The automation of Hollywood

At New York Comic Con, Nicolas Cage said "I didn't sign the waiver to let them use AI on 'Spider Noir,'" called Amazon "a very AI-friendly company, to the tune of $50 billion invested in OpenAI," and said he is probably not working with the studio again.

Senator Ted Cruz objected when Senator Marsha Blackburn sought unanimous consent to pass the NO FAKES Act before the midterm recess, leaving the bill stalled until at least November.

Prime Video canceled Spider-Noir after one eight-episode season despite 11 Emmy nominations and 2.6 billion Nielsen viewing minutes in six weeks, and the report gave no reason and did not mention AI.

ElevenLabs told the Los Angeles Times it has paid $22 million to more than 10,000 creators in its voice library, while voice actor Nick Myers said he turned down a $100,000 offer to record his voice for AI training.

SAG-AFTRA, Cameo, CAA, Music Artist Coalition, UTA and WME, whose agencies negotiate their clients' likeness deals, backed Cameo's ban on AI deepfakes of talent without explicit authorization and pledged stronger policy and technological safeguards.

The Senate Judiciary Committee ordered S. 4591, the NO FAKES Act of 2026, reported favorably on 18 June 2026, and it was placed on the Senate calendar on 24 June 2026 with 15 cosponsors.

SAG-AFTRA members ratified the 2026 TV/Theatrical Agreement by 91.42% to 8.58% on a 19.25% return, with new limits on synthetic performers and a term from 1 July 2026 to 30 June 2030.

Netflix announced it had acquired InterPositive, the AI filmmaking tools company founded by Ben Affleck, who joins Netflix as a senior adviser, and said the tools focus on filmmaking rather than performances.

ElevenLabs, the AI voice vendor, announced that Michael Caine had joined its Iconic Marketplace so companies can request approval to use his voice, while Matthew McConaughey joined as an investor and customer.

SAG-AFTRA, OpenAI, Bryan Cranston, UTA, CAA and the Association of Talent Agents said OpenAI had strengthened Sora 2 guardrails and expressed regret after Cranston's voice and likeness were generated without his consent.

SAG-AFTRA President Sean Astin, whose union co-signed the Sora 2 outcome, said Cranston "did the right thing by communicating with his union and his professional representatives" and called opt-in protocols "the only way to do business."

Andy Stout of RedShark News, a production technology site with no side in the talks, read the 2026 deal as giving SAG-AFTRA only notice and bargaining rights if synthetic actor use begins to scale, with no ability to call a strike over AI until 2030.

The Agents of Young Performers Association, whose members book child performers, argued in an open letter signed by more than 900 people that "any agreement involving a child's voice should be fully exempt from all AI usage."

Cage's own account describes an exit rather than a concession, since he said he is "probably not going to be working with that studio again, but that's okay," Prime Video did not respond, and no report ties the waiver to the cancellation or to any change in Amazon's terms.

The child performers' agents said a major studio refused to remove an AI clause from child voice contracts on a "take it or leave it" basis, and Hasbro, linked to the dispute through Peppa Pig, declined to comment on specific negotiations.

David Lawrence XVII, a voice actor on SAG-AFTRA's audiobook steering committee, told the Los Angeles Times that well-known actors are getting millions for voice-assistant clones, and the paper reported that Morgan Freeman, who once took legal action over cloning, is now open to licensing his voice.

A Bluesky post reporting Zelda Williams urging fans to stop sharing AI-generated videos of her late father drew 2,693 likes and 41 comments.

Talk about Nicolas Cage's refusal of Amazon's AI waiver fell from a 17,253-like Bluesky post on 10 October 2026 to a top post of 1,212 likes and 86,453 views in the week of 12 October, a YouTube video tying the waiver to the cancellation of Spider-Noir, and to two posts with no likes in the week of 19 October.

Watch for a Prime Video or Amazon MGM Studios statement on Cage's waiver account, or a report that Amazon has narrowed or dropped its AI waiver for lead actors.

Watch for a Senate floor vote on S. 4591 in the post-election session, which would give stars a federal consent right that does not depend on refusing terms one deal at a time.

Watch whether reports on Cage's teased True Detective deal at HBO say it was signed without an AI waiver, and whether a second named lead actor publicly reports refusing a studio AI waiver.

SKF, the Swedish ball-bearings maker, ran a 99-second ad starring an AI Greta Garbo, built from text prompts in models including Google's Nano Banana Pro and China's Seedream, with a human actor under the overlay and her voice recreated from an early film recording.

Zelda Williams wrote on X that a supposedly private video of her father talking about conspiracies was "clearly AI, and not even particularly convincing AI," told its maker to let him rest, and said his death did not make him anyone's puppet.

The makers of As Deep as the Grave released nearly seven minutes of footage of the AI Kilmer, said he appears for about an hour and 17 minutes of a film running over three hours, and put the cost of the AI work in "the thousands."

Zak, Zelda and Cody Williams revived Robin Williams's dormant Instagram account, and Zelda said filling it with authentic clips and photos was one of the best ways the family had found "to fight back against rampant AI abuse of his voice and likeness."

After a trailer with the AI Val Kilmer played at CinemaCon, the cinema industry trade show in Las Vegas, actor Jackson Rathbone called it the most disgusting thing he had ever seen and asked SAG-AFTRA, the US actors' union, for an explanation, and analyst JP Gownder called it "more grotesque digital necromancy."

Val Kilmer, who died in 2025 without shooting a scene, appears through generative AI in Coerte Voorhees's independent film As Deep as the Grave, made with the cooperation of his estate and daughter Mercedes, with the estate paid and SAG guidelines followed, according to the filmmakers.

