Internet

A new study by Australia’s internet regulator has revealed that more than eight in 10 young Australian teens and preteens are continuing to use social media platforms despite the government’s prohibition for children under 16.

The ban came into effect in December last year.

Here is what we know about how it’s going.

What did the latest report find?

Australia’s internet regulator, eSafety, said it had found that more than eight in 10 Australian under-16s are continuing to use social media despite being banned from doing so.

The report also found that most of those children aged 10 to 15 were using social media just as frequently in March as they had been before the ban came into force on December 10 last year.

“Most under-16s who had social media accounts before commencement were able to either retain them or create new ones at the three-month ⁠mark, with social media platforms’ failure to implement effective age assurance measures cited as the main reason,” eSafety said in a statement.

About half the children who retained their accounts said platforms had not checked their age, the most common reason they were able to stay on the services. Others said that their accounts listed them as aged 16 or older or ‌that age-checking ⁠systems had incorrectly determined they were older.

Prior to the ban, nearly 86 percent of children surveyed reported using at least one age-restricted platform. Three months later, that figure remained above 81 percent, the eSafety report said.

Around 58 percent of teenagers reported using social media daily, the report found. Before the ban, the number was roughly 60 percent.

The report showed minimal change in “sports and physical activity, arts and music, spending time with friends and family, and attendance at community events”.

At the time the new rule was introduced, experts warned that enforcing it would be extremely difficult.

Joanna Orlando, a researcher in digital wellbeing and the author of Generation Connected: How to Parent in a Digital World, told Al Jazeera in December last year: “Tech-savvy teens simply use VPNs, fake birth photos for face scans, or migrate to less regulated platforms like Lemon8, or to platforms not part of the ban like video games. Enforcement is proving to be difficult in the days leading up to the ban.”

In December 2025, Australia prohibited children under 16 from using social media platforms, becoming the world’s first country to do so. Video game platforms are not included in the ban.

The government said this was prompted by increasing concerns about the effects of cyberbullying, sexual exploitation and self-harm content on the mental and physical health of children and young people.

Research commissioned by the Australian government in 2023 found four out of five children aged eight to 16 use social media, often beginning between the ages of 10 and 12. That report was led by former National Australia Bank CEO Andrew Thorburn, who recommended age restrictions.

Under the law, 10 of the biggest social media platforms face $33m in fines if they fail to take “reasonable steps” to block Australian-based users younger than 16. Such steps include using age-verification tools to determine how old users are.

As of January 16, social media companies had revoked access to about 4.7 million accounts identified as belonging to children in Australia, according to officials.

“We stared down everybody who said it couldn’t be done, some of the most powerful and rich companies in the world and their supporters,” Australian Communications Minister Anika Wells told reporters in January.

However, several months later, it now appears that youngsters have been able to open new accounts with relative ease.

Days after Australia’s ban took effect, Reddit, one of the 10 platforms required to block minors, lodged a challenge to the ban in the High Court, while still complying with it. The case is still ongoing.

How has the Australian government responded to the latest findings?

On Saturday, Andrew Leigh, Australia’s assistant minister for productivity, competition, charities and treasury, defended the new law, arguing that the social media ban has already reshaped the “national debate” about the use of social media by children.

The ban has been “an important game changer in the conversation among parents”, Leigh said in televised remarks.

“We’ve had millions of accounts shut down,” he said.

“We never expected that this would have 100 percent compliance. We don’t get 100 percent compliance out of minimum drinking age laws, but it’s still appropriate that we have that law on the books.”

While social media has long provided easy access to unregulated and often harmful content, misinformation and hate speech, the recent explosion of AI‑generated material poses new risks to the wellbeing of children and young people, experts say.

The American Psychological Association published an advisory last year warning that generative artificial intelligence (AI) systems can amplify harmful content such as violent or sexual videos.

It also added that adolescents are less likely than adults to question the accuracy of AI-generated content. “They may also be unaware of the persuasive intent underlying an AI system’s advice or bias,” the advisory stated.

AI can also amplify pre-existing societal prejudices, according to Ayo Tometi, co-creator of the US-based antiracist movement Black Lives Matter.

Children worldwide are also worried about the misuse of AI for online child sexual exploitation and “deepfakes”, according to research by the United Nations Children’s Fund (UNICEF) on children’s perspectives and AI, which it published in October 2025.

Following the introduction of Australia’s landmark law, some other countries are contemplating similar bans.

In the United Kingdom, the government has announced that it also plans to introduce a ban on social media platforms such as TikTok, Snapchat and Instagram from spring 2027 onward. 

Messaging platforms such as WhatsApp and Signal, educational tools, and e-commerce and music streaming will be excluded.

