Business And Startup

Meta launches Muse Image: anyone can now generate AI pictures of you from Instagram

Meta's newly launched Muse Image tool lets anyone generate AI images using a public Instagram account's likeness, without notifying the account owner.

Meta launched its AI image generation model, Muse Image, earlier this week, integrating it directly with public Instagram accounts. Once live, anyone can tag another person’s Instagram handle in a prompt and use Meta AI to generate an image based on that account’s likeness.

According to Meta’s own policy, the account owner is not notified when someone does this, making it difficult to know when or how a person’s likeness has been used. Meta has described Muse Image as a way to make AI image generation ‘more personal’ by letting people reference public Instagram accounts in prompts.

Users can opt out, though the setting is not prominent. On Instagram, the path is Profile, then Menu, then Sharing and reuse; under ‘Allow people to reuse your content on Instagram and with AI features at Meta,’ the Posts and Reels toggles can be switched off. The wording may vary by app version, and the feature is still rolling out, starting in the US.

Opting out only prevents future image generation using an account’s likeness — it does not remove AI images already created before the setting was changed, and those remain in circulation. Switching an Instagram account to private is currently the only way to prevent the feature from being used on a profile at all.

Cybersecurity firm Malwarebytes has warned that the tool could be used for impersonation, scams or other abuse, noting in a blog post that public Instagram photos were already being harvested by attackers to create deepfakes for identity verification fraud before Muse Image existed.

Meta’s AI systems have faced other security scrutiny this year: researchers disclosed a ‘confused deputy’ flaw in the company’s AI support chatbot that allowed it to make account changes, including changing email addresses and resetting passwords, without adequately verifying who it was talking to. Enabling multi-factor authentication appeared to mitigate that issue.

Leave a Reply

Your email address will not be published. Required fields are marked *