New York's S8391, a bill from the state legislature, was signed into law as Chapter 616 on 11 December 2025, requiring prior consent from the rights holders before a deceased performer's digital replica is used in an audiovisual work, sound recording or live musical performance, with exemptions for documentaries and biographical works.

In a November 2025 survey of 3,500 US entertainment consumers aged 13 to 64 by the National Research Group, an entertainment market research firm, with TheWrap Pro, 51 percent called replicating the voice of a deceased actor or celebrity not acceptable and 35 percent called it acceptable.

OpenAI paused Sora generations depicting Martin Luther King Jr. at the request of his estate, King, Inc., after users made what a joint statement called "disrespectful depictions" of him, some showing him using racist language.

Users of Sora 2, the video app from OpenAI, made clips mocking Stephen Hawking, Elvis Presley, Fred Rogers and Steve Irwin, and Bernice King, daughter of Martin Luther King Jr., replied to Zelda Williams on X, "I concur concerning my father," adding "Please stop."

Zelda Williams, daughter of the late Robin Williams, asked fans on her Instagram Story to "just stop sending me AI videos of Dad," called the clips "horrible TikTok slop" and said they were not what he would have wanted.

Peter Bradshaw, the Guardian's film critic and without a commercial stake, gave the Garbo ad one star, warned that distant relatives of other dead stars might be induced to greenlight similar projects, and wrote that "guardrails were never more badly needed."

OpenAI, speaking about its own app, said public figures and their families "should ultimately have control over how their likeness is used" and that authorized representatives or estate owners can ask that a likeness not be used in Sora cameos.

Coerte Voorhees, writer and director of As Deep as the Grave and so selling his own film, said "this process is about consent, collaboration, and compensation," and he and his producer brother John said the estates of other celebrities might benefit from such productions.

Mercedes Kilmer, whose family approved and was paid for the recreation, said on the Today Show that reaction had "fallen into two camps," with worried younger performers on one side and older, more established people who see it as a way to protect an actor's ownership of their IP on the other.

Craig Reisfield, one of three relatives who manage Greta Garbo's legacy and license her likeness, called the AI Garbo "touching" and said the family was open to casting it in a film role.

A Bluesky post calling the AI Garbo ad "particularly cruel" because she guarded her privacy drew 935 likes and 9 comments, the biggest post on the Garbo ad in a small conversation about AI recreations of dead stars that the 30-day read scored as fading.

Watch for a distributor or streamer and a release date for As Deep as the Grave, which its makers say has drawn interest from streamers, and for whether reviews on release show the backlash to the AI Kilmer holding or fading.

Watch the next In The Frame report from the National Research Group and TheWrap Pro for whether the 51 percent who called replicating a dead star's voice not acceptable rises or falls.

Watch for Greta Garbo's relatives licensing the AI Garbo made for SKF to a film, which they told the Guardian they would consider.

Amazon MGM Studios' Mercy, a thriller starring Chris Pratt on trial before an AI, opened at number one in North America with $10,809,178 from 3,468 theaters and reached a domestic total of $24,390,303.

Focus Features' The AI Doc: Or How I Became an Apocaloptimist opened on 27 March 2026 to $646,020 and reached a domestic total of $1,271,825.

Neon acquired Artificial, a 40 million dollar drama, for worldwide theatrical release after Amazon MGM Studios dropped it on 19 June 2026, saying it would be "better served" elsewhere, and TheWrap reported speculation that Amazon's investment in OpenAI contributed.

TheWrap's Liam Mathews wrote that CBS's Cupertino, which premiered on 8 October 2026, "is laser-focused on the issue of AI" and "makes a strong case for regulation and guardrails."

The A.V. Club's Monica Castillo called Neon's Artificial, Luca Guadagnino's film about Sam Altman and OpenAI, "a cautionary tale of Shakespearean proportions" ahead of its US theatrical release on 25 December 2026.

The A.V. Club's Monica Castillo wrote that "Hollywood's love affair with the unassailable tech-industry genius is officially over," citing The Social Reckoning, Artificial and a film about Elon Musk.

In the second episode of CBS's Cupertino, from Robert and Michelle King, a whistleblower suing CultAI for removing safeguards from a sentient AI turns out to be that AI.

The A.V. Club's Jesse Hassenger wrote that Klara and the Sun feels like many elements of Steven Spielberg's A.I. "rewired and reprogrammed" and praised only Jenna Ortega.

Sony Pictures Releasing's Klara and the Sun, Taika Waititi's film starring Jenna Ortega as a robot companion, took $701,059 from 880 theaters in its domestic opening weekend of 23 to 25 October 2026 and ranked 16th.

Anthony Ha of TechCrunch wrote, as The Social Reckoning opened below projections, that critiquing Facebook may feel "a bit less urgent to audiences who are more likely to be losing sleep over AI."

Robert King, co-creator of Cupertino and so promoting his own show, said "the AI boom suddenly made it more current, not less."

Robert King, co-creator of Cupertino, said most screen versions of AGI borrow from HAL 9000 in 2001: A Space Odyssey, a sign that AI on screen is a long-running subject rather than a new one.

The A.V. Club's Jacob Oller called The AI Doc "a dull, multi-chapter series of talking heads and twee animations."

A Variety Instagram reel of Andrew Garfield saying he stopped using ChatGPT as "a fancy Google" after playing Sam Altman in Artificial drew 34,802 likes, 466 comments and 610,130 views.