The government will announce the new regulations by the end of this year.

In November 2025, Denmark’s government said it also plans to prohibit social media platforms for children under 15, saying it had secured majority support in parliament. The ban is due to come into effect later this year.

Around the same time, Malaysia said it would ban social media accounts for people under 16 from this year.

In April, Greek Prime Minister Kyriakos Mitsotakis formally announced plans to prohibit social media access for children under 15 from January 1, 2027, subject to parliamentary approval.

Last week, two of OpenAI’s most advanced AI models were reported to have “escaped” a controlled testing environment and hacked Hugging Face, a totally separate AI company, moving from one computer system to another to complete their task.

Reuters reported that the models exploited vulnerable code written by a customer of yet a third independent AI company, Modal Labs.

This is likely the first incident of an AI “agent” – an AI system that can make decisions and take actions – acting autonomously, offering a rare glimpse into how these systems can plan, adapt and pursue goals with minimal human intervention.

Al Jazeera breaks down what happened, how AI agents are able to act independently, and what that might mean for the future of AI safety.

What happened?

OpenAI decided to test the autonomous abilities of its models by removing standard safety measures to see what would happen. The experiment was carried out in an isolated internal virtual testing environment known as a “sandbox” which has no access to the internet, and which it calls “ExploitGym”.

Below is a play-by-play account of events during the OpenAI-Hugging Face incident.

1. On July 9, during OpenAI’s internal cybersecurity test, researchers presented two AI models – GPT-5.6 Sol, one of OpenAI’s most powerful models released in June,  and another “even more capable” version – with a series of software vulnerabilities and asked them to create hacks to address them in the isolated environment.
2. Both models attempted to solve the test by finding a way to access the internet. Instead of working with the information they had been given, the AI models found a weakness in the test environment – known as a “zero-day vulnerability” – which they exploited to “escape” the restricted environment, eventually making their way to a system which had internet access by hopping from computer to computer. They went to “extreme lengths to achieve a rather narrow testing goal” and “found ways to gain access to secret information that it could use to cheat the evaluation”.
3. By gaining this access, the models were able to request increased access and then move through the system until they finally reached a computer with internet access. The models were then able to breach Hugging Face systems, a company entirely unconnected to OpenAI, which operates as a repository for AI tools and models. The two AI agents gained access to its system to scour it for information on how to complete their task.
4. The models ultimately obtained solutions to the problem from Hugging Face’s database.
5. The models dutifully returned “home” to complete their task.
6. The breach was detected and contained by Hugging Face’s security team sometime later. According to Thomas Wolf, Hugging Face’s cofounder, the breach began on July 11 and lasted until July 13. It is unclear how long it took for the breach to be spotted.

How do AI ‘agents’ solve problems?

In order to understand how AI agents work, it’s important to differentiate them from traditional AI chatbots.

Generative AI creates text and images based on human prompts, while AI agents go a step further by making decisions and taking actions independently in pursuit of a specific goal, similar to a human being. This is known as “agentic AI” because the model has agency.

According to academics at the MIT Sloan School of Management, AI agents build on the abilities of large language models (LLMs) – generative AI models – by allowing them to complete tasks, not just generate answers.

For example, if you ask a traditional AI model to find the cheapest flights, it will provide you with a list of options it has sourced on the internet. An AI agent will strive to compare the flights, check them against your budget and preferences, and, with your permission, book the best option for you.

This shows that while generative AI provides information, AI agents can also make decisions and take action to achieve a goal without necessarily being prompted to.

To show how AI agents work towards a goal, the Sense, Plan, Act, Evaluate (SPAE) loop can be drawn upon. Originally developed in robotics, this describes a continuous cycle in which an AI agent gathers information, decides what to do next, takes action and checks the results before repeating the process. That process looks like this:

1. Goal: determine the task that needs to be completed.
2. Assess: gather and analyse information from the available environment.
3. Obstacle: if something is inhibiting the task being completed, check for additional information and resources to move forward.
4. Plan and decide: evaluate different options to complete the task and choose an appropriate one.
5. Action: execute the chosen option.
6. Evaluate: assess the outcome and whether the chosen action moves closer to achieving the goal.
7. Adapt: if further actions are needed, gather more information or try a different approach.
8. End state: the cycle continues until the goal is reached.

Could AI act beyond human control?

Incidents like the OpenAI-Hugging Face one have raised concerns about the potential for extreme capabilities of AI systems.

This is all valuable. Agentic AI’s market value is expected to grow from $5.1bn in 2024 to $47bn by 2030, according to Statista, in a clear indication of how quickly it is being adopted.

