×
How the ‘Demon Slayer’ Studio Broke Laws of Anime Physics to Bring ‘Infinity Castle’ Fights to Life

How the ‘Demon Slayer’ Studio Broke Laws of Anime Physics to Bring ‘Infinity Castle’ Fights to Life

As io9 has thoroughly glazed in detail, 2025 was a definitive year for anime films, thanks in part to the meteoric box-office success of Demon Slayer: Infinity Castle. As the film prepares for award season, its directors shared some behind-the-scenes tidbits about how studio Ufotable pushed past their limits (and the laws of physics) to bring the megapopular anime’s bombastic fights to life.

Speaking with Entertainment Weekly, Infinity Castle director Haruo Sotozaki, director of photography Yuichi Terao, and Ufotable president Hikaru Kondo reminisced about how the studio’s commitment to making the best anime possible contributed to the film being at the table this award season.

In the movie, hero Tanjiro Kamado and the rest of the Demon Slayer Corps dive straight into enemy territory, battling an onslaught of powerful enemies and overpowered bosses in the ever-shifting maze that is the Infinity Castle. While this isn’t Demon Slayer‘s first rodeo in cinemas, Infinity Castle marks the beginning of the end of the banner series as the first of a film trilogy. Having the first leg of that trilogy play in IMAX theaters inspired Ufotable to elevate the streamer anime into a cinematic experience. The studio wanted to ensure that every corner of the screen was popping with detail, which required collaboration in its compositing of 2D characters moving through 3D environments in its fluid action sequences.

“If you look at a lot of the keyframes, the key animation, the sakuga in this film, we worked really closely with our 3D department to make sure everyone’s in full lockstep because we didn’t want any weird bizarre effects where perhaps these 2D characters are sliding around on the 3D surroundings and environments,” Sotozaki told EW. “So it was very important that the 2D keyframe animators were talking to our 3D team to make sure the compositions were perfect.”

On that front, Sotozaki told EW that the most complicated aspect of bringing Infinity Castle to life was ensuring that its 3D environments and animations were all in sync. To prevent oopsies during production, he said Ufotable would often test keyframes in their respective 3D environments before finishing key animations. While hero Tanjiro’s secret to success is his impeccable breathing techniques, Terao explained that Ufotable’s animators would hype themselves up with music so that, once pencil hit paper, Infinity Castle‘s many set pieces and sequencing momentum would pour straight into its animation.

“Whenever you see Tanjiro saying, ‘Don’t give up,’ the animators are probably doing something similar with their headphones on,” Terao said. “I’m on the digital 3D side of things, but I know this is especially true for the keyframe animation team, the 2D animators. Whenever they want the characters to perform certain actions, they would sometimes get up from their desk and actually act out some of these action sequences. So if you catch this out of the corner of your eye, sometimes you might wonder, ‘Are they just cheering themselves on or dancing or are they trying to act something out to see what the animation would look like?’ It’s kind of an interesting sight to see.”

By far the biggest hurdle for the team was Infinity Castle‘s final fight between Tanjiro and Akaza, which Sotozaki said was led by four people at Ufotable. Key among them was key animator Masayuki Kunihiro, who Sotozaki explained handled 80 percent of Akaza and Tomioka’s physics-breaking fight.

“It’s quite tricky action, but very fast action. I mean this in a good way, but it’s unrealistic in the sense of how fast the characters can move and what they look like on screen. It’s very convincing to the viewer, but it defies physics,” Terao said. “If you’ve seen the film, you will understand what we’re talking about. The first time we’re trying different types of animation where parts of the Infinity Castle either shatter, breaks, or collapses, there’s a lot of destruction. And take, for example, when Akaza smashes the ground. It shatters and you have this smoke, but all of this needs to be simulated in CG in these different physics engines.”

“The irony is, when you look at Kunihiro’s animation, you’ll flip one frame of paper and the characters will move like 100 meters, which in real-world physics is impossible. So a lot of the simulation that we tried to run on how things shatter, making that hybrid between the 2D animation and the 3D simulation was quite challenging,” Sotozaki said, getting more into the nitty-gritty of how the anime sausage was made.

He continued: “It was fun, but challenging because there really are no settings that allow such fast physics to happen by how the physics engines are computing the different simulations. So I think he has a very unique pipeline, too. Oftentimes we’ll have 2D action animators match the action to the 3D environments or the shot composition, but Kunihiro is the exception to this and we allow him to make the action keyframes first and the 3D team actually has to match the 3D environments to whatever he gives.”