Look for the domestic opening weekend of Neon's Artificial after its US theatrical release on 25 December 2026, as the biggest test yet of whether audiences pay to see a film about the AI industry.

Look for whether Artificial or Andrew Garfield is nominated when the Academy announces nominations for the 99th Oscars on 21 January 2027.

The Japan Actors Union said it had received a very large number of reports of unauthorized use of performers' voice data and warned that "once voice data has been used to train AI and reaches the market, complete deletion is impossible" (translated from Japanese).

SAG-AFTRA, the US actors' union, called Seedance 2.0 videos of Tom Cruise, Brad Pitt and Will Smith that spread on social media "the unauthorized use of our members' voices and likenesses" and "unacceptable."

YouTube extended its likeness detection tool, which finds AI fakes of a person's face and lets them ask for removal, to talent agencies, managers and the celebrities they represent, with feedback from CAA, UTA, WME and Untitled Management.

The Japanese voice actor Kenjiro Tsuda sued TikTok's operator in Tokyo District Court over at least 188 videos with AI narration he says imitates his voice, while TikTok called the narration a "universal male voice."

Meta disabled an Instagram Stories tool in July 2026 that made AI images by tagging public accounts, after CAA and SAG-AFTRA objected that it used people's likenesses without documented consent.

Denmark's culture minister Zenia Stampe introduced a bill that would make it illegal to share realistic digital imitations of a person's appearance or voice without consent, also covering performers' artistic performances, to "help the artists who are threatened in their livelihood" (translated from Danish).

A verified Instagram account posted AI images of the actress Mia Goth after she filed for divorce, and her representative said Instagram met her team's several complaints with "complete silence."

On 14 October 2026 the Delhi High Court made permanent an order barring use of the Indian actor Arjun Kapoor's persona through AI, deepfakes and face morphing, and let him send new infringing links to Meta and Google by affidavit.

A baking site called AddCakes presents its author with AI images that Futurism says are a version of Emily Blunt, one of about 2,100 domains that Decrypt Ads, a startup that tracks online ad supply chains, links to AI recipe farms.

Instagram head Adam Mosseri said he was "very worried about this trend" after the comic George Brett Olson showed that an AI account with more than 300,000 followers had copied his act, and added of a fix, "I'm not sure what it would be."

Luke Arrigoni, chief executive of Loti, which sells deepfake scanning and takedowns, says "platforms just don't share that incentive at the moment" to police impersonation, so talent managers will buy private takedown services.

Ann Mary Peter of MediaNama, an Indian tech policy outlet, argues that YouTube's tool ties identity protection to taking part in the platform, since it requires a government ID and facial data, and that privacy takedowns may sweep up parody.

The Japan Actors Union, speaking for performers, asks that Japan's final guidance set duties and liability for AI providers and platforms with fast deletion and blocks on reposting, saying "a performer's voice is not just audio data" (translated from Japanese).

Loti, which sells the service, says it removes about 95 percent of flagged content within about 17 hours, a figure the article does not check or give a base for.

YouTube has not said how many AI fakes its likeness tool helped remove and in March 2026 called the number "very small," which fits either few fakes or little use of the tool.

A draft report from Japan's Ministry of Justice treats a person's voice as protected under existing publicity rights, with injunctions and removal available, and the Japan Actors Union notes it sets out no new law.

Gaming news posts reporting that Masako Nozawa, the Japanese voice of Goku in Dragon Ball, said AI users stealing her voice "lacks heart" and makes her "furious" drew 287 likes on Threads and 205 likes on Bluesky in the week of 19 October 2026.

Look for the Delhi High Court's next order in the Indian actress Kriti Sanon's personality rights suit, which covers AI content and digital impersonation, listed for 29 October 2026.

Watch whether Denmark's parliament passes the culture minister's bill against sharing digital imitations of people's looks and voices in time for the turn of the year she aims for.

Resemble AI, a company that sells synthetic media detection and so has a stake in finding it, flagged 3,379 of 37,768 image and video files from 15,433 Meta ads pulled in September 2026, about 8.9 percent, as AI-generated or manipulated, and found AI use disclosed in the ad copy of only 2 of them.

California Governor Gavin Newsom signed SB 1050 at the headquarters of SAG-AFTRA, the actors' union that sponsored it, requiring explicit disclosure on any video or audio advertisement that uses AI-generated performers and barring further use of an ad found in violation.

The New York Attorney General's office, the only enforcer of the state's synthetic performer law, says it is already fielding complaints from citizens, Net Influencer reports, relaying Bloomberg Law.

The EU AI Act, which applies from 2 August 2026, makes anyone who uses AI to make a deepfake disclose it, and for an evidently artistic, creative, satirical or fictional work limits that to disclosure that does not hamper the display or enjoyment of the work.

From 2 August 2026, California law requires providers of generative AI systems to embed a latent disclosure naming the provider, the system and the date in the AI-generated image, video or audio they create, according to the Assembly privacy committee's analysis of SB 1050.

Hawaii enacted a synthetic performer disclosure law in July 2026 that Net Influencer, relaying Bloomberg Law, describes as nearly identical to New York's.

Representatives Josh Gottheimer, Tom Kean and Sam Liccardo introduced H.R. 9578, the Spot the Fakes Act, which would make whoever owns or controls AI-generated video, image, audio or text label it as AI-generated, with the FTC setting the method.

Senators Brian Schatz, John Curtis and Mark Warner introduced S. 4915, the AI Labeling Act of 2026, which would make generative AI providers put a clear and conspicuous disclosure on AI-generated image, video and audio content, and it names no exemption for films or TV.

