Deepfake Liability Act
- Bill Number
- H.R. 6334
- Origin Chamber
- House
- Congress
- 119th Congress, Session 1
- Policy Area
- Science, Technology, Communications
- Status
- Introduced
- Latest Action
- 2025-12-01: Referred to the House Committee on Energy and Commerce.
- Last Updated
- 2025-12-10T17:47:46Z
AI-Generated Summary
Purpose
The Deepfake Liability Act aims to protect individuals from cyberstalking and violations of intimate privacy online, particularly those involving non-consensual sharing of intimate images or sexually explicit digital forgeries (like AI-generated deepfakes). It updates existing laws to hold online platforms more accountable while balancing free speech protections.
Key Provisions
- Amendments to Section 230 of the Communications Act of 1934:
- Introduces a "duty of care" for interactive computer services (e.g., social media or websites hosting user content). Platforms lose certain legal protections (immunity from lawsuits) unless they implement reasonable processes to prevent and address cyberstalking and intimate privacy violations. These processes must include:
- Prevention measures where feasible.
- A clear system for users to report and request removal of violating content, aligned with the TAKE IT DOWN Act.
- Data logging to preserve evidence for legal cases, without using it for other purposes.
- Removal or blocking of content ruled unlawful by a court.
- Additional requirements set by the Federal Communications Commission (FCC).
- Defines key terms:
- Cyberstalking: Repeated online actions targeting a specific person, causing significant emotional distress or fear of harm, which a reasonable person would also find distressing.
- Intimate privacy violation: Sharing intimate visual depictions (e.g., private photos or videos) or sexually explicit digital forgeries without the depicted person's consent.
- Sexually explicit digital forgery: AI-altered or created images/videos of identifiable people that appear real and depict explicit sexual content without consent.
- Expands the definition of an "information content provider" (someone who can be held liable for content) to include those who create or develop content using generative AI models, even if they solicit or encourage it.
- Amendments to the TAKE IT DOWN Act:
- Expands coverage to include both intimate privacy violations and cyberstalking content.
- Requires "covered platforms" (public websites, apps, or online services, excluding email, messaging, data storage, or broadband providers) to:
- Set up an easy-to-use reporting process for affected individuals (or their representatives) to notify the platform and request removal. Reports must include proof of identity, location of content, and a sworn statement of good faith.
- Provide clear, prominent notices about the process on the platform.
- Remove reported violating content within 48 hours and search for/delete identical copies.
- Offers platforms legal protection (limited liability) for good-faith removals, even if the content is later found not to violate the law.
- Strengthens criminal prohibitions by focusing on sexually explicit digital forgeries and removing some prior exceptions (e.g., for certain educational or artistic uses).
- Defines terms like "covered individual" (the targeted or depicted person) and "covered platform."
- General Provisions:
- The Federal Trade Commission (FTC), with input from the FCC and Attorney General, must issue implementing regulations within 180 days of enactment.
- Changes apply only to content posted after the law takes effect.
- Includes a safeguard stating the law does not violate First Amendment free speech rights.
Significant Changes to Existing Law
- Section 230: Previously provided broad immunity to platforms for user-generated content. This bill carves out an exception, conditioning immunity on proactive compliance with anti-cyberstalking and privacy measures, marking a shift toward greater platform responsibility.
- TAKE IT DOWN Act: Originally focused on non-consensual intimate images; now broadens to cyberstalking and deepfakes, adds a strict 48-hour removal deadline (replacing looser timelines), eliminates some exceptions to criminal bans, and requires more detailed reporting processes.
- Introduces new criminal focus on AI-generated explicit forgeries and ties platform duties to evidence preservation for lawsuits.
Potential Impacts
- On Citizens: Enhances online safety for victims of cyberstalking or non-consensual image sharing by speeding up content removal and preserving evidence for legal action, potentially reducing emotional harm and encouraging reporting.
- On Government Agencies: Increases workload for the FTC (leading regulations), FCC (setting additional requirements), and Department of Justice (enforcing criminal provisions), requiring coordination on tech enforcement.
- On Online Platforms: Imposes operational costs for compliance (e.g., reporting systems, data logging), with risk of lawsuits if they fail to meet the duty of care; may lead to more content moderation.
- On International Relations: No direct impacts mentioned, though it could influence global standards for handling deepfakes on U.S.-based platforms used worldwide.
Main Stakeholders Affected
- Victims and Individuals: People targeted by cyberstalking or whose intimate images are shared without consent, gaining easier recourse.
- Online Platforms and Tech Companies: "Covered platforms" like social media sites, facing new compliance duties and potential liability.
- Content Creators and AI Developers: Those using generative AI tools, now more likely to be treated as responsible for harmful content they produce or encourage.
- Government Entities: FTC, FCC, and law enforcement agencies, tasked with regulation and prosecution.
- Legal and Advocacy Groups: Organizations focused on privacy, free speech, and tech accountability, which may influence or challenge implementation.
Notable Legal, Constitutional, or Political Implications
- Legal: Reduces platform immunity under Section 230 for specific harms, potentially increasing civil lawsuits against non-compliant sites and strengthening criminal penalties for deepfake creation/sharing; promotes evidence preservation to aid prosecutions.
- Constitutional: Explicitly protects First Amendment rights, avoiding broad censorship, but could face challenges if removal processes are seen as overreach in moderating speech.
- Political: Addresses rising concerns over AI-driven harms like deepfakes (e.g., in elections or personal attacks), signaling bipartisan interest in tech regulation; may set precedent for future limits on Section 230 amid debates on online responsibility.
This summary was generated by AI and may contain inaccuracies. Refer to the official source document for the authoritative text.
Sponsor
Rep. Auchincloss, Jake [D-MA-4]
Cosponsors (1)
Recent Actions
- 2025-12-01: Referred to the House Committee on Energy and Commerce.
- 2025-12-01: Introduced in House
- 2025-12-01: Introduced in House
Bill Versions
- Deepfake Liability Act — issued 2025-12-01 — PDF (16 pages)