'Terminator' director James Cameron flip-flops on AI, says Hollywood is 'looking at it all wrong'
James Cameron's stance on artificial intelligence has evolved over the past few years, and he feels Hollywood needs to embrace it in a few different ways.
Cameron joined the board of directors for Stability AI last year, explaining his decision on the "Boz to the Future" podcast last week.
"The goal was to understand the space, to understand what's on the minds of the developers," he said. "What are they targeting? What's their development cycle? How much resources you have to throw at it to create a new model that does a purpose-built thing, and my goal was to try to integrate it into a VFX workflow."
He continued by saying the shift to AI is a necessary one.
"And it's not just hypothetical. We have to. If we want to continue to see the kinds of movies that I've always loved and that I like to make and that I will go to see – 'Dune,' 'Dune: Part Two' or one of my films or big effects-heavy, CG-heavy films – we've got to figure out how to cut the cost of that in half.
"Now that's not about laying off half the staff at a VFX company. That's about doubling their speed to completion on a given shot, so your cadence is faster and your throughput cycle is faster, and artists get to move on and do other cool things and then other cool things, right? That's my sort of vision for that."
Cameron doesn't think films are ultimately "a big target" for companies like OpenAI.
"Their goal is not to make GenAI movies. We're a little wart on their butt," he laughed.
"Movies are just a tiny application, tiny use case, and it's too tiny right now. That's the problem. It's going to be smaller, boutique-type GenAI developer groups that I can get the attention of and say, 'Hey, I've got a problem here, it's called rotoscope,' or it's called this or that. How do I outpaint, upscale, whatever needs to be done in a workflow that already exists that's CG-based? And there are answers."
Rotoscoping is a task involving drawing or tracing live-action footage frame by frame.
Cameron said people don't get into filmmaking and VFX to do "the underwater needlepoint of roto[scoping]. And there are a lot of tasks like that," like changing pieces frame by frame for motion capture that he feels AI can make quicker and more cost effective."
"It's all common sense, right? All you have to do is maybe something like feed in a couple million images of the real world and have a dumb algorithm" figure it out, he said.
Cameron previously denounced the use of AI in 2023, citing his 1984 film "The Terminator" as a warning.
"I think the weaponization of AI is the biggest danger," he told Canadian CTV at the time.
"I think that we will get into the equivalent of a nuclear arms race with AI, and if we don't build it, the other guys are for sure going to build it, and so then it'll escalate," Cameron said.
He continued, "You could imagine an AI in a combat theater, the whole thing just being fought by the computers at a speed humans can no longer intercede, and you have no ability to deescalate."
"I warned you guys in 1984, and you didn't listen," Cameron said, referring to the plot of the film and its sequels featuring a sentient AI destroying humanity.
At the time, Cameron said he wasn't concerned about AI replacing creatives, saying, "It's never an issue of who wrote it, it's a question of is it a good story?"
"I just don't personally believe that a disembodied mind that's just regurgitating what other embodied minds have said – about the life that they've had, about love, about lying, about fear, about mortality – and just put it all together into a word salad and then regurgitate it … I don't believe that have something that's going to move an audience," he said.
On the "Boz to the Future" podcast, Cameron downplayed the negative feelings much of the entertainment industry has shared about generative AI.
"A lot of the hesitation in Hollywood and entertainment in general are issues of the source material for the training data and who deserves what and copyright protection and all that sort of thing, and I think people are looking at it all wrong, personally," he said.
He explained that everyone's brains already work essentially like an AI: "You're creating a model as you go through life to process quickly through that model every new situation that comes on. You don't say, 'Wait a second, I'm going to go back.' … It doesn't work that way."
"My point is, as a screenwriter, as a filmmaker, if I exactly copy 'Star Wars,' I'll get sued. Actually, I won't even get that far. Everybody will say it's too much like 'Star Wars,' we're going to get sued. I won't even get the money. As a screenwriter, you have a sort of built-in ethical filter that says, 'I know my sources, I know what I like. I know what I'm emulating. I also know I have to move it far enough away that it's my own independent creation.'"
Cameron agreed that there needs to be management of AI "from a legal perspective," but he thinks the focus should be on what the output is and if it's being monetized versus the input.