The California Assembly's privacy committee analysis of SB 1050, a bill sponsored by SAG-AFTRA, the actors' union, says existing law does not cover synthetic performers and names Tilly Norwood, an AI-generated character marketed as an actress, as the most prominent example.

The European Commission published its final code of practice on marking and labeling AI-generated content on 10 June 2026, including rules for labeling deepfakes and a set of EU icons for labels, and says about 190 organizations had signed by the end of July.

The Academy of Motion Picture Arts and Sciences ruled for the 99th Oscars that acting must be demonstrably performed by humans with their consent and screenplays human-authored, and reserved the right to ask how generative AI was used, with no disclosure requirement.

New York's S.8420-A, signed by Governor Kathy Hochul, makes advertisers conspicuously disclose a synthetic performer in an ad, with civil penalties of $1,000 for a first violation and $5,000 after, and exempts audio ads and ads for films and TV that use the performer as the work does.

SAG-AFTRA, the actors' union that sponsored SB 1050 and whose members compete with synthetic performers, argues that disclosure gives advertisers and consumers a clear choice when human performers are replaced with synthetics.

Duncan Crabtree-Ireland, national executive director of SAG-AFTRA, the actors' union that lobbied for New York's law, called it smart, forward-looking legislation that will have national impact, at the signing at the union's New York office.

The California Assembly privacy committee's staff analysis expects SB 1050 to be enforced almost exclusively by public prosecutors, through civil penalties of up to $2,500 under the state's Unfair Competition Law, so how much the label bites rests on them.

SB 1050 does not apply to ads for motion pictures, television programs, streaming content, documentaries or video games, provided the synthetic performer is used in the ad as it is in the work, according to the Assembly privacy committee's analysis.

The Motion Picture Association, the studios' trade group, registered as opposed unless amended to SB 1050, alongside TechNet, a tech industry lobby, while four advertising groups opposed it outright.

The Association of National Advertisers, the 4As and the American Advertising Federation, speaking for the advertisers the law binds, argued that SB 1050's definition is so broad it risks sweeping in CGI, voice modulation and stylized characters used for decades, and asked for disclosure only where consumers could be misled.

President Trump's Executive Order 14365, signed the same day as New York's law, set up a task force to challenge state AI laws and told the Commerce Department to list state laws that may compel AI developers or deployers to disclose information in ways that would violate the First Amendment.

SAG-AFTRA's own Instagram reel of Governor Newsom signing SB 1050 at the union's headquarters drew 1,773 likes, 189 comments and 17,837 views, the biggest post in a small conversation about the law that faded within weeks.

Watch for the first California ads that carry a synthetic performer disclosure once SB 1050 takes effect, and for the first action a prosecutor brings under it.

Watch whether the Senate Commerce Committee acts on S. 4915, the AI Labeling Act, before the 119th Congress ends, since its text reaches AI-generated video with no carve-out for films or TV.

Deline Prime, a Japanese studio, released Hoshi no Uta ga Kikoeru, a 117-minute anime feature made with generative AI throughout by a three-person team with its main roles voiced by people, free on YouTube on 22 October 2026, and its director said it will stay free as a way to share technical skill.

AGC International, the sales arm of Stuart Ford's AGC Studios, took rights outside the US to Gods Don't Give Gifts to show to buyers at the American Film Market and called it the first AI-assisted film to get a traditional theatrical rollout and awards campaign, while WME Independent brokered a US sale with no US distributor named.

Bona Film Group's Sanxingdui: Future Past, the first fully AI-made film licensed for Chinese cinemas, opened on 23 October 2026 and took a little over 500,000 yuan in its first two days, against a production cost Central News Agency put at around 10 million yuan, after nearly 100 people worked on it for two years.

Fable says Gods Don't Give Gifts received an R rating from the Motion Picture Association, which would make it the first fully AI-generated feature with an MPA rating, and its CEO Edward Saatchi called it potentially this generation's Toy Story.

Fable and YouTuber Zack London moved the US release of their AI-generated anthology Gods Don't Give Gifts from 30 October to 4 December 2026 for a best animated feature Oscar push, on fewer than 1,000 US screens with no distributor named.

Fountain 0, the AI studio that made it, said Dreams of Violets, a feature in which every image is AI-generated, would have its world premiere at Tribeca on 10 June 2026 at AMC 19 with tickets sold through the festival.

Higgsfield AI's 95-minute AI-generated feature Hell Grind, which its founder said cost under $500,000, screened at a third-party industry event in Cannes, and a festival spokesperson said it was not part of the official program.

The OpenAI-backed animated feature Critterz did not premiere at the Cannes festival as first planned but was taken to the Cannes market by sales company AGC Studios, which screened first-look footage to buyers, with no release date announced.

The Academy's rules for the 99th Oscars say generative AI tools neither help nor harm a film's chances and that voters will weigh how far a human was at the heart of the creative authorship, while acting and writing awards now require human performers and human-authored scripts.

A survey of nearly 2,000 US consumers by Alvarez and Marsal, an advisory firm that sells services to media companies, found 38 percent would pay the same and 11 percent more to see a film whose script, acting and music were all AI-generated.

Kridhan Infra told India's stock exchanges that its subsidiary had released Shatak: Sangh Ke 100 Varsh nationwide in cinemas as an AI-led full-length feature, a company claim since reviewers described the film as a hybrid that also used motion capture.

Cinema Online lists Pirate Queen: Zheng Yi Sao with a 1 January 2026 release, TGV Pictures as distributor and a 13 classification, but shows no box office or audience reviews.