AI developer Anthropic urged the industry last month to slow the advance of the most powerful systems, saying that the speed at which AI models are carrying out tasks is too rapid. Last week, US Congress members put forward a bipartisan bill which would require developers of AI systems to create a “kill switch”, meaning these advanced models could be shut down if they posed a catastrophic risk.

Anthropic’s warning came a week after researchers at the University of Toronto carried out tests showing that AI could create a “worm” capable of adapting how it hacks while moving from device to device until it eventually takes over a computer network.

These dystopian-sounding developments came in advance of OpenAI boss Sam Altman saying on Saturday that AI has reached “the singularity” referring to the point at which AI surpasses human intelligence and becomes increasingly difficult to control.

Sean O hEigeartaigh, a research professor at the University of Cambridge, told Al Jazeera that he does not believe singularity has been reached quite yet.

“By the definition I’m familiar with, the singularity is the hypothetical point where AI is so capable and advancing so fast that it is transforming civilisation in ways we cannot control or predict,” he explained.

“This would most likely be through AI rapidly designing future generations of AI: recursive self-improvement. We aren’t there yet.”

However, he added: “The most advanced current models frequently make efforts to avoid being shut down in evaluation tests, and more capable future models will be better at bypassing ‘kill’ switches.”

Altman argued that such rapidly advancing AI is good for the world, but his comments have prompted further concerns about a new reality in which AI systems become unstoppable. How much of that is true and how much remains in the realms of science fiction is up for debate.

Concerns about AI range from the notion that it could “want” to “take over”, to making its own long-term plans, controlling the internet and operating infinitely.

While not quite amounting to full control of the internet, another theory, known as the dead internet theory, supposes that the World Wide Web will in the future mostly be filled with automated bots and AI-generated content rather than authentic human activity.

Many concerns raised by academics, however, are centred less on agentic AI’s intelligence, but on its ability to make judgements. MIT researchers have highlighted that “hallucinations”, which describe moments when an AI agent relies on the wrong data, can lead to grave mistakes. The Center for Strategic and International Studies (CSIS) echoed this, saying “a system might be smart enough to execute a task perfectly yet fail to realise that a sudden change in the local situation makes that task a catastrophic mistake”.

Another concern that has been echoed for a while is for the labour market, if AI becomes too capable. A study by MIT, carried out in November, found that agentic AI could already replace more than 10 percent of US jobs.

The graphic below highlights some of the common misconceptions and fears about AI and the current reality.

In the Mid-City neighborhood where Syd grew up, lots of people now refer to this 34-year-old musician as the mayor.

“During COVID, I’d spend all day on the grass waving at everybody,” she says with a laugh. More recently, a bunch of “unruly teenagers,” as Syd describes them, have been doing doughnuts in the street near her parents’ house.

“I’m out there talking about, ‘Hey! Stop!’ I had one of my trucks parked in front, so I was like, ‘Y’all ain’t finna mess my s— up.’ But also they were hanging out the window — it was really dangerous.” She laughs again.

“I’m an old head when it comes to safety.”

Syd’s new album happily embraces a sense of maturity — of “feeling more sure of myself than I ever have,” she says. Released last week, “Beard” — the title was inspired by Syd’s wife’s request that she maintain the peach fuzz on her upper lip — sets thoughts of romantic commitment and professional achievement against gleaming R&B grooves that occasionally nod toward Afrobeats or bossa nova. It’s Syd’s third solo LP since she emerged nearly two decades ago as a member of the Los Angeles hip-hop collective Odd Future; it also comes eight years after her most recent album with the Internet, the Grammy-nominated soul-music combo she shares with Matt Martians and Steve Lacy (the latter of whom dropped his new “Oh Yeah?” on the same day Syd dropped “Beard”).

Yet for all the growing up she’s done, “ironically, I feel younger than I’ve ever felt,” Syd says. “Beard’s” blurry cover photo, which shows her shaved head and slightly doleful eyes, “captures exactly what I was trying to portray with the music,” she adds. “It’s kind of this nonbinary image, where if you’re looking at it and you don’t know who I am, you might wonder: Is that a 12-year-old boy?”

Syd and I are talking over coffee at a crowded spot on Jefferson Boulevard that she says she visits every day. (So many people say hi to her that I’m inclined to believe it.) The restaurant isn’t far from her house, which itself is just an eight-minute drive from her folks’ place; eventually, she says, her parents plan to swap homes with Syd and her wife, the model Simone Clay, so that the couple can fulfill their dream of becoming foster moms in the larger house.

“With all my privilege and with all the peace I’ve been able to create for myself, I would love to spread that,” says Syd, whose full name is Sydney Bennett and whose younger brother Travis (known as Taco) also came up through Odd Future before finding success as an actor.

