity roundtable
Artificial intelligence continues to be in the news in the lead-up to next week’s Economic Reform Roundtable, often referred to in the media as the productivity roundtable. Here’s just some of the stories.
- Human rights commissioner Lorraine Finlay says AI could worsen racism and sexism. “When you combine algorithmic bias with automation bias — which is where humans are more likely to rely on the decisions of machines and almost replace their own thinking — there’s a real risk that what we’re actually creating is discrimination and bias in a form where it’s so entrenched, we’re perhaps not even aware that it’s occurring,” she said.
- “It would be harsh, but not entirely inaccurate, to observe that [Tech Council chair] Scott Farquhar’s contribution to the intellectual property debate has been ironically light on intellect,” wrote media analyst Tim Burrowes. Meanwhile the Guardian reckons Farquhar has forgotten that even in the US the application of their copyright law’s fair use doctrine is far from settled.
- Job Skills Australia says that the doomsday predictions about the end of work as we know it are overblown. You’ll be fine if you’re in cleaning, construction, labouring, and perhaps hospitality.
Cybercrime hits fewer, but impacts are greater
The Australian Institute of Criminology (AIC) has released its report Cybercrime in Australia 2024 — yes, nearly eight months into the following year — and there are signs that some things could be getting better. Maybe.
Respondents were less likely to have been a victim of online abuse and harassment, malware, and fraud and scams in 2024 when compared with 2023. Small and medium business owners were more likely to seek help from police or otherwise report the crime.
But it’s all relative.
Overall harm to victims was up, thanks to increased social and health impacts.
Respondents were also less likely to have used important online safety behaviours, such as checking and changing privacy settings, using antivirus, avoiding clicking on unknown links, and independently contacting companies when unsure about texts and emails.
Parents were less likely to have used parental controls.
“Overall, we found that rates of victimisation remain high, formal help-seeking remains low, and a large proportion of victims are negatively impacted by cybercrime,” the AIC writes.
NEED A MUSIC BREAK? Why not try the second episode (not counting the pilots) of Another Untitled Music Podcast, in which my good friend Snarky Platypus and I bring you a selection of songs with “Stay” in the title. Look for it in your podcast app of choice.
Also in the news
- From The Mandarin, the observation that first it was robodebt, and now it’s robodole. “Unlawful automation, vulnerable people hurt, and the APS again accused of moving too slowly. It’s déjà vu all over again.”
- From InnovationAus, “A Federal Court judge found the smartphone app stores of Apple and Google engaged in anticompetitive conduct in Australia, in a partial victory for Fortnite game developer Epic Games against the tech giants”.
- From iTnews, “NBN Co is embedding AI and automation ‘into the fabric of [its] operations‘.”
- One of Australia’s most respected experts on the ethics of AI, Kobi Lens, told a medical specialist she didn’t want them use AI transcription software. She was told to go elsewhere.
- Also via *iTnews, “The Australian Taxation Office (ATO) is exploring the use of artificial intelligence to assist its 800 core developers with addressing common programming issues.”
- The ATO has also enabled instant account locking to counter real-time payments scams.
- “‘Australiana’ images made by AI are racist and full of tired cliches,” reports The Conversation. Nit a surprise, of course.
- CSIRO has developed technology which might help stop unauthorised AI learning from photos, artwork, and other image-based content.
- From Cyber Daily, a NSW man out on bail has to send a selfie to NSW Police every hour of the day, at least from 8am to 8pm — and to pay for an iPhone for the supervising cop. Still, he could have a lot of fun with this, I reckon.
- The Australian Bureau of Statistics has started recruiting its temporary workers for next year’s Census, and they want to reflect community diversity. So I guess they weren’t already?
IF YOU’VE FOUND THIS NEWSLETTER HELPFUL, PLEASE SUPPORT IT: The Weekly Cybers is currently unfunded. It’d be lovely if you threw a few dollars into the tip jar at stilgherrian.com/tip.
Elsewhere
- Indonesia is considering banning Roblox over child safety concerns and “immoral” content.
- Meta’s rules for AI chatbots have allowed them to hold “sensual” chats with kids, and offer false medical information. In one case it led to a man’s death.
- Wikipedia has lost a UK court case relating to that country’s Online Safety Act and the need to verify users’ ages.
- A new study suggests that using AI made doctors less skilled at spotting cancer in just a few months.
- From The Observer last week, “America’s plan to use AI as a political weapon”. It’s also a reflection on how Silicon Valley once considered itself above politics.
- Apparently “people with lower AI literacy perceive AI to be more magical, and thus experience greater feelings of awe when thinking about AI completing tasks,” which means there’s an incentive to make AI seem like magic rather than properly informing potential users.
- The UK government issued the remarkable advice to delete your old emails to save water. Over at Pivot to AI, David Gerard explains my that’s very silly and tries to get an answer.
- Finally, some digital economy magic. “Independent study reveals USD 1.2 billion economic impact of foodpanda in Pakistan for 2023-2024,” reads the press release. Lol no it didn’t. The people of Pakistan were just spending that $1.2 billion some other way. But sure. (Economists are welcome to explain why I’m wrong, for I am not an economist.)
Inquiries of note
Nothing new this week.
Related news digests
I’ve mentioned Cam Wilson a couple of times this week. Yes, he writes for Crikey. But he also produced the daily newsletter The Sizzle, “Australia’s favourite daily technology email newsletter featuring the top news, deals and fun things”.
It’s much better than my terrible newsletter for people more into the technology itself rather than the policy discussions. It isn’t free, but you can give it a two-week free trial to see if it’s right for you.
What’s next?
Parliament returns on 25 August, which is 10 days away.
Before then, though, the government’s Economic Reform Roundtable runs for three days next week, Tuesday 19 to Thursday 21 August. The agenda (PDF) shows a session on “AI and innovation” on Wednesday afternoon.
Amusingly, Treasury’s list of pre-approved possible outcomes was leaked, so watch this space.
DOES SOMETHING IN THE EMAIL LOOK WRONG? Let me know. If there’s ever a factual error, editing mistake, or confusing typo, it’ll be corrected in the web archives.
The Weekly Cybers is a personal look at what the Australian government has been saying and doing in the digital and cyber realms, on various adjacent topics, and whatever else interests me, Stilgherrian, published every Friday afternoon (nearly).
If I’ve missed anything, or if there’s any specific items you’d like me to follow, please let me know.
If you find this newsletter useful, please consider throwing a tip into the tip jar.
This is not specifically a cyber security newsletter. For that that I recommend Risky Biz News and Cyber Daily, among others.