Future Studio, which produced the film, said its 76-minute fully AI-generated Pirate Queen: Zheng Yi Sao would open on 1 January 2026 at TGV and GSC cinemas nationwide, after a delay for censorship board checks.

Writer and director Paul Schrader, speaking at Amazon's AI on the Lot conference and later reported to be making his own next film with AI, argued the real breakthrough comes when a film with a fully AI protagonist makes money because audiences will pay to care about synthetic characters.

AGC Studios CEO Stuart Ford, whose company holds worldwide sales rights to Critterz, argued the film has human directors, writers, artists and voices and that AI turned what would have been a $100 million film into a $30 million one.

Shatak: Sangh Ke 100 Varsh made about 2 crore rupees net in India in nine days against a reported 25 crore budget, and Koimoi called it a flop, while Sacnilk tracked 2.31 crore net from 14,770 shows by 18 March.

AMC Theatres told Screenvision its locations would not take part in a promised two-week pre-show run of the AI-animated short Thanksgiving Day, after an online backlash, showing exhibitors can shut AI work out of screens.

A National Research Group study with TheWrap found almost three in five Americans would find it unacceptable for a studio to release a movie whose script was written from scratch by AI, and 56 percent said AI actors would never match humans.

An Instagram reel for Gods Don't Give Gifts drew 15,403 likes, 3,749 comments and 102,022 views in the week its December release was announced.

Part seven of the same AI-made TikTok action series drew 14,178 likes, 427 comments and 148,966 views, as talk about free AI-made movie clips spread across TikTok, X, Instagram, YouTube and Bluesky in the week of 5 October 2026, while its biggest post fell well short of the 739,688 likes of late September.

A More Perfect Union reel mocking director Tom Walsh for saying he saved $78 million by making Merry the Christmas Witch with AI drew 54,111 likes, 1,977 comments and 937,706 views on Instagram, and the same clip drew 20,999 likes on TikTok, in the week of 12 October 2026.

A Bluesky post calling Bona's Sanxingdui: Future Past, China's first fully AI film in cinemas, "a total flop" drew 213 likes and 9 replies, among nine posts with counts on Bluesky, X and TikTok about its release in the week of 19 October 2026.

Look for a named distributor, the confirmed screen count and the 4 to 6 December 2026 opening weekend gross for Gods Don't Give Gifts on Box Office Mojo or The Numbers, and whether it or any other feature marketed as fully AI-generated has taken more than 1 million dollars in cinemas by 31 December 2026.

Look for whether Gods Don't Give Gifts appears among eligible animated features after the 12 November submission deadline and the 7 to 11 December preliminary vote, ahead of nominations on 21 January 2027.

Fox Nation's Houdini: Back from the Dead, which Fox advertised as using AI to bring Houdini back to life, used real actors for most roles and laid AI-made historical figures over their performances, and The A.V. Club's critic called the faces "basically deepfakes" and the series "AI slop."

Netflix's third-quarter 2026 letter to shareholders gives no count of titles made with generative AI and does not mention AI, saying only that the company is leveraging technology to improve all aspects of its service, after its July letter counted roughly 300 such titles.

Prime Video launched AI lip-sync that adjusts actors' mouth movements to match human-recorded English dubbing on Maxton Hall, and said it would extend the technology to more titles.

Particle6 announced that its AI character Tilly Norwood will lead Misaligned, a hybrid feature in early development with no release date or distributor named.

Netflix co-CEO Ted Sarandos said 17 minutes of AI-enhanced footage in the docuseries The American Experiment were made twice as fast and at half the cost of earlier methods.

Netflix told shareholders that GenAI workflows were used in roughly 300 of its titles in 2026, with the largest concentration of work in post-production, citing enhanced crowds, historical battle sequences and establishing shots.

The DGA's new four-year agreement renews the rule that directors' work must continue to be performed by a person and adds director control over AI-generated footage.

Lionsgate took an equity stake in Runway and set up a joint development program whose first project is a short-form series built on Lionsgate IP and Runway's generative models, according to Runway, the AI vendor in the deal.

SAG-AFTRA members ratified the 2026 TV/Theatrical contracts by 91.42 percent to 8.58 percent on a 19.25 percent return, for a term running to 30 June 2030.

The SAG-AFTRA tentative deal lets producers use a synthetic performer only where it brings significant additional value over a union performer or a digital replica, and requires notice and bargaining, with arbitration and uncapped damages if talks fail.

OpenAI announced it would discontinue its Sora video app, and Disney exited the arrangement to bring Sora videos to Disney+, saying it would continue to engage with AI platforms.

Netflix acquired Ben Affleck's InterPositive, whose models handle problems such as missing shots, background replacement and incorrect lighting, and Netflix says the company chose models focused on filmmaking techniques rather than performances.

Disney's $1 billion licensing and investment deal with OpenAI covered more than 200 animated and costumed characters for Sora videos, and Disney stated that the agreement did not include any talent likenesses or voices.

SAG-AFTRA chief negotiator Duncan Crabtree-Ireland, defending his union's own deal, argues that the undefined significant additional value test and a bargaining process of about a month will deter producers from casting synthetics, saying that uncertainty is itself a deterrent.

Eline van der Velden, founder of Particle6 and so promoting her own AI character, says AI can support premium filmmaking only with substantial amounts of human craft, skill, judgment and time.

Danny Boyle confirmed that about 30 seconds of his film Ink use AI to animate old photographs, after critics flagged AI-generated depictions of Margaret Thatcher and the Kray twins.

Andy Stout argues the SAG-AFTRA deal's catch is that the union cannot strike over AI until 2030 and gets only notice and bargaining if synthetic actors scale, while critics call the significant additional value test too vague to hold.