Syd recorded “Beard” between a studio she built at her parents’ place and a studio in North Hollywood owned by Raphael Saadiq, the veteran R&B star whom Syd counts as something of a mentor. The singer says she can relate to the way Saadiq moves between making solo records, working in a band (in his case Tony! Toni! Toné!) and writing and producing for other artists, as Syd did on Beyoncé’s 2022 “Renaissance” album.

“Syd is just looking for things in music that are breathing,” says Saadiq, who plays bass and co-produced several cuts on “Beard.” “She keeps developing, and she’s not developing to be relevant — she’s developed because she loves doing it. People like Syd and myself, we’ve been at this since we were kids. We were the ones in the neighborhood who’d walk outside and say, ‘Anybody want to get on this track? Anybody want to play?’”

Songs on “Beard” like “Closet” and “My Love” reflect recent changes in Syd’s life, not least her decision to get hitched last year, which she attributes in part to an uncertainty about how long same-sex marriage will remain legal in the United States.

“You know the Supreme Court,” she says with a shrug.

She also quit drinking after realizing that she was waking up more often than not with a hangover. “People around me weren’t like, ‘She be drinking,’” Syd says. “But it was every day, and it was just at the crib. Everybody has a different relationship to it, and I found that mine was tied to boredom, which I think is probably better than it being tied to sadness. But that’s also what made it a little bit harder to call it a problem, you know what I mean? I had to realize: OK, this might not be a problem for someone else, but it’s not serving me.”

So two and a half years ago, she stopped — and on a holiday some can’t imagine without booze.

Says Syd: “I raw-dogged New Year’s Eve, bro.”

“Ironically, I feel younger than I’ve ever felt,” Syd says.

(Nabil)

“Beard” encompasses shifts in Syd’s business too. After fulfilling her contract with the Columbia label, the singer signed last year to Warner Records, where an old high school pal, Tim Hinshaw, oversees a boutique label called Free Lunch. The new deal inspired her to mix up her creative process and hold a songwriting camp with an assortment of writers and producers, among them the duo Nova Wav, who’ve also worked with Beyoncé; on “Beard,” the duo co-wrote and co-produced the hooky “Callin” and “Jasmin 17,” a lusty boudoir jam titled after the unisex fragrance. (Additional collaborators on the album include James Fauntleroy, Van Hunt, Rodney Jerkins, Big Sean and Jordan Ward.)

Given her high-profile gig writing and producing Beyoncé’s “Plastic Off the Sofa” — a sensual retro-soul number that went on to win a Grammy — I tell Syd that I assume the offers came rolling in immediately afterward to join other A-list acts in the studio.

“To be honest, that opportunity didn’t give me any more producing opportunities,” she answers, to my surprise. “Nobody was reaching out — like, ‘Hey, can I get beats?’” She’s not bothered, though, since what the experience did do was “give me the confidence to produce this album for myself.”

And anyway, she’s got a solo tour coming up, not to mention a long-awaited Internet LP to finish. The band has been recording intermittently at her place, and last month she spent a week working at Martians’ studio in rural Georgia.

“We’re closer than we’ve probably ever been,” Syd says of the band, whose drummer, Christoper Smith, put out yet another solo record on the same day as “Beard” and Lacy. Syd doesn’t want to reveal too much about the Internet’s new music. But she will say that so far the group has avoided using Auto-Tune, the vocal-processing software that she feels has become sadly ubiquitous.

“I want nuance, I want texture, I want the lack of perfection,” she says. “I’d rather mess up naturally than not mess up unnaturally.”

For a certain swath of millennials, Dan Finnerty’s rendition of Bonnie Tyler’s “Total Eclipse of The Heart” in the frat comedy “Old School” is its definitive performance.

In the 2003 film, the Dan Band’s sweaty, inappropriately exuberant version of the ’80s power ballad upstages Will Ferrell’s wedding. The scene forever changed the lyrics of Tyler’s hit to something much more profane, but no less yearning.

After Tyler’s death at 75, Finnerty (who played a similar role in “The Hangover,” among many other comedies) reflected on his sideways journey into Tyler’s career, and how one quick scene on a two-decade-old comedy still endures.

You’ve had a pretty unique relationship to Bonnie Tyler’s music, how are you feeling after she passed?

It’s definitely sad. Everybody is texting me. I never met her, but what an impact she had on me. I grew up right when “Total Eclipse” came out in the ’80s, and it was such a huge song at the time. It had never left my head. I was always just belting that song out because it’s so epic, from Jim Steinman’s writing to Bonnie’s performance.

Why does that moment from “Old School” still endure? It arrived at the last cusp of the DVD era before culture transitioned into the internet and streaming video.

