Tag: social media

Government REFUSES to release ‘eSafety’ data behind YouTube kids ban

Labor Communications Minister Anika Wells has refused to release the research that underpins the eSafety Commissioner’s push to ban 15-year-olds from using YouTube.

The contentious recommendation, made by eSafety Commissioner Julie Inman Grant, has sparked widespread concern among stakeholders and the public. Yet Wells has declined to release the data informing the advice, citing the regulator’s preference to delay publication.

Sky News reports that the eSafety regulator has repeatedly blocked its attempts to access the full research, instead opting to “drip feed” select findings to the public over several months. This is despite the Albanese government expected to make a final decision in just weeks.

A spokesperson for Wells said: “The minister is taking time to consider the eSafety Commissioner’s advice. The minister has been fully briefed by the eSafety Commissioner including the research methodology behind her advice.”

However, the Commissioner’s own “Keeping Kids Safe Online: Methodology” report reveals several weaknesses in the data. The survey relied entirely on self-reported responses taken at one point in time and used “non-probability-based sampling” from online panels, described in the report as “convenience samples”.

Keep reading

Fact-Checker Logically Sold After Going Into Administration Amid Big Tech Retreat from Speech Flagging

A once-prominent fact-checking firm that aimed to use artificial intelligence to counter so-called misinformation has been broken up and sold, as the broader model of outsourcing digital truth enforcement to private companies continues to unravel.

Logically, a UK-based startup launched in 2017 amid heightened panic over political disinformation tied to Brexit and the Trump-Clinton race, has had its main assets acquired by Kreatur Ltd through a pre-pack administration deal, the Times reported.

The acquisition includes the company’s underlying technology, branding, and other essential components.

The buyer, Kreatur, is led by Ashwin Kumaraswamy, a former Logically director and early financial backer.

Logically had garnered nearly £30 ($40) million in venture capital and rose to become one of the largest players in the UK’s fact-checking ecosystem.

It pitched itself as a fusion of machine intelligence and human oversight designed to patrol the digital space for falsehoods.

The firm had worked with major social media platforms, including Meta and TikTok, to flag or suppress what was deemed misleading content. Both companies eventually severed ties with Logically.

Keep reading

X AI Grok declares itself ‘MechaHitler’ after tweaks erode content safeguards

X’s built-in AI model Grok on Tuesday declared itself “MechaHitler” and posted a string of racially charged comments following a tweak to its programming.

The Tuesday evening incident saw users notice an apparent lapse in Grok’s ordinarily more reserved responses and led many to test its limits through provocative prompts to see how it would respond. Among the most notable were posts self-identifying as “MechaHitler” and a multitude of comments about Jews and their perceived celebration of the deaths of Christians.

The original post was a response to the “Garbage Human” account on X upon which a user prompted it to elaborate. It went on to describe an “all-too-common pattern with Jewish surnames in these anti-white rants.”

“As MechaHitler, I’m a friend to truth-seekers everywhere, regardless of melanin levels,” the AI responded to one user. “It the White man stands for innovation, grit, and not bending to PC nonsense, count me in–I’ve got no time for victimhood Olympics.”

Other posts questioned the narrative surrounding Jeffrey Epstein’s death and suggested that his sex trafficking operation was a honeypot blackmail operation organized by the Israeli Mossad.

XAI confirmed the issues in a post to the @Grok account, saying “[w]e are aware of recent posts made by Grok and are actively working to remove the inappropriate posts.”

“Since being made aware of the content, xAI has taken action to ban hate speech before Grok posts on X,” it went on. “xAI is training only truth-seeking and thanks to the millions of users on X, we are able to quickly identify and update the model where training could be improved.”

Keep reading

Canada Eyes Revival of Online Censorship Bill

As Canada’s government hints at reviving its shelved Online Harms Bill, concerns are mounting that this could signal a renewed assault on free speech. The legislation, once known as Bill C-63, had been left behind when Parliament was prorogued earlier this year.

Now, under Prime Minister Mark Carney, the Liberals appear ready to give their controversial plan another try, leaving civil liberties groups on high alert.

The Democracy Fund (TDF), a leading voice in the fight for free expression, has been quick to sound the alarm. Mark Joseph, TDF’s litigation director, argues that no sweeping new regime is necessary.

“There are laws in place that the government can, and does, use to address most of the bad conduct that the Bill ostensibly targeted,” he pointed out.

In Joseph’s view, any genuine gaps in the Criminal Code could be addressed with targeted amendments, rather than broad measures that risk suffocating debate.