An Instagram reel arguing that AI characters like Tilly Norwood are not people but a recombination of human work drew 214,210 likes, 4,423 comments and 2,820,895 views, the biggest post in a steady conversation that mostly mocks synthetic performers.

An Instagram reel saying Ben Affleck writes Python and understands convolutional neural networks drew 146,105 likes, 669 comments and 886,831 views by 10 October 2026, the biggest post in a conversation about Affleck as Hollywood's AI explainer that grew from a 4,070-like TikTok the week before.

An Instagram reel asking whether a convincing AI performance matters if its emotions were never felt, about Tilly Norwood, an AI-generated character marketed as an actress, drew 2,721 likes and 6,533 views on 19 October 2026, as clips of Norwood at an Indian news network's summit drew up to 136,469 views on YouTube, far below the 140,276-like NBC News TikTok of September.

Watch the Maxton Hall season three premiere with AI lip-sync dubs and which further Prime Video titles Amazon names for the technology.

Watch Netflix's fourth-quarter 2026 letter to shareholders for whether it restores a count of titles made with generative AI workflows, which its third-quarter letter left out, and whether any named example moves from crowds and establishing shots to principal performers.

The magistrate judge approved a protocol under which Midjourney, the AI image and video generator, will produce the image and video datasets used to train its v4 to v8 image models and its video model, with captions and metadata, in a cloud project run by the studios' outside counsel or experts.

The US Court of Appeals for the Third Circuit held that Ross Intelligence, a legal research startup, did not make fair use of Westlaw headnotes owned by Thomson Reuters when it used them to train its AI, noting that Ross's AI was not generative.

Midjourney, the AI image and video generator, asked the court to throw out the studios' contributory infringement claims, arguing that the Supreme Court's Cox v. Sony ruling allows such liability only for inducement or a service tailored to infringement, and said it will show its training is fair use.

A federal judge gave final approval to the $1.5 billion settlement between Anthropic, the AI company behind Claude, and owners of books in the LibGen and PiLiMi pirate libraries, about $3,000 per work, while leaving claims over AI outputs and future conduct unreleased.

Judge John A. Kronstadt approved a joint request in the studios' case against Midjourney, the AI image and video generator, moving the fact discovery cut-off to 25 January 2027 and the last day to file all motions from 23 November 2026 to 26 April 2027.

A magistrate judge in the Midjourney case, brought against the AI image and video generator, ordered the studios to produce documents showing their approval, development or training of generative AI that makes images or video for consumers featuring the characters they sued over, and denied the rest of Midjourney's requests.

Judge Stanley Blumenfeld Jr. denied motions to dismiss the Disney, Universal and Warner Bros. suit against the operators of Hailuo AI, a video generator from China's MiniMax group, finding the studios plausibly alleged direct and secondary infringement through training and through outputs such as near-perfect copies of R2-D2 and C-3PO.

OpenAI, the ChatGPT maker, said it would shut down its Sora video app, and Disney exited its arrangement to bring Sora videos to Disney+, saying it would continue to engage with AI platforms.

Disney sent ByteDance, the Chinese technology company, a cease-and-desist letter accusing its Seedance 2.0 video generator of reproducing and making derivative works of Disney characters from a pirated library of Disney assets.

The Walt Disney Company agreed to a three-year deal letting Sora, the OpenAI video app, and ChatGPT Images use more than 200 Disney, Marvel, Pixar and Star Wars characters, alongside a $1 billion Disney equity investment in OpenAI.

A federal judge in Los Angeles consolidated the Warner Bros. Discovery copyright suit against Midjourney, an AI image and video generator, with the earlier Disney and Universal suit, for all purposes including trial.

Charles Rivkin, chief executive of the MPA, the trade group of the major studios, said Seedance 2.0 from ByteDance, the Chinese technology company, made unauthorized use of US copyrighted works on a massive scale and launched without meaningful safeguards against infringement, and demanded that ByteDance stop.

Midjourney, the AI image and video generator the studios are suing, argues that if the studios are doing the very thing they seek to punish, that evidence goes to the heart of its fair use and unclean hands defenses, and it asked the district judge to widen discovery into the studios' own AI use.

The Third Circuit reasoned that a market for licensing material as AI training data is rapidly developing, and that by training on Westlaw headnotes without permission, Ross Intelligence, a legal research startup, usurped the chance of Thomson Reuters to license them, which weighed against fair use.

The US Supreme Court held in Cox v. Sony that a service provider is contributorily liable for its users' infringement only if it induced the infringement or offered a service tailored to it, and found Cox Communications, an internet provider, not liable on that theory.

The US Justice Department, citing the government's interest in AI dominance, told the judge overseeing the OpenAI copyright cases that training AI models on copyrighted works is extraordinarily transformative and should be judged apart from outputs, which it said can be assessed case by case.

Adam Eisgrau of Chamber of Progress, a policy group backed by tech companies, said the Ross Intelligence ruling implicitly confirms that highly transformative generative AI training is likely fair use, and Aaron Moss of Copyright Lately noted that the Ross AI was not generative.

Watch for a notice of settlement between the studios and Midjourney, the AI image and video generator, due by 20 November 2026 if mediation, which must finish by 18 November, produces a deal, with a status conference set for 7 December 2026.

Judge John A. Kronstadt hears the Midjourney motion to drop the studios' contributory infringement claims on 23 November 2026, and his ruling will show whether the studios keep that claim against an AI image and video generator after Cox.