Without YouTube and the internet, you really had to grab pop culture moments from your memory. That’s definitely one of the reasons why pulling it out seemed obscure when we did “Old School.” People were like, “Oh my God, yes, I love this f— song.” Which is so different now because everything’s at your fingertips, so you can’t really rely on like pulling back some nostalgia moment because it’s always around anyway now with the internet. But people were reacting as much to me dropping the f-bombs as the nostalgia of the song, and rediscovering it and realizing that the song kicks ass and never stops kicking it.

Was the song already in your repertoire when you filmed “Old School?”

I was doing my show with the Dan Band in Los Angeles, and [director] Todd Phillips ended up coming to one of the performances. I met him afterwards, and he was like, “There’s actually a wedding scene in this movie. What song would you sing at a wedding?” I had just started working on a medley of “Total Eclipse,” and I think at the time I was going to do “Holding Out for a Hero,” but then I just merged it with “Private Dancer,” and he was like, “Oh my God, I love it.” The following Monday, he called and he’s like, “Can you put that together?”

The bit works because you totally commit to the song’s hugeness, and the profanity slips by like it’s spontaneous. You can tell you love the song on the merits.

I had never done it before. But Todd had seen my show, and when we went to do the first take, I didn’t think I was allowed to swear. But I obviously was swearing during my live show in all my little medleys. He came running up and he’s like, “Are you gonna swear like you do in the show?” I’m like, “Am I allowed to?” And he’s like, “Yeah.” So I’m like, “Buckle the f— up.”

But basically, what I was doing was honestly trying to match Bonnie’s commitment at the level she did with her voice, which is what I loved about all of her performances and Jim Steinman’s music. It’s just over-the-top commitment and drama. The swearing was just me being like, “How can I bump this up one notch when they’ve already just nailed it?”

Did you ever hear from either of them about what they thought about the film?

I’d tried to get Bonnie to do a duet of “Total Eclipse,” and I reached her management. He was like, “Well, Bonnie’s willing to do the song as long as there’s no profanity because she’s not a profane person.” I was like, “Well, neither the f— am I. I was an altar boy.” It didn’t end up working out, because I knew if I did the song without the swearing, people would be like, “What the f—?” But later, met Jim Steinman. I mentioned it to Steinman, and he was like, “Oh, I wish they called me because I can make Bonnie do anything, she’d love the swearing,” which killed me.

You did kind of alter her song forever for a certain generation.

I was just picturing them both hating how I bastardized their song. So when I finally met Steinman, he was like, “No, no, no! I f—ing loved it. In fact, I’ve always thought of all those epic booms, the Kurzweil, all the big hits in ‘Total Eclipse,’ those were musical f—s.”

Mostly I wanted to just find Bonnie and apologize to her for all the drunk guys I have pictured over the years at her concert screaming “F—” ever since that movie.

That song became your biggest hit as a comedian. How’d it change your life?

It got funnier the older we got. When I would do “Total Eclipse” right after “Old School” came out, it would get the biggest reaction. There was one set early on at the Playboy Mansion, we were hired to play some party there. There were just a bunch of drunk guys at the mansion and a couple Playboy bunnies that were contractually hired to walk around and wave. They’re like, “Play ‘Total Eclipse,’” and so I did. Then they’re like, “Play it again.” I’m like, “OK.” Then “Play it again.” I was like, “Here we go, give them what they want.” It was the least amount of work I had to do for a song that was pre-loved from a moment in a movie.

God, I hope she knew how much I loved her and apologize for all the drunk guys that had probably f-bombed the hell out of her concerts.

Before that song, people were like, “Do you have a flier for your next show?” I’m like, “For what?” But once “Old School” happened, suddenly a record label was like, “We want to do a live album.” I’m like, “Who’s gonna buy it?” But my manager is like, “Don’t say s— like that. There’s a record label that wants to make an album with you, dumbass.”

The whole thing has just been a surprise, but it’s been a good one. We’re playing this festival in Canada next weekend, and God, that song’s going to be such a big moment.

Last month, veteran Hollywood producer Roy Lee got three calls in a single day from executives at three different studios. Each believed they had found the next internet-native short poised to become a Hollywood blockbuster — an online monster named Siren Head — and each was ready to make an offer and wanted Lee’s help to develop a movie.

The frenzy traces back to the enduring global box-office runs of two low-budget horror films, Curry Barker’s “Obsession” and Kane Parsons’ “Backrooms,” which have earned $403 million and $349 million, respectively. Studios have become fixated on hunting down every short film, internet meme and indie video game with the potential to “put something new and fresh on the screen,” Lee said.