© Ufotable

The trio unfortunately said they couldn’t provide an update on when fans can expect the second installment of Demon Slayer‘s film trilogy finale or what the future of Demon Slayer might look like once the trilogy concludes. This response, in concert with the Crunchyroll president’s FOMO-fueled “better watch it in theaters while you can” response to anime fans asking when they can expect the film to arrive on the streamer, is pretty disheartening.

However, Sotozaki assured fans that Ufotable will continue striving to deliver a consistent level of animation quality for the final legs of Demon Slayer‘s theatrical series finale.

“At the studio, we always try to meet and exceed audience expectations and create an experience that everyone can enjoy and the entire team and staff share this sentiment, so do with that information what you will,” Sotozaki said.

In io9’s review of Infinity Castle, we too lauded the film for its flashy animation and action sequencing, writing, “Ufotable’s animation is as dazzling as ever, with slick camera movement that has its characters ping-ponging through 3D CG backgrounds, complemented by compositing finesse that gives the film a cinematic grandeur exceeding that of past films like Mugen Train. Its action choreography is a rollercoaster for the eyes—fluid, explosive, and occasionally slowed down, as if to let sparks fly off katana blades like fireworks behind your eyes. And its final battle kicks things into third gear.”

Clearly, Ufotable’s reputation as the anime studio to watch has borne fruit this year, because Infinity Castle got a Golden Globe nomination for Best Animated Feature Film. Chainsaw Man – The Movie: Reze Arc, despite being equally excellent, didn’t make the cut, so all Demon Slayer has to do is beat out its fellow nominees (including heavy-hitters KPop Demon Hunters and Zootopia 2.) Like Issa Rae at the Emmys, we’re rooting for everyone anime.

Want more io9 news? Check out when to expect the latest Marvel, Star Wars, and Star Trek releases, what’s next for the DC Universe on film and TV, and everything you need to know about the future of Doctor Who.

Source link
#Demon #Slayer #Studio #Broke #Laws #Anime #Physics #Bring #Infinity #Castle #Fights #Life

According to the Wall Street Journal, the export control directive that led to Anthropic cutting off access to Fable 5 and Mythos 5 was triggered in part by cybersecurity research from Amazon and conversations between CEO Andy Jassy and the White House. According to the report, the paper from Amazon claims that, through a series of prompts, it was able to get Fable 5 to serve up information that could be used in cyberattacks. Amazon has yet to respond to a request for comment.

Shortly after Jassy shared the company’s findings with the government, it made the call to block its use by foreign nationals. Complicating this issue is that many of Anthropic’s researchers are foreign-born, meaning they were barred from accessing their own product.

In a statement, Anthropic disputed the government’s characterization of the issue as a “jailbreak.” It argued that many of the same vulnerabilities could be discovered using other publicly available models, including GPT 5.5. Some security researchers appear to back the company’s interpretation. Katie Moussouris, the founder and CEO of LutaSecurity posted on BlueSky that “I’ve seen the paper. It’s not a jailbreak.” Former Commerce Department official Kate Koren speculated to the WSJ that the White House’s dislike of Anthropic may have influenced the decision.

Anthropic and the Trump administration have been at odds for some time over the company’s refusal to allow its AI to be used for mass surveillance of Americans or to power lethal autonomous weapons. In February, Trump instructed federal agencies to stop using Anthropic’s AI. And just hours later, Secretary of Defense Pete Hegseth designated the company a supply chain risk.

The government and the company seemed to have made amends, and the two had worked together to expand access to Mythos. However, now the two seem destined to clash again.

#Amazon #security #research #reportedly #led #White #Houses #Anthropic #Fable #banAI,Amazon,Anthropic,News,Policy,Politics,Security,Tech">Amazon security research reportedly led to the White House’s Anthropic Fable banAccording to the Wall Street Journal, the export control directive that led to Anthropic cutting off access to Fable 5 and Mythos 5 was triggered in part by cybersecurity research from Amazon and conversations between CEO Andy Jassy and the White House. According to the report, the paper from Amazon claims that, through a series of prompts, it was able to get Fable 5 to serve up information that could be used in cyberattacks. Amazon has yet to respond to a request for comment.Shortly after Jassy shared the company’s findings with the government, it made the call to block its use by foreign nationals. Complicating this issue is that many of Anthropic’s researchers are foreign-born, meaning they were barred from accessing their own product.In a statement, Anthropic disputed the government’s characterization of the issue as a “jailbreak.” It argued that many of the same vulnerabilities could be discovered using other publicly available models, including GPT 5.5. Some security researchers appear to back the company’s interpretation. Katie Moussouris, the founder and CEO of LutaSecurity posted on BlueSky that “I’ve seen the paper. It’s not a jailbreak.” Former Commerce Department official Kate Koren speculated to the WSJ that the White House’s dislike of Anthropic may have influenced the decision.Anthropic and the Trump administration have been at odds for some time over the company’s refusal to allow its AI to be used for mass surveillance of Americans or to power lethal autonomous weapons. In February, Trump instructed federal agencies to stop using Anthropic’s AI. And just hours later, Secretary of Defense Pete Hegseth designated the company a supply chain risk.The government and the company seemed to have made amends, and the two had worked together to expand access to Mythos. However, now the two seem destined to clash again.#Amazon #security #research #reportedly #led #White #Houses #Anthropic #Fable #banAI,Amazon,Anthropic,News,Policy,Politics,Security,Tech