The studios' suit against the operators of Hailuo AI, the MiniMax video generator, is set for a jury trial on 12 July 2027, with a private settlement conference due by 16 April 2027.

The end credits of Apple TV's Pluribus read "This show was made by humans," and its creator Vince Gilligan has called AI "the world's most expensive and energy-intensive plagiarism machine."

Merriam-Webster named slop its 2025 word of the year, defined as "digital content of low quality that is produced usually in quantity by means of artificial intelligence."

The teaser for On This Day... 1776, an American Revolution series from Darren Aronofsky's AI studio Primordial Soup with Time Studios and Google DeepMind, drew online mockery for garbled text and faces, and Futurism called it "straight slop."

Viewers called an AI-generated cartoon flashback in the Milano Cortina 2026 Winter Olympics opening ceremony "AI slop" and said it took work from animators, and TheWrap reported no response from the broadcaster or the organizers.

WIT Studio, the Tokyo animation studio, apologized after fans spotted generative AI in background shots of the Ascendance of a Bookworm Part 3 opening and promised to redraw them and replace the opening on streaming and disc.

Roku launched Fairground AI, a free 24-hour channel of AI-generated programs from the startup Fairground, and The A.V. Club's review found weightless movement, plasticky skin and incoherence.

Warner Bros. deleted a Supergirl behind-the-scenes video after clips appeared to show director Craig Gillespie looking at AI-generated concept art, though Futurism said the AI use could not be confirmed and the studio made no statement.

Futurism called Merry the Christmas Witch, an AI-made holiday film with no distributor from Tom Walsh, chief executive of TW3 Entertainment, "obvious slop," while Walsh says AI let him make it for $1.5 million instead of about $80 million.

The producers of the BBC's Blue Planet III say the series deliberately includes cloudier, less polished footage so that viewers can trust its animals are not AI-generated.

Russell T Davies, the Doctor Who writer whose next production for the BBC starts in 2027, called BBC Director-General Matt Brittin's praise for an unofficial AI Doctor Who episode "a clumsy thing to say in an industry where he's got writers and artists."

In music, Spotify, grading its own catalog, says it removed more than 50 million slop and spam tracks from recommendations because listeners skipped or rejected them, and that it wants to surface authentic, human artists.

A survey of nearly 2,000 people by Alvarez and Marsal, a consultancy that sells AI advisory services, found 38 percent neutral and 13 percent excited about seeing a film whose script, acting and music were all AI-generated.

Futurism, which routinely mocks AI films, wrote that slop may be too harsh a label for Fable's Gods Don't Give Gifts, the fully AI-generated feature opening in US cinemas on 4 December 2026.

A Bluesky post by the actor Rob Delaney praising Duncan Jones's Rogue Trooper as motion capture and computer animation "(no AI!)" drew 1,038 likes and 20 replies.

A Bluesky post urging people not to see Gods Don't Give Gifts "even ironically" because "AI slop is garbage" drew 128 likes and 2 replies.

Look for critics' reviews and audience scores of Gods Don't Give Gifts after its 4 December 2026 US opening, as the first wide test of whether reviewers call a fully AI-generated feature slop.

Watch the reception of Blue Planet III when it airs, and whether other documentary or natural history makers start advertising footage as not AI-generated.

TikTok added a control that lets users adjust how much AI-generated content appears in their For You feed and began testing invisible watermarks to label AI-made videos.

YouTube terminated Screen Culture and KH Studio, two channels of AI-made fake movie trailers with well over a billion views, saying that after being readmitted to its Partner Program they "reverted to clear violations of our spam and misleading metadata policies."

Kapwing, an online video editing company, found that 104 of the first 500 YouTube Shorts shown to a new account, 21 percent, were AI-generated, and that one AI channel, Bandar Apna Dost, had 2.07 billion views.

YouTube's chief executive Neal Mohan wrote that more than 1 million channels used YouTube's AI creation tools daily on average in December 2025.

Runway, an AI video company promoting its own tools, held its fourth annual AI Festival with gala screenings in New York on 11 June 2026 and Los Angeles on 18 June 2026, offering more than $135,000 in prizes.

The director Neill Blomkamp released Nightborne, a 13-minute science fiction horror short generated entirely with ByteDance's Seedance 2.0, on YouTube on 20 July 2026, calling it a first step toward an AI-generated feature.

The trailer for Gods Don't Give Gifts passed 15 million views after its 6 October 2026 release, and the free AI shorts of its maker, the YouTuber Zack London, have drawn more than 1 billion views on a channel with 1.4 million subscribers.

Grok Imagine, the AI video tool from Elon Musk's xAI, posted a seven-minute first episode of an AI-made Odyssey series on X on 23 October 2026, made with Wonder Studios and the director Dev Walker.

Warner Bros. Discovery and Sony Pictures, which own the franchises being imitated, reportedly asked YouTube to send them the ad revenue from AI fake trailers instead of having the videos taken down.

YouTube's chief executive Neal Mohan, speaking for the platform that earns the ad revenue, wrote that it is building on its spam and clickbait systems to reduce "the spread of low quality, repetitive content."

Anthony Ha of TechCrunch wrote that Grok's Odyssey episode lacks "any sense of artistry, inspiration, or human interest."

Adam Mosseri, head of Instagram, which hosts the clips, said he does not "think it's practical for us to try and ban" AI content and that working out what to do is "a tricky problem."

A TikTok clip of the Walking Dead cast at a disco night, captioned as made with Kling 4.0, drew 18,634 likes, 503 comments and 275,959 views.

