Internet Cinematic launches India-Japan AI film studio, Yoroll’s AI Odyssey turns Homer’s epic into a playable film, Warner Bros. faces backlash over AI dog promotion
In this week’s Entertainment and AI Tech News — Yoroll’s AI Odyssey turns Homer’s epic into a playable, branching film built in two days on a ~$250 budget, while independent channel IncredibleXvision has crossed 10,000 YouTube subscribers with AI-generated shorts, suggesting synthetic content can build repeat audiences, not just one-off virality. On the studio side, Warner Bros. is facing significant backlash after promoting The End of Oak Street with an AI-generated dog podcast, a campaign many viewers say made them less likely to see the Anne Hathaway and Ewan McGregor thriller. AI is reshaping production and policy: Roku added a 24-hour AI channel via Fairground to its FAST lineup; Autodesk is rolling out 3D controls in Flow Studio for finer command over AI scenes; Brad Pitt argues AI could revive mid-budget films if used as a cost-saving tool rather than a performer replacement; Chinese models like MiniMax’s Hailuo 3 and ByteDance’s Seedream 5.0 are reigniting Hollywood’s copyright concerns; SAG-AFTRA is pushing back on weakening performer protections around digital replicas; the EU has begun enforcing transparency rules requiring clear labeling of AI-generated or significantly altered content; and Internet Cinematic’s new India–Japan AI film studio in Tokyo signals growing cross-border collaboration aimed at making ambitious filmmaking more accessible while keeping human artists central.
Yoroll’s AI Odyssey turns Homer’s epic into a playable film
Yoroll has launched AI Odyssey, an interactive film inspired by Homer’s The Odyssey that allows viewers to influence the story through branching choices. Created by filmmaker Amy Sun on Yoroll’s AI-native video platform, the project was reportedly completed in two days on a budget of approximately $250. It includes more than 140 AI-generated cinematic video segments, 11 possible endings, and new choices every 30 to 50 seconds. Yoroll said the experience attracted more than 10,000 plays across 30 countries during its first week, while the hashtag #AIOdysseyByYoroll briefly ranked among the top 10 trending topics on social media.
The project was created using Yoroll’s integrated workflow, which combines AI video generation, character consistency, dialogue and voice tools, branching narrative design, and gameplay systems. The company said the platform is intended to reduce the technical and financial barriers associated with interactive entertainment.
Warner Bros. faces backlash over AI dog promotion
Warner Bros. is facing online criticism over an AI-generated dog podcast created to promote The End of Oak Street, the upcoming sci-fi survival thriller starring Anne Hathaway and Ewan McGregor. The studio partnered with DogPack, a podcast project featuring two AI-generated canine hosts, Goldie and Frenchie. In the promotional video, the dogs interview an AI version of Starbuck, the dog featured in the film.
The clip was released across social media as Warner Bros. increased marketing ahead of the movie’s Aug. 14 theatrical release. The campaign quickly drew negative reactions from viewers who objected to the studio’s use of generative AI in a project connected to the film and creative industries. Some commenters said the promotion had reduced their interest in seeing the movie, while others described the campaign as an example of AI being used unnecessarily in entertainment marketing.
The End of Oak Street, written and directed by David Robert Mitchell, follows the Platt family after a mysterious cosmic event transports their suburban neighborhood to a dangerous environment populated by prehistoric creatures. J.J. Abrams is among the film’s producers.
Independent AI channel surpasses 10,000 YouTube subscribers
IncredibleXvision, an independent YouTube channel built around AI-generated short-form video, has surpassed 10,000 subscribers, marking a notable milestone for creator-led synthetic entertainment. The channel publishes short videos featuring fictional animals, mythical creatures, cyborgs, and futuristic environments. Much of its output centers on dogs and cats placed in situations that would be difficult or impossible to film traditionally, including imagined adventures and action sequences.
Although 10,000 subscribers is small compared with the audiences of major entertainment channels, the figure suggests sustained interest beyond a single viral video. Viewers are returning for a consistent style and recurring themes rather than simply sampling AI content as a novelty. That distinction is significant as the entertainment industry debates whether AI-generated video can attract loyal audiences. IncredibleXvision’s model relies on rapid production and familiar emotional subjects, particularly pets, combined with fantastical settings.