Wall Street Journal, the export control directive that led to Anthropic cutting off access to Fable 5 and Mythos 5 was triggered in part by cybersecurity research from Amazon and conversations between CEO Andy Jassy and the White House. According to the report, the paper from Amazon claims that, through a series of prompts, it was able to get Fable 5 to serve up information that could be used in cyberattacks. Amazon has yet to respond to a request for comment.

Shortly after Jassy shared the company’s findings with the government, it made the call to block its use by foreign nationals. Complicating this issue is that many of Anthropic’s researchers are foreign-born, meaning they were barred from accessing their own product.

In a statement, Anthropic disputed the government’s characterization of the issue as a “jailbreak.” It argued that many of the same vulnerabilities could be discovered using other publicly available models, including GPT 5.5. Some security researchers appear to back the company’s interpretation. Katie Moussouris, the founder and CEO of LutaSecurity posted on BlueSky that “I’ve seen the paper. It’s not a jailbreak.” Former Commerce Department official Kate Koren speculated to the WSJ that the White House’s dislike of Anthropic may have influenced the decision.

Anthropic and the Trump administration have been at odds for some time over the company’s refusal to allow its AI to be used for mass surveillance of Americans or to power lethal autonomous weapons. In February, Trump instructed federal agencies to stop using Anthropic’s AI. And just hours later, Secretary of Defense Pete Hegseth designated the company a supply chain risk.

The government and the company seemed to have made amends, and the two had worked together to expand access to Mythos. However, now the two seem destined to clash again.

#Amazon #security #research #reportedly #led #White #Houses #Anthropic #Fable #banAI,Amazon,Anthropic,News,Policy,Politics,Security,Tech">Amazon security research reportedly led to the White House’s Anthropic Fable ban

According to the Wall Street Journal, the export control directive that led to Anthropic cutting off access to Fable 5 and Mythos 5 was triggered in part by cybersecurity research from Amazon and conversations between CEO Andy Jassy and the White House. According to the report, the paper from Amazon claims that, through a series of prompts, it was able to get Fable 5 to serve up information that could be used in cyberattacks. Amazon has yet to respond to a request for comment.

Shortly after Jassy shared the company’s findings with the government, it made the call to block its use by foreign nationals. Complicating this issue is that many of Anthropic’s researchers are foreign-born, meaning they were barred from accessing their own product.

In a statement, Anthropic disputed the government’s characterization of the issue as a “jailbreak.” It argued that many of the same vulnerabilities could be discovered using other publicly available models, including GPT 5.5. Some security researchers appear to back the company’s interpretation. Katie Moussouris, the founder and CEO of LutaSecurity posted on BlueSky that “I’ve seen the paper. It’s not a jailbreak.” Former Commerce Department official Kate Koren speculated to the WSJ that the White House’s dislike of Anthropic may have influenced the decision.

Anthropic and the Trump administration have been at odds for some time over the company’s refusal to allow its AI to be used for mass surveillance of Americans or to power lethal autonomous weapons. In February, Trump instructed federal agencies to stop using Anthropic’s AI. And just hours later, Secretary of Defense Pete Hegseth designated the company a supply chain risk.

The government and the company seemed to have made amends, and the two had worked together to expand access to Mythos. However, now the two seem destined to clash again.

#Amazon #security #research #reportedly #led #White #Houses #Anthropic #Fable #banAI,Amazon,Anthropic,News,Policy,Politics,Security,Tech
Meta has begun dismantling its $2 billion acquisition of Manus, completing an operational separation from the Chinese-founded AI startup and halting data sharing between the two companies. This is the most concrete step yet toward complying with a divestiture order Beijing issued roughly two months ago on national security grounds.