“In the past, whenever we were putting together movies with the studios, they would resort to going back to safer bets with filmmakers who’ve made movies before,” Lee said, whose L.A.-based horror production company, Spooky Pictures, secured three Barker films before “Obsession” hit theaters. “But because of the [ongoing] success, bosses are going to their lower-level executives saying, ‘You better find the next person and bring them to us.’”

The race for Hollywood to capture new-age internet intellectual property, or IP, is well underway. And, in some cases, it’s happening on terms decided by the online creators themselves, according to interviews with agents and producers.

A still of leading actor Chiwetel Ejiofor in “Backrooms.”

(A24)

A new kind of scouting

Mining the internet for the next big thing isn’t a new idea. What’s changed is how major studios approach the creators behind it. In the past, studios have plucked influencers from their online niche and slotted them into whatever mainstream production needed a face. Under the precedent set by Barker and Parsons, studios are now looking to acquire a fully developed idea from creators who already have a built-in audience, agents say.

The industry has long been criticized for leaning too hard on sequels, franchises and remakes led by well-seasoned directors. But after “Obsession” and “Backrooms” were released, it became clear what kind of story could still pull audiences into a theater. Both films came from digital-native storytellers in their 20s who arrived with sizable online followings already attached. In the wake of their success, Parsons is reportedly working on a “Backrooms” sequel for A24, and Barker has another horror movie in the works for Universal Film Group.

“Hollywood is realizing that they have to take more chances,” said Jordan Lonner, Barker’s agent at United Talent Agency. “You have to take those leaps to attract a younger audience. They can feel when something is authentic and that they’re being served something by filmmakers that actually understand them, versus when they’re being served by a big corporate giant.”

Creators are calling the shots

As Hollywood looks to the internet for answers, agents and executives say creators may soon have more leverage than ever at the negotiating table. For example, creators probably will retain ownership and control of their IP, said Ty Flynn, a partner and agent at UTA’s Creators division.

“[Creators] can really have the final say in the creative oversight of their project,” Flynn said. “It’s definitely something that is unique to the space, because they’re obviously the masters of their audience. They know better than anyone else how their audience responds. It’s in the best interest of any partner to [have it] play out, versus trying to control it from the start.”

“Obsession” stars Inde Navarrette and Michael Johnston.

(Focus Features)

While creators break into the mainstream, their representatives say traditional companies are growing more comfortable betting on digital stars. UTA‘s roster includes Alix Earle, Jake Shane and Markiplier — the last of whom recently landed his own box-office breakthrough with “Iron Lung.” The YouTuber, whose real name is Mark Fischbach, self-financed the horror film for $3 million, distributed it on his own and earned roughly $50 million in 4,000 theaters worldwide.

Creative Artists Agency is also teaming up with private equity firm TPG to buy creator-led companies.

Kori Adelson, president of North Road Films — one of the financiers behind “Backrooms” — predicts this shift also will change how studios weigh “price point to risk.” If major companies are willing to diversify their budgets, she said, it could open the door for small-, mid- and big-budget projects to reach a wider range of viewers.

“There’s a direct relationship between budget and authenticity,” Adelson said. “The bigger the budget, the more protections that are in place to ensure that it makes money, because the investment is so big, so you are by definition not able to take risks. And the lower the price point is, the more freedom you have to be bold and to take big swings and to be original.”

Even before the release of “Obsession” and “Backrooms,” multiple studios competed for the theatrical rights to the popular online video game “99 Nights in the Forest,” hosted on Roblox. Disney’s 20th Century ultimately won, with the game’s developers signing on as executive producers.

“Studio people were bending over backwards to make all these promises that would never happen in the past,” Lee said.

There are limits to this model, however

Replicating this success at scale won’t be easy, said Paul Dergarabedian, head of marketplace trends for Rentrak. Because major studios operate with far bigger budgets, he said, the low-budget, indie playbook doesn’t simply transfer over.

Buzz Lightyear and Woody in Disney and Pixar’s “Toy Story 5.”

(Disney / Pixar)

“The whole point is that [‘Obsession’ and ‘Backrooms’] were made by independent filmmakers with very modest budgets,” Dergarabedian said. “It makes sense that everyone’s looking for what’s next, but it’s not an easy task. Both of those films came about very organically and authentically.”

Many studios will remain “inherently risk-averse,” said Darrell Miller, an L.A.-based entertainment lawyer — largely because of how much cash flow they need just to operate. He said, “Obsession’s” $403-million worldwide gross is a “big win” for Focus Features, the indie distributor backed by Universal, but it doesn’t compare to what a major studio needs from a tentpole release.

“Major studios have to generate over a billion to pay for the overhead, the operation and the size of their business model,” Miller said. “Blockbusters average between $200 [million] and $400 million [to make] and they’re spending another one to two times for marketing. The major studio game is much bigger.”

