Creatives and artists long treated generative AI as a straightforward threat to their work. That posture is shifting: many are now negotiating terms with the technology — and, in some cases, regaining control over how their content is used.
The distance traveled since 2022 is considerable. Early adopters were sometimes shamed for using image generators, and artists who won competitions with AI-assisted work faced public backlash. Meanwhile, fear spread among artists and writers as platforms like OpenAI’s DALL-E and Stability AI’s Stable Diffusion appeared to reproduce styles and content without attribution. That fear translated into litigation that has defined the era: Getty Images sued Stability AI over training on its image library; a class action by book authors against Anthropic ended in a $1.5 billion settlement, the largest publicly reported copyright recovery to date; and Hollywood studios including Disney, Universal and Warner Bros. have sued Midjourney, alleging the service created unauthorized copies of their copyrighted characters.
Two sides
Inside the entertainment industry, positions remain sharply divided. Speaking at the AI Summit in New York on December 10, BET+ executive vice president and general manager Jason Harvey summed up the prevailing mood: the industry expects AI either to save Hollywood or to burn it down, depending on which side of the fence one stands.
Opponents of AI in creative work argue that the tools strip out the human element and yield uninspired output — and that genuine artists simply do not need AI to make good work. Advocates counter that the technology can serve rather than replace human creators; one documentary-platform founder argued at the summit that AI agents could even become protective infrastructure for artists, monitoring whether a person’s name or likeness is being used without consent.
A middle ground
Between the poles, a pragmatic center is forming. Joe Livecchi, CEO of Wrigley Media Group, articulated it plainly: “I definitely think you need the human touch in anything you’re creating.” Audiences, he argues, want something new — not an average of a thousand existing things.
Livecchi is no opponent of the technology. His entire team works with agentic tools; he personally uses Microsoft Copilot and ChatGPT, while his staff experiments with Google’s Veo video generator and Nano Banana image model as well as xAI’s Grok. His assessment is that low-quality AI content is proliferating and hard to keep pace with, but that the expanding toolbox is ultimately good news — running a modern media business without AI in some form is becoming nearly impossible.
He also frames the tools as an opportunity rather than a gate: where earlier generations faced steep barriers to entry in film and media, anyone can now make something, publish it online and potentially reach millions. At the same time, he supports standards for how models are trained on intellectual property, arguing that IP protection is what lets a creator eventually get paid for work that took real time to make.
Revenue-sharing models
Compensation structures are beginning to emerge. Getty Images maintains a revenue-sharing model for contributors whose content trained its AI tools, and OpenAI has struck publishing partnerships with news organizations including the Associated Press and The Atlantic.
Smaller companies are exploring similar arrangements. Pinaki Saha, founder and CEO of digital product engineering firm Anshar Labs, presented TechFlix, an AI platform that converts unstructured video into short, avatar-based learning videos, built on open models including Meta’s Llama and DeepSeek. With growing interest from small e-commerce sellers, the company plans features that let partners serve content on behalf of customers and share in the resulting revenue. Saha rejects the idea that avatar-led video removes the human element: the process begins with human creativity — defining the content and its direction — and in his view humans reposition themselves in the loop rather than leaving it.
The case for using the tools
For working creatives outside the industry’s power centers, the calculus is often practical. Anthony Bailey, pastor of a New York congregation, sees tools like TechFlix as a way to turn sermon notes into visuals that help his community engage — while drawing a firm personal line at using AI to write sermons themselves. His advice to fellow creators is to treat AI as a tool to extend their own creativity rather than something to fear, since the ideas still originate with the person using it.
That instinct — adopt the tools, keep the human at the center — echoes concerns being voiced well beyond the creative industries, including recent warnings from Pope Leo XIV about AI and human agency.
What remains unsettled
The optimism on display at industry conferences deserves some counterweight. The major copyright cases are far from resolved: the Midjourney litigation is still in discovery, the legal boundaries of fair use for AI training remain unsettled in US courts, and the Anthropic settlement — while enormous — addressed pirated acquisition of books rather than the underlying question of whether training on lawfully obtained work requires a license. Revenue-sharing programs currently cover a small fraction of the creators whose work has influenced commercial models, and their payout terms are largely opaque. And the economic effects on working artists — particularly in illustration, stock photography and entry-level production roles — are still being measured. The creative industries are negotiating with generative AI, not yet at peace with it.
