Scarlett Johansson, R.E.M., and 800+ Artists Push Back on AI Slop and Theft

Scarlett Johansson, R.E.M., and 800+ artists back "Stealing Isn't Innovation" to end unlicensed AI training. They want consent, licensing, and pay, warning of deepfakes.

Categorized in: AI News Creatives
Published on: Jan 23, 2026
Scarlett Johansson, R.E.M., and 800+ Artists Push Back on AI Slop and Theft

Scarlett Johansson, R.E.M., and hundreds more push back on "AI slop" with new campaign

A coalition of 800+ artists - including Scarlett Johansson, R.E.M., and Cate Blanchett - has endorsed Stealing Isn't Innovation, a new effort backed by the Human Artistry Campaign. The message is simple: stop training AI on creative work without permission or payment.

Johansson's history with AI voice concerns makes her presence notable. And yes, Common signed too, despite appearing in high-profile AI ads in the past. The roster signals a broad front across film, music, and performance.

What the campaign argues

In a press release cited by The Verge, the group alleges that "profit-hungry technology companies… have copied a massive amount of creative content online without authorization or payment." It calls that practice "theft at a grand scale," warning it fuels deepfakes, misinformation, and a flood of "AI slop" - low-quality machine-made content swamping feeds.

The goal: enforce licensing, protect opt-outs, and ensure compensation when creative work trains models. Ads will run across major outlets and social platforms to drive the message.

The bigger context

Some publishers have cut deals with AI companies - Disney reportedly signed a $1 billion agreement with OpenAI. Still, many creators argue training requires consent and payment, while AI companies lean on fair-use defenses. The fight is moving from opinion to policy, contracts, and courtrooms.

Why this matters if you're a creative

  • Your portfolio has value twice: the work itself and the data it becomes when scraped. Unlicensed training captures that second value without paying you.
  • "AI slop" dilutes attention and undercuts rates. It's harder to stand out if feeds are saturated with cheap, low-quality imitations.
  • Deepfakes and voice clones threaten reputation and brand deals. Provenance and consent become part of your business strategy.

Practical moves to protect your work

  • Update contracts: add explicit clauses against training use without permission and payment. Require written consent for any AI training, with separate licensing fees.
  • Block known AI crawlers: configure robots.txt to disallow common bots (e.g., GPTBot, CCBot, ClaudeBot) where appropriate. It's not perfect, but it sets terms.
  • Use provenance tools: adopt the C2PA standard or similar content credentials to verify authorship and signal usage terms.
  • Register your copyrights: faster takedowns, stronger leverage, and clearer damages if infringement happens.
  • Monitor misuse: set up alerts for your name, voice, and distinctive visuals; file DMCA takedowns and platform reports for deepfakes or impersonations.
  • Choose ethical tools: prefer models and platforms that document licensed data sources and offer opt-in revenue sharing.
  • Watermark selectively: for public portfolio previews, consider smart watermarking or lower-res versions to reduce high-value scraping.
  • Join the conversation: track the Human Artistry Campaign and similar efforts for templates, legal updates, and collective bargaining power.

If you use AI in your workflow

  • Credit sources, keep process notes, and label AI-assisted assets when clients need transparency.
  • Stick to licensed model providers for commercial work. Ask for dataset documentation and terms in writing.
  • Price for value: AI can speed drafts; your taste and direction are the product. Bill for strategy, not just outputs.

What to watch next

The campaign plans sustained media buys to push for licensing and enforcement. On the legal front, more suits and settlements are likely as lines get tested.

Disclosure: Ziff Davis, Mashable's parent company, filed a lawsuit against OpenAI in April 2025, alleging copyright infringement tied to training and operation of its AI systems.

Bottom line

Consent and compensation are the center of this fight. Creatives don't need to avoid AI - they need fair terms, clear provenance, and tools that respect their work. Set your boundaries now, in your contracts and your tech stack, and keep shipping your best work while the rules catch up.


Get Daily AI News

Your membership also unlocks:

700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)
Advertisement
Stream Watch Guide