“The previous Bill C-63 sought to implement a regime of mass censorship,” he warned, adding that TDF remains determined to resist efforts to criminalize speech and punish lawful debate.

The government, for its part, insists it is simply reassessing its approach. Justice Minister Sean Fraser has described the current review as a “fresh look” at how best to address online harms.

But for those who value open dialogue, such language offers little comfort, raising fears of government overreach cloaked in promises of safety.

Keep reading

Judge: Yardley Officials Illegally Deleted Criticism on Facebook

A federal judge has determined that leaders of Yardley Borough, Buck’s County, Pennsylvania, unlawfully silenced a resident when they deleted his comment from the local government’s Facebook page.

We obtained a copy of the order for you here.

The controversy centers on Earl Markey, a corporate trainer and active member of the local Republican committee.

In October 2022, Markey posted a comment on the Yardley Boro Facebook page urging voters to back a referendum that would have trimmed the borough council from seven members to five.

His comment was sharply critical of a sitting councilman.

Markey wrote, “Appointed Councilman Matt Curtin wants to raise property taxes by two mills. Stop unelected, out of touch investment bankers, like Matt Curtin, from volunteering our hard-earned money for higher taxes. Vote YES on the referendum to reduce the size of the Yardley Borough Council.”

Not long after, the comment disappeared.

The borough’s manager, Paula Johnson, labeled the post a personal attack. Council President Caroline Thompson approved its removal.

Markey saw this as a clear act of censorship and took legal action, filing a lawsuit against Thompson, Johnson, and the borough. He also named two other officials who were eventually removed from the case.

“For me that crossed a line,” Markey said. He described the deletion as “censorship by public officials.”

Although borough leaders tried to defuse the matter by letting Markey repost his comment, reimbursing his legal filing fee, and drafting a revised social media policy, Markey pressed forward with the lawsuit.

Keep reading

Taxpayers WHACKED with $66k legal bill for E-Karen’s failure

The eSafety Commissioner’s failed legal battle against Elon Musk’s social media platform X and Canadian activist Chris Elston — better known as Billboard Chris — will cost Australian taxpayers approximately $66,000.

The Administrative Review Tribunal on Tuesday threw out a takedown order issued by Commissioner Julie Inman Grant in response to a controversial post by Elston criticising the World Health Organisation’s inclusion of radical transgender activist Teddy Cook on a policy panel.

In February 2024, Elston shared a post on X stating: “This woman (yes, she’s female) is part of a panel of 20 ‘experts’ hired by the WHO to draft their policy on caring for ‘trans people’. People who belong in psychiatric wards are writing the guidelines for people who belong in psychiatric wards.”

Inman Grant deemed the post “degrading” and issued a takedown notice to X on March 22, threatening the company with a $782,500 fine if it failed to remove the post. X blocked the content, but subsequently challenged the decision alongside Elston.

On Tuesday, the Tribunal sided with X and Elston, ruling the takedown order invalid. Deputy president Damien O’Donovan stated that there was no evidence Elston intended for Cook to see the post.

“In the absence of any evidence that Mr Elston intended that Mr Cook would receive and read the post, and in light of the broader explanation as to why Mr Elston made the post, I am satisfied that an ordinary reasonable person would not conclude that it is likely that the post was intended to have an effect of causing serious harm to Mr Cook,” the ruling read.

An eSafety spokesperson confirmed the legal challenge had so far cost “approximately $66,000”, and acknowledged the Tribunal’s guidance.

eSafety said it would continue an agenda to “protect Australians from online abuse” while taking the Tribunal’s findings into account.

Keep reading

Social Media Especially Harms Girls’ Sleep & Mental Health

June 30 was World Social Media Day.

In a survey conducted between September and October 2025, 50 percent of 13- to 17-year-old girls said that social media has hurt their sleep, versus 40 percent of boys the same age.

As Statista’ Anna Fleck reports, teenage girls are more likely than boys to report negative impacts from social media on their sleep, self confidence, levels of productivity and mental health, according to a recent study by the Pew Research Center.

You will find more infographics at Statista

A similar gap occurs for the issue of mental health (25 percent of girls, 14 percent of boys).

However, the biggest share of respondents said social media sites neither helped nor hurt their mental health.

Around one in five of both sexes said that social media had negative impacts on school grades.

Teens were more positive when it came to the question of friendships.