Meta has cut Manus off from its internal systems, Bloomberg reported, preventing employees from using Manus tools for internal projects as the two companies move toward a full separation.

Meanwhile, according to May reports, the co-founders of Manus have held preliminary discussions about raising approximately $1 billion from outside investors to reclaim the startup from Meta, a move that could pave the way for a Chinese joint venture structure and an eventual listing in Hong Kong, a venue that has seen a surge in AI listings this year for Chinese AI startups like MiniMax and Zhipu.

What was supposed to be a landmark exit for Chinese AI is quickly unraveling. The move underscores Beijing’s determination to retain control over strategically sensitive technology, regardless of a company’s offshore incorporation.

In addition to the forced divestiture, Chinese authorities have since expanded travel restrictions to researchers and executives at private firms, requiring government approval before heading abroad. China is also tightening its grip on foreign capital, with reports indicating that top AI firms, including Moonshot AI, StepFun, and ByteDance, will need government sign-off before accepting U.S. investment, adding another layer to Beijing’s sweeping effort to control its AI sector.

Even as Meta moves to sever ties with Manus, the agentic AI startup has continued to ship new features, rolling out integrations with Similarweb and Shopify.

Manus drew widespread attention with a viral agent demo relocated its staff to Singapore in mid-2025 before announcing a $2 billion acquisition by Meta in December. Chinese regulators moved to scrutinize the transaction earlier this year, citing potential violations of technology export controls and foreign investment rules.

Manus investors, including California-based venture firm Benchmark, have already received their proceeds from the acquisition, while Asian backers, including Tencent, HSG, and ZhenFund, have indicated they will cooperate with the unwinding process, according to the WSJ.

Manus’ Chinese origins with parent company Butterfly Effect drew scrutiny on both sides of the Pacific, with Senator John Cornyn questioning whether American capital should flow to a Chinese-linked firm.

Meta and Manus did not immediately respond to a request for comment outside regular business hours.

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

#Meta #reportedly #moves #unwind #Manus #deal #Beijings #demand #TechCrunchChina,manus,Manus AI,Meta">Meta reportedly moves to unwind B Manus deal after Beijing’s demand | TechCrunch
Meta has begun dismantling its  billion acquisition of Manus, completing an operational separation from the Chinese-founded AI startup and halting data sharing between the two companies. This is the most concrete step yet toward complying with a divestiture order Beijing issued roughly two months ago on national security grounds.

Meta has cut Manus off from its internal systems, Bloomberg reported, preventing employees from using Manus tools for internal projects as the two companies move toward a full separation.







Meanwhile, according to May reports, the co-founders of Manus have held preliminary discussions about raising approximately  billion from outside investors to reclaim the startup from Meta, a move that could pave the way for a Chinese joint venture structure and an eventual listing in Hong Kong, a venue that has seen a surge in AI listings this year for Chinese AI startups like MiniMax and Zhipu.

What was supposed to be a landmark exit for Chinese AI is quickly unraveling. The move underscores Beijing’s determination to retain control over strategically sensitive technology, regardless of a company’s offshore incorporation. 

In addition to the forced divestiture, Chinese authorities have since expanded travel restrictions to researchers and executives at private firms, requiring government approval before heading abroad. China is also tightening its grip on foreign capital, with reports indicating that top AI firms, including Moonshot AI, StepFun, and ByteDance, will need government sign-off before accepting U.S. investment, adding another layer to Beijing’s sweeping effort to control its AI sector. 

Even as Meta moves to sever ties with Manus, the agentic AI startup has continued to ship new features, rolling out integrations with Similarweb and Shopify. 

Manus drew widespread attention with a viral agent demo relocated its staff to Singapore in mid-2025 before announcing a  billion acquisition by Meta in December. Chinese regulators moved to scrutinize the transaction earlier this year, citing potential violations of technology export controls and foreign investment rules.


Manus investors, including California-based venture firm Benchmark, have already received their proceeds from the acquisition, while Asian backers, including Tencent, HSG, and ZhenFund, have indicated they will cooperate with the unwinding process, according to the WSJ.

Manus’ Chinese origins with parent company Butterfly Effect drew scrutiny on both sides of the Pacific, with Senator John Cornyn questioning whether American capital should flow to a Chinese-linked firm.

Meta and Manus did not immediately respond to a request for comment outside regular business hours.
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.#Meta #reportedly #moves #unwind #Manus #deal #Beijings #demand #TechCrunchChina,manus,Manus AI,Meta

roughly two months ago on national security grounds.

