Muse Image ranks No. 2 on Arena: for text-to-image, single-image editing, and multi-image editing by human preference Elo as of July 5, 2026.
Muse Video ranks No. 3: on Arena for text-to-video, with native audio support and competitive visual fidelity.
Agentic capabilities: Muse Image uses search and coding tools, self-refines its own outputs, and scales performance with test-time compute — similar to how language models improve with more reasoning.
Content Seal: An invisible watermarking system embeds provenance signals that survive cropping, compression, resizing, and screenshots.
Instagram integration is opt-out by default: Public accounts are automatically included as source material for AI generations unless users manually disable the setting.
Hollywood backlash: CAA and SAG-AFTRA have publicly condemned the rollout, calling for an opt-in model instead of opt-out.
No notification policy: Users whose photos are used in AI generations will not be notified, according to Meta's help center.
Unlike traditional AI image generators that map prompts directly to outputs, Muse Image operates as an agent. It can invoke search tools to ground generations in real-time facts, write and execute code for accurate plots and QR codes, and self-reflect on its own outputs to improve them. This self-refinement behavior emerged naturally during reinforcement learning training — the model learned that checking and improving its work led to better results.
The model also benefits from test-time compute scaling. The more "thinking" Muse Image does during inference, the better its outputs become. This scaling spans both text tokens (for reasoning) and visual tokens (for generation), and the company reports an approximately log-linear relationship between compute and quality.
The core of the backlash centers on how Meta has integrated Muse Image with Instagram. By default, public Instagram profiles are opted in as training material for AI generations. Users can tag a public account in a prompt, and Muse Image will generate new images using that person's likeness from their public photo library.
Key concerns raised by critics include:
Opt-out, not opt-in: Users must manually navigate to Settings → Sharing and Reuse → disable "Allow people to use your content on Instagram and with AI features on Meta."
No deletion of existing generations: Even after opting out, previously generated AI images using a user's content will not be removed.
No notifications: Users will not be informed when their likeness is used in an AI generation.
Persistent watermarks: While Content Seal watermarks are embedded, they are invisible to the naked eye and require a detection tool to verify.
CAA (Creative Artists Agency) issued a strong statement: "We call on Meta to make protection the default on Muse Image AI, not the exception, and enable individuals to opt-in if they want to allow usage of their image or likeness."
SAG-AFTRA called the rollout "an utter miscalculation of public sentiment" and demanded a clear opt-in system.
Lori Fena, former chair of the Electronic Frontier Foundation, described the approach as a "dark pattern of AI overreach" that undermines trust in the AI ecosystem.
Meta responded by stating that private accounts and users under 18 are automatically excluded, and that adult users with public accounts can opt out with "a couple clicks."
How do I opt out of having my Instagram photos used for AI generation?
Go to your Instagram profile, tap the three lines in the top-right corner, scroll to "Sharing and reuse," and toggle off the option labeled "Allow people to use your content on Instagram and with AI features on Meta." You can disable it for both Posts and Reels.
Will existing AI images of me be deleted if I opt out?
No. Meta has stated that already generated AI images made with your content will not be removed after you opt out.
Will I be notified if someone uses my photos to create an AI image?
No. According to Meta's help center, "You will not be notified about content created using AI features at Meta."
What is Content Seal?
Content Seal is Meta's invisible watermarking system embedded in images generated by Muse Image. The watermark survives cropping, compression, resizing, and screenshots. Meta has previewed a detection tool to verify whether an image carries a Content Seal watermark.
When will Muse Video be available?
Muse Video is currently in preview and is "coming soon to creators and Meta AI," according to Meta's announcement.
Check your Instagram settings now: If you have a public account, navigate to Settings → Sharing and Reuse and disable the AI content toggle to prevent your photos from being used in AI generations.
Understand the watermark: Images generated by Muse Image carry Content Seal watermarks. Use Meta's detection tool to verify if an image was AI-generated.
Who this affects most: Public figures, influencers, actors, and content creators with public Instagram accounts are most at risk of having their likeness used without consent.
How to prepare: Consider switching to a private account if you are concerned about AI usage. Monitor Meta's announcements for updates to their opt-out policies.
Stay informed: The conversation around AI rights and consent is evolving rapidly. Follow industry updates from organizations like SAG-AFTRA and CAA for policy changes.
Do you think Meta's opt-out approach is fair, or should AI training on user content require explicit consent? Share your thoughts with others who are navigating this new AI landscape.
*Share this with others who need to stay ahead of this trend!*
Share on: Twitter | LinkedIn | Reddit
What's your take? Do you think this trend will last, or will regulatory pressure force Meta to switch to an opt-in model? Let us know!
On July 24, 2026, a coalition of 25 major technology companies — including **Nvidia, Microsoft, Meta, Palantir, Google, IBM, and Hugging Fac...
On Saturday, July 25, 2026, OpenAI experienced a significant global outage that left thousands of users unable to access ChatGPT, the compan...
A 55-year-old Florida pastor, Scott Winters, has filed a lawsuit against OpenAI and its founder Sam Altman, alleging that medical advice fro...
On July 20, 2026, a federal court in San Francisco gave final approval to what is now the largest copyright class action settlement in US hi...
⚠ Disclaimer: Yanuki provides article summaries and links for reference only. Yanuki does not endorse, verify, or guarantee the accuracy of third-party sources. Please review original sources and verify information independently. Managed by the Yanuki Data Engine. Full Disclaimer