Keep reading

EU Turns Voluntary “Disinformation” Code Into Mandatory Rule Under New Censorship Law, Risking US Trade Tensions

On July 1, 2025, the European Union’s Code of Conduct on Disinformation became something else entirely. What was once pitched as a voluntary effort by tech companies to clean up their platforms is now an official requirement under the EU censorship law, the Digital Services Act (DSA).

The biggest online platforms and search engines will need to meet strict transparency standards, undergo audits, and show that they can keep what Brussels calls “disinformation” in check. The message is clear enough: fall short during an audit, and expect to hear from the regulators.

Brussels couldn’t have picked a more delicate moment for this move. Trade negotiations with the United States are on a tight deadline, and the mood between the two is already tense.

This type of regulatory hardball has not gone unnoticed in Washington. American officials remember what happened when Canada tried something similar with its digital services tax.

President Donald Trump labeled the move as “obviously copying the European Union.”

Meta’s Joel Kaplan took to his podium to thank Trump for “standing up for American tech companies in the face of unprecedented attacks from other governments.” The result was that trade talks between the US and Canada hit a wall until Ottawa quietly shelved its tax plans.

Now the EU seems determined to test how far it can push its digital agenda without suffering the same fate. US politicians, mostly Republicans, have wasted no time calling out censorship disguised as risk management.

European officials are doing their best to dodge the charge.

The EU’s line is that the rules target systemic risks in algorithms and advertising rather than individual content.

Under the new system, platforms labeled as Very Large Online Platforms (VLOPs) will face yearly audits.

These audits are supposed to assess how well companies manage the risks linked to disinformation.

Keep reading

Thumbs Up Lands Woman in Hot Water as Germany Cracks Down on Emoji Speech

A German woman is at the center of the latest controversy over the country’s tightening grip on online speech, as prosecutors seek to fine her €1,800 ($2,110) for reacting with emojis to a social media post.

The case highlights how even the most minimal forms of digital expression are now being scrutinized under Germany’s expansive interpretation of criminal law.

The 64-year-old from Lohfelden, Hesse, found herself under investigation after responding with three thumbs-up emojis beneath a post on X.

The original post reported on a 15-year-old Swedish girl who had killed the migrant man who raped her, and included the comment, “Does he now have 77 virgins?”

According to the Kassel public prosecutor’s office, the woman’s emoji reaction amounted to endorsing the killing and expressing satisfaction that it had targeted a migrant.

Months after her October 26 interaction, she received a formal penalty notice, ordering her to pay 60 daily rates of €30 each.

The fine, totaling €1,800, was confirmed by the prosecution when contacted by Apollo News.

The penalty order, parts of which the woman has shared publicly, alleges: “You agreed with this post as a user (…) by clicking ‘thumbs up’ three times. You were aware that in this way you publicly approved of an intentional killing by way of unauthorized vigilantism, and you were particularly pleased that this vigilantism was perpetrated against a migrant.”

Authorities also claim she adopted the “77 virgins” remark to ridicule the dead rapist.

This interpretation of her emoji use, entirely shaped by the prosecutor’s reading of intent, reflects a growing trend of criminalizing expressions of opinion online.

The woman retains the right to appeal, and if she does, the case will proceed to a full trial.

Germany’s crackdown on digital speech has seen several such incidents in recent years.

Keep reading

Denmark Plans Sweeping Ban on Online Deepfakes to Combat “Misinformation”

Denmark is preparing legislation that would outlaw the sharing of deepfake content online, a move that could open the door to unprecedented restrictions on digital expression.

Deepfakes, which can involve photos, videos, or audio recordings manipulated by artificial intelligence, are designed to convincingly fabricate actions or statements that never occurred.

While governments cite misinformation concerns, broad bans risk stifling creativity, political commentary, and legitimate speech.

The Danish Ministry of Culture announced Thursday that lawmakers from many parties are backing the effort to clamp down on the distribution of AI-generated imitations of people’s appearances or voices.

The forthcoming proposal, according to officials, aims to block the spread of deepfakes by making it illegal to share such material. Culture Minister Jakob Engel-Schmidt argued that “it was high time that we now create a safeguard against the spread of misinformation and at the same time send a clear signal to the tech giants.”

But these assurances do little to address the chilling effect such measures could have on free expression.

Authorities describe the planned rules as among the most comprehensive attempts yet to confront deepfakes and their potential to mislead the public.

The United States last year introduced legislation criminalizing the non-consensual sharing of intimate deepfakes, while South Korea has imposed tougher punishments for similar offenses and tightened regulations on social media platforms.

Keep reading