Meta has cut Manus off from its internal systems, Bloomberg reported, preventing employees from using Manus tools for internal projects as the two companies move toward a full separation.

Meanwhile, according to May reports, the co-founders of Manus have held preliminary discussions about raising approximately $1 billion from outside investors to reclaim the startup from Meta, a move that could pave the way for a Chinese joint venture structure and an eventual listing in Hong Kong, a venue that has seen a surge in AI listings this year for Chinese AI startups like MiniMax and Zhipu.

What was supposed to be a landmark exit for Chinese AI is quickly unraveling. The move underscores Beijing’s determination to retain control over strategically sensitive technology, regardless of a company’s offshore incorporation.

In addition to the forced divestiture, Chinese authorities have since expanded travel restrictions to researchers and executives at private firms, requiring government approval before heading abroad. China is also tightening its grip on foreign capital, with reports indicating that top AI firms, including Moonshot AI, StepFun, and ByteDance, will need government sign-off before accepting U.S. investment, adding another layer to Beijing’s sweeping effort to control its AI sector.

Even as Meta moves to sever ties with Manus, the agentic AI startup has continued to ship new features, rolling out integrations with Similarweb and Shopify.

Manus drew widespread attention with a viral agent demo relocated its staff to Singapore in mid-2025 before announcing a $2 billion acquisition by Meta in December. Chinese regulators moved to scrutinize the transaction earlier this year, citing potential violations of technology export controls and foreign investment rules.

Manus investors, including California-based venture firm Benchmark, have already received their proceeds from the acquisition, while Asian backers, including Tencent, HSG, and ZhenFund, have indicated they will cooperate with the unwinding process, according to the WSJ.

Manus’ Chinese origins with parent company Butterfly Effect drew scrutiny on both sides of the Pacific, with Senator John Cornyn questioning whether American capital should flow to a Chinese-linked firm.

Meta and Manus did not immediately respond to a request for comment outside regular business hours.

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

#Meta #reportedly #moves #unwind #Manus #deal #Beijings #demand #TechCrunchChina,manus,Manus AI,Meta">Meta reportedly moves to unwind $2B Manus deal after Beijing’s demand | TechCrunch

Meta has begun dismantling its $2 billion acquisition of Manus, completing an operational separation from the Chinese-founded AI startup and halting data sharing between the two companies. This is the most concrete step yet toward complying with a divestiture order Beijing issued roughly two months ago on national security grounds.

Meta has cut Manus off from its internal systems, Bloomberg reported, preventing employees from using Manus tools for internal projects as the two companies move toward a full separation.

Meanwhile, according to May reports, the co-founders of Manus have held preliminary discussions about raising approximately $1 billion from outside investors to reclaim the startup from Meta, a move that could pave the way for a Chinese joint venture structure and an eventual listing in Hong Kong, a venue that has seen a surge in AI listings this year for Chinese AI startups like MiniMax and Zhipu.

What was supposed to be a landmark exit for Chinese AI is quickly unraveling. The move underscores Beijing’s determination to retain control over strategically sensitive technology, regardless of a company’s offshore incorporation.

In addition to the forced divestiture, Chinese authorities have since expanded travel restrictions to researchers and executives at private firms, requiring government approval before heading abroad. China is also tightening its grip on foreign capital, with reports indicating that top AI firms, including Moonshot AI, StepFun, and ByteDance, will need government sign-off before accepting U.S. investment, adding another layer to Beijing’s sweeping effort to control its AI sector.

Even as Meta moves to sever ties with Manus, the agentic AI startup has continued to ship new features, rolling out integrations with Similarweb and Shopify.

Manus drew widespread attention with a viral agent demo relocated its staff to Singapore in mid-2025 before announcing a $2 billion acquisition by Meta in December. Chinese regulators moved to scrutinize the transaction earlier this year, citing potential violations of technology export controls and foreign investment rules.

Manus investors, including California-based venture firm Benchmark, have already received their proceeds from the acquisition, while Asian backers, including Tencent, HSG, and ZhenFund, have indicated they will cooperate with the unwinding process, according to the WSJ.

Manus’ Chinese origins with parent company Butterfly Effect drew scrutiny on both sides of the Pacific, with Senator John Cornyn questioning whether American capital should flow to a Chinese-linked firm.

Meta and Manus did not immediately respond to a request for comment outside regular business hours.

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

#Meta #reportedly #moves #unwind #Manus #deal #Beijings #demand #TechCrunchChina,manus,Manus AI,Meta

Post Comment