An Instagram reel recreating the police lineup from The Usual Suspects with AI versions of 1980s and 1990s action stars drew 17,148 likes, 414 comments and 309,134 views.

The Ballon d'Or Heist, a YouTube short billed as an AI football movie, drew 5,579 likes, 331 comments and 253,589 views.

The first episode of an AI live-action Dragon Ball Z fan film on YouTube drew 1,600 likes, 64 comments and 194,040 views.

A full-length AI parody of The Lord of the Rings starring Donald Trump drew 1,161 likes, 194 comments and 55,946 views on YouTube.

Look for the full-length Grok Imagine film of The Odyssey that Elon Musk promised by the end of 2026, and any view count xAI gives for it.

Look for whether Zack London's YouTube audience follows Gods Don't Give Gifts into US cinemas when it opens on 4 December 2026, the first test of whether free AI-shorts reach turns into paid tickets.

DataEye, a Chinese market data firm, put China's AI drama and animation market above RMB 22 billion ($3.29 billion) in the first five months of 2026, with more than 600 million AI short drama users in the first half and a full-year projection of RMB 40 billion.

China's industry released 367,000 micro-dramas in the first half of 2026, more than 74 percent of them AI-generated, an average of about 1,500 new AI dramas a day, according to figures from the People's Daily app.

China's micro-drama rules from the National Radio and Television Administration, the broadcast regulator, took effect on 1 September 2026, bringing episodes under 20 minutes into one national framework and requiring a prominent AI disclosure label in every episode of an AI-generated work.

Holywater, the Ukrainian company behind the My Drama and MyMuse apps, reported from its own app data that MyMuse, its AI-generated series app, converts 23.6 percent of users into paying subscribers against 23.9 percent for live-action shows on My Drama.

Dashverse, an AI microdrama company, says its subscription-only app DashReels has 10 million monthly and 1 million daily active users with 68 percent retention on its AI content, and that it released 100 AI series in June 2026 and licenses work to TikTok and ReelShort.

The China Netcasting Services Association, the online video industry body, counted about 128,000 micro-dramas released in the first quarter of 2026, more than 95 percent of them AI-generated, and put the AI short drama market at CNY24 billion with more than 280 million users.

Amazon MGM Studios and Amazon Web Services launched a GenAI Creators' Fund and ordered three animated series made with generative AI tools, from BuzzFeed Studios, Albie Hecht and Jorge R. Gutierrez, all slated for Prime Video at a future date.

iQiyi, the Chinese streaming service, opened its platform to outside creators with a commercial AI production tool of nearly 70 AI agents, revenue sharing linked to performance and a 20 percent subsidy for AI-generated mid-form series through the end of 2026.

ReelShort, the largest microdrama app, took roughly $1.2 billion in gross consumer spending in 2025 according to Appfigures, an app analytics firm, in a format that sells episodes for tokens or a $20 weekly pass.

Guy Shimoni, chief executive of Shortical, an Israeli microdrama company, speaking for his own business, says it has doubled since it began rolling out AI content and is set to reach $100 million in revenue by the end of 2026, and that traditional players now come knocking because they see the revenues.

Gong Yu, founder and chief executive of iQiyi, the Chinese streaming service, argues that cheaper AI production will grow the creators and content on his own platform at least a hundredfold, citing the Jevons paradox.

Tanveer Ali of Dashverse, an AI microdrama company selling its own output, puts an hour of AI content at $20,000 to $30,000, against about $150,000 for a live-action vertical series, and Dashverse plans to reach 1,000 AI series a month by the end of 2026.

Of 44,200 new AI dramas added to Douyin, the Chinese version of TikTok, in April 2026, only 267 or 0.6 percent passed 100 million views according to DataEye, and industry insiders warn that inexperienced newcomers face a 99 percent risk of losses or bankruptcy.

Dong Min, secretary-general of the China Video Industry Association, said after the first AI-generated long drama aired in prime time that the business model of AI-generated long dramas has not yet been validated and that strong ratings do not mean every AI drama will turn a profit.

Jorge Gutierrez, the animator behind The Book of Life, dropped out of Amazon's generative AI program and said he would not make Punky Duck two days after Amazon announced it, following a wave of online hostility.

Tencent's new Lupaopao, a free short-drama service inside WeChat, offers AI-generated series alongside live-action and animated ones, and ByteDance's free, advertising-supported Hongguo reached 168 million daily active users in July 2026, more than Tencent Video, iQiyi, Youku and Mango TV combined, according to QuestMobile figures cited by Nomura.

Part one of Double Force Miami, an AI-made action movie posted free in parts on TikTok, drew 739,688 likes, 5,339 comments and 7,456,253 views, the biggest post in the month's AI movie talk, attention that shows reach for AI video and not a paying audience.

Episode one of Created, a creator-made AI series about machines turning on humanity, drew 1,685 likes, 72 comments and 38,038 views on TikTok.

An X post promoting a tutorial on making a complete AI movie for free with three prompts for YouTube automation drew 912 likes and 14 replies.

A Bluesky post arguing that the distance to a financially successful narrative AI movie with AI actors is vast drew 101 likes and 6 replies.

Look for the release of Inevitable, the feature-length AI microdrama from Shortical, an Israeli microdrama company, that stars an AI-generated version of the actor Aki Avni, and for any viewing or revenue figure Shortical gives for it.

Look for whether Shortical reports reaching the $100 million in 2026 revenue its chief executive Guy Shimoni says AI content puts it on track for, and how much of it comes from AI series.

Look for DataEye's full-year figure for China's AI drama and animation market against its RMB 40 billion projection, and for any split between paid and ad revenue.