Roku adds 24-hour AI-generated channel to FAST lineup
Roku has added a 24-hour channel featuring AI-generated video content to its free, ad-supported streaming television lineup through a partnership with Fairground, a startup led by Colin Petrie-Norris. The channel is one of four new offerings added to Roku’s platform. Fairground’s service is the most significant experiment because it provides continuous programming created with generative AI.
The move reflects the economic pressures facing the FAST industry, where platforms must supply a steady stream of programming without charging viewers subscription fees. Traditional channels depend heavily on licensing existing shows and movies, while AI-generated content could potentially be produced more quickly and at lower cost.
Fairground appears to be targeting the lean-back viewing experience associated with traditional television, producing structured programming intended to run continuously rather than isolated viral clips. The approach could give Roku a way to expand its channel inventory while reducing dependence on expensive content deals.
Chinese AI models expose Hollywood’s copyright dilemma
AI-generated videos using characters from The Office and other major franchises are renewing concerns about copyright enforcement in generative video tools, according to Deadline. The debate centers on Hailuo 3, a video model developed by Chinese AI company MiniMax. Although the model is not directly available to users in the United States or European Union, third-party services have enabled people elsewhere to access it. One example cited by Deadline involved an AI-generated clip featuring Jim and Dwight from The Office.
MiniMax is currently defending a lawsuit brought by Disney, Universal and Warner Bros. The studios allege that Hailuo illegally reproduces copyrighted works on a large scale. MiniMax denies the claims. The company has also removed AI videos featuring characters including Deadpool, Wolverine and the Joker from its official YouTube account.
ByteDance, another major Chinese AI developer, has faced similar concerns. Researchers at VN.ai said they were able to generate recognizable images of copyrighted characters, including Elsa, Shrek and Buzz Lightyear, through ByteDance’s Seedream 5.0 image tool by modifying prompts and accessing the system through a third-party API.
ByteDance said its models include safeguards against unauthorized intellectual-property use and that it is working to strengthen them with rights holders and creative organizations.
SAG-AFTRA lawyer warns against weakening performer protections
SAG-AFTRA Chief Legal Officer Jeffrey Bennett has renewed the union’s focus on consent, compensation and control as Hollywood negotiates the use of generative artificial intelligence. Speaking with other industry experts at Digital Hollywood 2026, Bennett discussed the legal and creative risks associated with AI-generated performances, digital replicas and the use of copyrighted material to train AI systems. SAG-AFTRA’s position is that a performer’s voice, likeness and recorded performance should not be replicated or reused without clear authorization.
The union has negotiated contractual protections requiring producers to provide notice and obtain consent for certain digital replicas. Its broader campaign also includes support for legislation addressing unauthorized digital replicas and synthetic media.
Bennett’s comments come as studios, game companies and technology firms explore AI tools that can generate voices, faces and performances. For performers, the central concern is that a digital scan or recording could be reused beyond the original production or used to replace human labor without a new agreement. The dispute also extends to copyright. SAG-AFTRA has argued that AI-generated material incorporating an unauthorized replica should not receive copyright protection for that portion of the work.
AI visual-effects market forecast to expand through 2032
Artificial intelligence is expected to play a growing role in visual-effects production through 2032, according to a market report released by Market Research Corridor. The report examines the use of generative AI for asset creation, real-time virtual production, automated compositing, neural rendering and deep-learning-based performance capture. It identifies Adobe, NVIDIA, Autodesk, Foundry, Weta Digital, Industrial Light & Magic, Framestore, DNEG and Pixar among the prominent companies and studios operating across the sector.
The research divides the market by application, including film and television, video games, architecture, advertising and scientific visualization. It also considers machine learning, deep learning, computer vision, natural-language processing and motion capture, as well as on-premises, cloud-based and hybrid deployment models.
According to the report, studios and production companies are pursuing AI tools to address rising costs, shorten production schedules and reduce bottlenecks in areas such as rotoscoping, rendering and asset creation. The technology is also converging with virtual production, game engines and remote collaboration systems.