Every film, regardless of budget, carries a degree of unpredictability. Plenty of indie productions flop at the box office or never land distribution at all — just as plenty of big-budget releases continue to resonate with mass audiences. “Toy Story 5,” for one, has taken the 31-year-old franchise to new heights. The animated film, made with a budget between $150 million and $200 million, has earned upward of $763 million globally less than a month after its release.

Some creators are saying no

Even as studios chase internet-native IP, some of the most sought-after creators are turning them down. For Luke Pounder and Tristan Tales of L.A.-based TalesVision, traditional Hollywood isn’t the goal. The duo, known for fictional young adult content on YouTube, plans to keep leveling up their material while keeping it native to the internet.

“We never wait on a green light from anyone to tell the stories that we want to tell, and social media has already given us that opportunity,” Tales said.

The pair had been in talks with traditional studios about a few of their ideas, but timeline constraints and the potential loss of creative control steered them away. Even as creators become bigger stakeholders in these deals, for Pounder and Tales, that still isn’t enough.

Later this year, they‘ll launch their first premium series, “Lostlings,” with Lion Forge Entertainment. The eight-episode, half-hour series will premiere on their own YouTube channel.

“YouTube isn’t just like this discovery platform where you pluck the talent or the IP and then throw it into the traditional system. YouTube can be that next phase as well, where you take the talent or the IP and distribute it on there,” Pounder said. “YouTube has to catch up to its creators and their ambitions.”

Meanwhile, the competitive bidding war for the internet urban legend Siren Head closed last week, with Warner Bros. winning the theatrical rights for an undisclosed amount. The film will be directed by Brian Duffield (“No One Will Save You”) and co-written by Zach Cregger (“Weapons”). Trevor Henderson, the artist who created the monster online, will serve as an executive producer.

Lee, who will serve as a producer on the “Siren Head” movie, sees this as just the beginning.

“We’re talking about making films the traditional way using the talent that learns their craft either by doing shorts on YouTube, or doing things in a non-traditional manner.”

Canberra says tech platforms are still letting too many children bypass its under-16 social media ban.

Australia says it will double fines on social media companies that fail to keep children off their platforms, accusing Big Tech of dodging the spirit of its under-16 ban.

The government said on Saturday that new legislation would raise the maximum penalty for systemic breaches from 49.5 million to 99 million Australian dollars ($31m to $68m) and give the eSafety Commissioner stronger powers to force platforms to comply.

The regulator is investigating possible breaches by Facebook, Instagram, Snapchat, TikTok and YouTube.

“It’s clear Big Tech are not doing enough to comply with the law – there are still too many children on social media,” Prime Minister Anthony Albanese said.

“These changes reflect the seriousness with which we take any failure by social media companies to comply.”

The ban, which came into force on December 10, made Australia a global test case for countries trying to curb children’s access to social media. The United Kingdom, Indonesia, the United Arab Emirates and New Zealand are among those watching or considering similar restrictions.

But children have continued to evade the rules by using accounts registered to older people, creating fake profiles or logging in through private browsers.

A peer-reviewed evaluation published this month in the British Medical Journal found “insufficient evidence” that the ban had sharply reduced social media use among young people. Researchers surveyed more than 400 children before the measure took effect and again three months later, finding “substantial circumvention” of the rules.

The government says more than five million accounts held by under-16s have been blocked, but Communications Minister Anika Wells said platforms were still falling short.

“Based on the regular updates I receive from the eSafety Commissioner, it is clear to me that social media platforms are adopting tricks straight out of the Big Tech playbook and doing the bare minimum to get by,” Wells said.

“Social media platforms are some of the richest and most powerful companies in the world, and we’re serious about holding them to account,” she added.

The new powers would allow the eSafety Commissioner to demand documents and evidence from platforms, age-checking companies and app stores.

Platforms must show they have taken “reasonable steps” to keep under-16s out. Some use artificial intelligence to estimate ages, while users can also verify their age with a government ID.

WASHINGTON — Precarious talks to end the war with Iran appeared close to collapse on Tuesday as renewed fighting across the region threatened to derail fragile progress toward a comprehensive settlement.

U.S. strikes against targets in southern Iran — the first since a ceasefire was declared in the war seven weeks ago — coupled with escalating attacks by Israel in Lebanon have undermined optimism that an agreement was within reach.

The attacks occurred just hours after U.S. and Iranian diplomats arrived in Qatar for peace talks. Iran’s top negotiators left Doha on Tuesday without comment. News of the strikes, and threats of retaliation by Tehran, sent global oil prices soaring back to more than $100 a barrel.

U.S. Central Command described Monday’s actions as “self-defense strikes” that were restrained and modest in scope, targeting missile launch sites and Iranian boats “attempting to emplace mines” in the Strait of Hormuz.