"You can't control my input. You can't tell me what to view and what to see and where to go. My input is whatever I choose it to be and whatever has accumulated throughout my life. My output should be judged on whether it's too close or plagiaristic," he said, acknowledging that people are considering "the monetization possibilities of libraries and all that sort of thing."
Some companies are already allowing AI companies access to their back catalogs.
Lionsgate, the studio behind films like "The Hunger Games" and "Twilight," partnered with AI research company Runway last September for the "creation and training of a new AI model, customized to Lionsgate's proprietary portfolio of film and television content," according to a statement from the company at the time.
Cameron admitted he doesn't love the idea of someone using an AI to mimic him.
"I think we should discourage the text prompt that says 'in the style of James Cameron' or in the style of 'Zac Snyder' or 'in the style of.' Make up your own style. That makes me a little bit queasy," he said. "But on the other hand, it's kind of cool to have an 'in the style of' as part of a text prompt."
The trend of generating images in specific styles has taken off on social media, creating controversy over copyright and originality.
Most recently, people were regularly sharing images generated in the style of Japanese animation company Studio Ghibli, further fueling the debate.
But Cameron didn't see an issue.
"How is Studio Ghibli harmed by fan art?" Cameron asked on the podcast, noting fan art has been around "forever."
"You can't stop fan art, you don't want to stop fan art. It's an expression. It's them echoing back something they responded to, it makes them feel participatory in that universe. These things should be encouraged."
However, he drew the line at financial gain.
"If somebody goes out and tries to make money off it in a way that harms Studio Ghibli and is provable, I think that's where you have to draw the line."
He continued, "How is that any different than selling a T-shirt and on the T-shirt is an image of Luke Skywalker? You can't do that. LucasArts can do that. You can't do that. That's piracy. It's going to get shut down. However, if you just wanted to wear one and express your fandom for 'Star Wars,' you're not harming anything."
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Yahoo
28 minutes ago
- Yahoo
CoreWeave (CRWV) Jumps 8% on Rosy Growth Prospects
CoreWeave, Inc. (NASDAQ:CRWV) is one of the CoreWeave grew its share prices by 7.99 percent on Friday to hit a new all-time high, as investors continued to load up positions amid rosy growth prospects. At intraday trading, shares of CoreWeave, Inc. (NASDAQ:CRWV) jumped as high as 10 percent to hit $187, before paring gains to finish the day at $170, as investors appeared to have flocked to the booming Artificial Intelligence industry for safety to mitigate risks from the ongoing geopolitical tensions. In recent news, CoreWeave, Inc. (NASDAQ:CRWV) announced a record-breaking performance from using 2,496 of Nvidia Corp.'s latest Grace Blackwell Chips on its AI-optimized cloud platform, making its submission the largest-ever benchmarked under MLPerf. In March this year, CoreWeave, Inc. (NASDAQ:CRWV) bagged an $11.9-billion deal with OpenAI and welcomed it as a new investor through the sale of $350 million CRWV stocks to the latter. Last month, OpenAI upsized the deal with another $4 billion worth of contract. A close-up of a network administrator's hands working on a cloud computing server. Last month, it was tapped by Aston Martin Aramco as its official AI cloud computing partner, where it will provide AI-accelerated engineering opportunities to support car design efficiency. While we acknowledge the potential of CRWV as an investment, our conviction lies in the belief that some AI stocks hold greater promise for delivering higher returns and have limited downside risk. If you are looking for an extremely cheap AI stock that is also a major beneficiary of Trump tariffs and onshoring, see our free report on the best short-term AI stock. READ NEXT: 20 Best AI Stocks To Buy Now and 30 Best Stocks to Buy Now According to Billionaires. Disclosure: None. This article is originally published at Insider Monkey.