Brad Pitt says AI could help revive mid-budget movies
Brad Pitt says artificial intelligence could help more mid-budget films get made, provided filmmakers use the technology as a production tool rather than a replacement for human creativity. In an interview with Esquire, Pitt said AI could reduce the cost of visual effects for movies budgeted between $40 million and $90 million. Those films have increasingly struggled to secure financing as studios concentrate on lower-cost productions and large-scale franchises.
“There’s a lot of pushback against AI,” Pitt said, adding that the technology could help mid-budget projects move forward if used appropriately. Pitt expressed greater concern about AI being used to replicate actors’ performances. He said a generated version of an actor might imitate a voice or image but would not necessarily reproduce the choices and discoveries that the performer made during production.
The actor also discussed body scanning, a practice used in stunts, battle scenes and de-aging. Pitt said he has traditionally required contracts to state that his scans be destroyed, reflecting performers’ concerns about how digital likenesses could be stored and reused.
Autodesk adds 3D controls to AI filmmaking tools
Autodesk Flow Studio is introducing tools designed to give filmmakers greater control over AI-generated scenes, addressing one of the technology’s central weaknesses: the difficulty of maintaining precise camera positions, depth and character movement. The new package, called 3D Editor + Canvas, combines prompt-based AI image generation with a three-dimensional editing environment. Canvas uses multiple AI models to create visual material, while the 3D Editor allows users to manipulate shots and characters in a 360-degree space.
The system is intended to reduce the trial-and-error process common in AI filmmaking, in which users repeatedly regenerate a scene until it approximates the desired result. Filmmakers can instead adjust viewpoints, environments and character placement more directly.
Autodesk Flow Studio co-founder Nikola Todorovic said the tools are designed to bridge the gap between predominantly two-dimensional AI generation and the 3D control available in professional computer-generated imagery software.
In demonstrations, the company showed users placing an AI-generated character into an existing environment, building a setting around a character, and transferring live-action performances into synthetic locations. The approach is sometimes described as performance transfer.
EU begins enforcing AI-content transparency rules
The European Union has begun enforcing new transparency requirements for artificial intelligence, requiring companies to disclose certain AI-generated content and tell users when they are interacting with an AI system. The rules, contained in Article 50 of the EU AI Act, took effect Aug. 2. They apply to providers and deployers of systems that create or significantly alter realistic images, audio, and video that could be mistaken for authentic material.
Content created entirely by AI must carry an appropriate disclosure, while substantially manipulated material involving real people or events must also be identified. The rules extend beyond media production to customer-service systems, search tools, social platforms and systems that analyze biometric information or infer emotions.
The European Commission says the requirements are intended to help people distinguish between human-made and synthetic material without banning generative AI. Private communications and clearly artistic, fictional or satirical works are among the reported exemptions.
The regulations also introduce responsibilities for providers of general-purpose AI models, including technical documentation, training-data summaries and copyright-compliance policies. Violations could result in fines of up to €15 million or 3% of a company’s annual global turnover, depending on the offense.
The rules are likely to affect studios, streaming services, advertisers and independent creators distributing AI-generated material in the EU.
Internet Cinematic launches India-Japan AI film studio
Internet Cinematic launched an AI-focused film research and production studio in Tokyo on Aug. 2, bringing together filmmakers, technologists and creative executives from India and Japan. The launch event drew nearly 300 film professionals and was designed to strengthen collaboration between the two countries in cinema, storytelling and emerging technology. The studio was created by Kei Hayashi, Hardeep Gambhir, Suhas Sumukh and Chandan Perla and grew out of LocalHost HQ, a Bengaluru-based media and technology research lab.
Internet Cinematic is developing AI-assisted workflows for visual design, animation, editing and post-production. Its founders say the goal is not simply to produce AI-generated content, but to make ambitious filmmaking more accessible while keeping artists and filmmakers at the center of the creative process.
The Tokyo launch builds on the group’s previous work in India, including the Mumbai AI Film Festival at the Royal Opera House and the India AI Film Festival at New Delhi’s Qutub Minar. Those events reportedly received more than 1,800 international submissions and included participants from Netflix India, JioStar and Google.
The studio’s supporters see India and Japan as potential partners in developing new AI filmmaking practices. Its Tokyo presence is intended to create a cross-border network linking filmmakers, researchers and technology companies.