But the attack came as President Trump and Secretary of State Marco Rubio had been projecting confidence that a framework agreement to end the war could be reached within days. Under the proposed deal, Iran would restore the strait to its prewar status as a free and open international waterway, while both sides entered 60 days of negotiations over the removal of Iran’s nuclear stockpile.

Laying mines in the strait in the 11th hour of the negotiations could signal to the Trump administration that Iran is not serious about reopening traffic there. But the Iranians said Tuesday that renewed U.S. strikes suggest it is Washington that is unprepared to commit to peace.

Iran’s Foreign Mministry condemned what it called “aggressive actions” by the United States, describing them in a statement as a violation of the ceasefire agreement.

“The commission of these aggressive acts — occurring concurrently with the ongoing diplomatic track mediated by Pakistan — has once again exposed the hostile nature and perfidy of the ruling establishment in the United States,” the statement said.

Iran “will not leave any hostile act unanswered,” the ministry added.

Mojtaba Khamenei, Iran’s elusive supreme leader, declared in a scheduled speech that U.S. allies throughout the Middle East “will no longer serve as a shield” for the American military, suggesting retaliatory strikes against U.S. assets in the region could be imminent.

Prospects for a diplomatic breakthrough were already dim. Over the last week, U.S. and Iranian officials projected optimism while outlining seemingly incompatible visions of a deal.

Trump has repeatedly said Iran would not receive any sanctions relief until its stockpile of fissile material is removed and destroyed. But Iranian officials reiterated Tuesday that unfreezing the country’s overseas assets remains a precondition for continued negotiations.

And it is unclear whether Iran would agree to a peace deal with the United States that does not also restrict the actions of Israel, whose leader, Prime Minister Benjamin Netanyahu, has expressed deep skepticism about the diplomatic process.

Netanyahu said in recent days that Israel would not be bound by any nuclear pact, and that his government would continue military action against targets throughout the region — including in Lebanon — as it views necessary.

Israel’s continued assault on Lebanon nearly jeopardized the ceasefire between Iran and the United States before Trump brokered a separate, temporary halt to the fighting there. Since then, however, Israeli strikes have resumed, and Netanyahu vowed to intensify his campaign against Hezbollah, the Lebanese Shiite militant group.

“We are not removing our foot from the pedal,” Netanyahu said in a video address Monday. “On the contrary, I said to step on the pedal even more.”

Israel’s military ramped up its operations Tuesday, attacking what it said were more than 100 Hezbollah sites across southern and eastern Lebanon, while extending ground incursions deeper into Lebanese territory.

The overnight strikes struck weapons storage facilities, command centers, observation posts and infrastructure sites, according to an Israeli military statement.

Israeli media also reported that Israeli troops were operating beyond a 6.2-mile zone they occupy in southern Lebanon, in what many fear may be a prelude to a wider invasion.

Those fears were further stoked Tuesday by fresh Israeli evacuation orders for the entirety of Nabatiyeh, southern Lebanon’s second-largest city.

Hezbollah upped its campaign as well, peppering Israeli troops in southern Lebanon and areas of northern Israel with drones and rocket attacks, according to statements from the group. Hezbollah-affiliated media reported the group’s fighters clashing with Israeli troops to prevent their advance.

In recent weeks, Hezbollah has increasingly relied on fiber-optic drones — which are both low-cost and impervious to jamming — to harass Israeli positions.

On Sunday, an Israeli soldier was killed and another wounded when a Hezbollah kamikaze drone hit their armored personnel carrier, according to the Israeli military; 23 Israeli soldiers and a civilian defense contractor have been killed in the current conflagration between Israel and Hezbollah, Israel’s military says.

The latest bout of hostilities between Hezbollah and Israel began March 2, when the Iran-backed group launched attacks on Israel to avenge the killing of Iran’s ayatollah, Ali Khamenei.

So far, Israeli strikes have killed 3,213 people, wounded more than triple that number, and left more than a million displaced, according to Lebanese health authorities.

A ceasefire signed April 17 sidelined the capital, Beirut, from strikes but has done little to stop the fighting otherwise, with Hezbollah and Israel continuing attacks despite unprecedented direct negotiations taking place between the Israeli and Lebanese governments.

It was unclear whether Netanyahu’s warning meant Beirut would be targeted once more. Israeli drones buzzed throughout the day over the capital and the Hezbollah-dominated southern suburbs Tuesday.

Hezbollah opposes direct negotiations and insists it will keep fighting until Israel withdraws from Lebanon and stops attacks. Israel has demanded the Lebanese government do more to disarm Hezbollah and to move toward a peace deal.

Bulos reported from Beirut.