Yahoo
3 hours ago
- Yahoo
Palantir, Meta, OpenAI, And Thinking Machines Just Had Their Executives Sworn Into The US Army Reserve
Four top tech executives have joined the U.S. Army Reserve as lieutenant colonels, skipping basic training and stepping directly into roles aimed at helping modernize the military. The initiative is part of a broader push by the Army to bring in private-sector innovation and reshape how the service approaches technology, talent, and modernization. The executives—Palantir (NYSE:PLTR) Chief Technology Officer Shyam Sankar, Meta (NASDAQ:META) CTO Andrew Bosworth, OpenAI Chief Product Officer Kevin Weil, and advisor at Thinking Machines Lab and former OpenAI Chief Research Officer Bob McGrew — will serve in a new unit called Detachment 201, also known as the Army's Executive Innovation Corps. Don't Miss: Maker of the $60,000 foldable home has 3 factory buildings, 600+ houses built, and big plans to solve housing — Peter Thiel turned $1,700 into $5 billion—now accredited investors are eyeing this software company with similar breakout potential. Learn how you can 'Detachment 201 is being created to bring in tech innovation executives to help the Army ... on broader conceptual things like talent management, how do we bring in tech-focused people into the ranks of the military, and then, how do we train them,' Army Chief of Staff spokesperson Col. Dave Butler, told Breaking Defense on June 13. Unlike traditional recruits, these executives will not attend boot camp. Instead, they will go through an express training program that covers marksmanship, physical fitness, Army history, and protocols. They will be expected to serve about 120 hours per year and pass annual fitness tests. 'You could think of it as a pilot' of a lighter version of basic training, Butler told Business Insider. The detachment's name, 201, references the HTTP status code indicating a newly created resource—a fitting metaphor for a new kind of Army asset. Trending: Maximize saving for your retirement and cut down on taxes: . According to an Army statement, the new officers will work on 'targeted projects to help guide rapid and scalable tech solutions to complex problems.' Their advisory roles will include input on AI-powered military systems and optimization tools for soldier fitness. However, safeguards will be in place to avoid conflicts of interest with their current or former employers. 'We've done this over and over when our nation needed top talent,' Butler told Breaking Defense. 'The difference is we used to do it in wartime. Now we're doing it ahead of wartime so that we can prepare and deter.' This marks another move by the Trump administration to align more closely with Silicon Valley. Palantir, Anduril, and other VC-backed defense tech startups have increasingly become major players in national security. Meta recently partnered with Anduril to develop augmented reality tools and AI systems for military direct commissioning has been used to bring in specialized talent, such as doctors or chaplains, during times of war. This move represents a peacetime shift aimed at long-term transformation. 'Their swearing-in is just the start of a bigger mission to inspire more tech pros to serve without leaving their careers,' the Army statement said. 'Showing the next generation how to make a difference in uniform.' Read Next: How do billionaires pay less in income tax than you?.UNLOCKED: 5 NEW TRADES EVERY WEEK. Click now to get top trade ideas daily, plus unlimited access to cutting-edge tools and strategies to gain an edge in the markets. Get the latest stock analysis from Benzinga? PALANTIR TECHNOLOGIES (PLTR): Free Stock Analysis Report This article Palantir, Meta, OpenAI, And Thinking Machines Just Had Their Executives Sworn Into The US Army Reserve originally appeared on © 2025 Benzinga does not provide investment advice. All rights reserved.
Yahoo
5 hours ago
- Yahoo
Why is AI halllucinating more frequently, and how can we stop it?
When you buy through links on our articles, Future and its syndication partners may earn a commission. The more advanced artificial intelligence (AI) gets, the more it "hallucinates" and provides incorrect and inaccurate information. Research conducted by OpenAI found that its latest and most powerful reasoning models, o3 and o4-mini, hallucinated 33% and 48% of the time, respectively, when tested by OpenAI's PersonQA benchmark. That's more than double the rate of the older o1 model. While o3 delivers more accurate information than its predecessor, it appears to come at the cost of more inaccurate hallucinations. This raises a concern over the accuracy and reliability of large language models (LLMs) such as AI chatbots, said Eleanor Watson, an Institute of Electrical and Electronics Engineers (IEEE) member and AI ethics engineer at Singularity University. "When a system outputs fabricated information — such as invented facts, citations or events — with the same fluency and coherence it uses for accurate content, it risks misleading users in subtle and consequential ways," Watson told Live Science. Related: Cutting-edge AI models from OpenAI and DeepSeek undergo 'complete collapse' when problems get too difficult, study reveals The issue of hallucination highlights the need to carefully assess and supervise the information AI systems produce when using LLMs and reasoning models, experts say. The crux of a reasoning model is that it can handle complex tasks by essentially breaking them down into individual components and coming up with solutions to tackle them. Rather than seeking to kick out answers based on statistical probability, reasoning models come