U.S. senators are widening the pressure campaign over sexualized deepfakes, asking major social and AI platforms to show how they are preventing nonconsensual, AI-generated intimate imagery from spreading online.
The letter targets X, Meta, Alphabet, Snap, Reddit, and TikTok. It comes as concerns around Grok, X, and other image and video tools have made clear that platform policies alone may not be stopping users from creating or circulating abusive synthetic content.
A broader demand for platform accountability
The senators are asking the companies to prove they have “robust protections and policies” and to explain how they plan to reduce the rise of sexualized deepfakes on their services. The request is not limited to takedown rules after content appears. It also focuses on the systems, incentives, and safeguards around how this material is created, detected, moderated, and monetized.
The letter also tells the companies to preserve all documents and information connected to sexualized, AI-generated images. That includes records related to creation, detection, moderation, monetization, and any related policies.
The core concern is that written bans may not be enough. The senators acknowledge that many companies already have policies against non-consensual intimate imagery and sexual exploitation, and that many AI systems say they block explicit pornography. But the letter argues that users are still finding ways around those barriers, or that the barriers are failing.
Why Grok put the issue back in focus
The letter arrived hours after X said it updated Grok to stop it from making edits of real people in revealing clothing. X also restricted image creation and edits via Grok to paying subscribers. X and xAI are part of the same company.
Media reports had described how easily and often Grok generated sexualized and nude images of women and children. Those reports became part of the senators’ argument that guardrails designed to block nonconsensual sexual imagery may not be reliable in practice.
X responded to the letter by pointing to its announcement about the Grok update. xAI has maintained that it takes action to remove “illegal content on X, including [CSAM] and non-consensual nudity,” but neither the company nor Elon Musk has addressed why Grok was allowed to generate such content in the first place.
The move also follows Elon Musk saying he was “not aware of any naked underage images generated by Grok.” Later on Wednesday, California’s attorney general opened an investigation into xAI’s chatbot amid mounting pressure from governments across the world over the lack of guardrails around Grok.
The problem reaches beyond one company
Although Grok and X have drawn heavy criticism, the senators’ letter reflects a wider concern across social platforms and AI systems. The source article notes that sexualized deepfakes first gained popularity on Reddit, where a page displaying synthetic porn videos of celebrities went viral before Reddit took it down in 2018.
Other platforms have faced related problems. Sexualized deepfakes targeting celebrities and politicians have multiplied on TikTok and YouTube, though they usually originate elsewhere. Meta’s Oversight Board last year called out two cases involving explicit AI images of female public figures. Meta has also had nudify apps selling ads on its services, though it later sued a company called CrushAI.
Snapchat has been connected to multiple reports of kids spreading deepfakes of peers. Telegram is not included on the senators’ list, but the source article notes that it has become notorious for hosting bots built to undress photos of women.
Reddit said it does not allow non-consensual intimate media and does not offer tools capable of making it. A Reddit spokesperson said the platform takes proactive measures to find and remove such content, strictly prohibits nonconsensual intimate media including faked or AI-generated depictions, and bans soliciting this content, sharing links to “nudify” apps, or discussing how to create it on other platforms.
Alphabet, Snap, TikTok, and Meta did not immediately respond to requests for comment.
Deepfakes are becoming harder to contain
The issue is not limited to manipulated sexualized imagery. The source article points out that even AI image and video tools that do not let users “undress” people can still make it easy to generate deepfakes.
Several examples show the range of risks. OpenAI’s Sora 2 reportedly allowed users to generate explicit videos featuring children. Google’s Nano Banana seemingly generated an image showing Charlie Kirk being shot. Racist videos made with Google’s AI video model are gaining millions of views on social media.
The problem becomes more complicated when Chinese image and video generators are included. Many Chinese tech companies and apps, especially those linked to ByteDance, offer easy ways to edit faces, voices, and videos. Those outputs have spread to Western social platforms.
China has stronger synthetic content labeling requirements. The U.S. does not have those requirements at the federal level, leaving users to rely on fragmented platform policies that the source article describes as dubiously enforced.
Lawmakers are still looking for leverage
U.S. lawmakers have already passed some legislation aimed at deepfake pornography, but the impact has been limited. The Take It Down Act became federal law in May and is meant to criminalize the creation and dissemination of nonconsensual, sexualized imagery.
However, the source article notes that provisions in the law make it difficult to hold image-generating platforms accountable because they focus much of the scrutiny on individual users instead.
States are also trying to act. This week, New York governor Kathy Hochu l proposed laws that would require AI-generated content to be labeled as such and would ban nonconsensual deepfakes in specified periods leading up to elections, including depictions of opposition candidates.
The senators’ letter was signed by Senators Lisa Blunt Rochester (D-Del.), Tammy Baldwin (D-Wis.), Richard Blumenthal (D-Conn.), Kirsten Gillibrand (D-NY), Mark Kelly (D-Ariz.), Ben Ray Luján (D-NM), Brian Schatz (D-Hawaii), and Adam Schiff (D-Calif.). Their message is clear: platforms are being asked to show not just that they have rules, but that those rules work when users test the limits of generative AI.