up with strategies to solve a problem, much like how humans think. In order to develop creative, and potentially novel, solutions to problems, AI needs to hallucinate —otherwise it's limited by rigid data its LLM ingests. "It's important to note that hallucination is a feature, not a bug, of AI," Sohrob Kazerounian, an AI researcher at Vectra AI, told Live Science. "To paraphrase a colleague of mine, 'Everything an LLM outputs is a hallucination. It's just that some of those hallucinations are true.' If an AI only generated verbatim outputs that it had seen during training, all of AI would reduce to a massive search problem." "You would only be able to generate computer code that had been written before, find proteins and molecules whose properties had already been studied and described, and answer homework questions that had already previously been asked before. You would not, however, be able to ask the LLM to write the lyrics for a concept album focused on the AI singularity, blending the lyrical stylings of Snoop Dogg and Bob Dylan." In effect, LLMs and the AI systems they power need to hallucinate in order to create, rather than simply serve up existing information. It is similar, conceptually, to the way that humans dream or imagine scenarios when conjuring new ideas. However, AI hallucinations present a problem when it comes to delivering accurate and correct information, especially if users take the information at face value without any checks or oversight. "This is especially problematic in domains where decisions depend on factual precision, like medicine, law or finance," Watson said. "While more advanced models may reduce the frequency of obvious factual mistakes, the issue persists in more subtle forms. Over time, confabulation erodes the perception of AI systems as trustworthy instruments and can produce material harms when unverified content is acted upon." And this problem looks to be exacerbated as AI advances. "As model capabilities improve, errors often become less overt but more difficult to detect," Watson noted. "Fabricated content is increasingly embedded within plausible narratives and coherent reasoning chains. This introduces a particular risk: users may be unaware that errors are present and may treat outputs as definitive when they are not. The problem shifts from filtering out crude errors to identifying subtle distortions that may only reveal themselves under close scrutiny." Kazerounian backed this viewpoint up. "Despite the general belief that the problem of AI hallucination can and will get better over time, it appears that the most recent generation of advanced reasoning models may have actually begun to hallucinate more than their simpler counterparts — and there are no agreed-upon explanations for why this is," he said. The situation is further complicated because it can be very difficult to ascertain how LLMs come up with their answers; a parallel could be drawn here with how we still don't really know, comprehensively, how a human brain works. In a recent essay, Dario Amodei, the CEO of AI company Anthropic, highlighted a lack of understanding in how AIs come up with answers and information. "When a generative AI system does something, like summarize a financial document, we have no idea, at a specific or precise level, why it makes the choices it does — why it chooses certain words over others, or why it occasionally makes a mistake despite usually being accurate," he wrote. The problems caused by AI hallucinating inaccurate information are already very real, Kazerounian noted. "There is no universal, verifiable, way to get an LLM to correctly answer questions being asked about some corpus of data it has access to," he said. "The examples of non-existent hallucinated references, customer-facing chatbots making up company policy, and so on, are now all too common." Both Kazerounian and Watson told Live Science that, ultimately, AI hallucinations may be difficult to eliminate. But there could be ways to mitigate the issue. Watson suggested that "retrieval-augmented generation," which grounds a model's outputs in curated external knowledge sources, could help ensure that AI-produced information is anchored by verifiable data. "Another approach involves introducing structure into the model's reasoning. By prompting it to check its own outputs, compare different perspectives, or follow logical steps, scaffolded reasoning frameworks reduce the risk of unconstrained speculation and improve consistency," Watson, noting this could be aided by training to shape a model to prioritize accuracy, and reinforcement training from human or AI evaluators to encourage an LLM to deliver more disciplined, grounded responses. RELATED STORIES —AI benchmarking platform is helping top companies rig their model performances, study claims —AI can handle tasks twice as complex every few months. What does this exponential growth mean for how we use it? —What is the Turing test? How the rise of generative AI may have broken the famous imitation game "Finally, systems can be designed to recognise their own uncertainty. Rather than defaulting to confident answers, models can be taught to flag when they're unsure or to defer to human judgement when appropriate," Watson added. "While these strategies don't eliminate the risk of confabulation entirely, they offer a practical path forward to make AI outputs more reliable." Given that AI hallucination may be nearly impossible to eliminate, especially in advanced models, Kazerounian concluded that ultimately the information that LLMs produce will need to be treated with the "same skepticism we reserve for human counterparts."
