Bill C-16 won’t hold platforms accountable for online harms, lawyer says, as Ottawa remains silent on X deepfakes

Despite the government’s assertion that Bill C-16, the recently proposed Protecting Victims Act, will quell AI-generated sexual abuse material, Canadian litigator and internet lawyer Maanit Zemel says it leaves a critical gap unaddressed. While the bill may regulate the most serious offenses by individuals, a recent spur of non-consensual sexual deepfakes on X (formerly Twitter) raises questions about how to address online harms that are inflicted by the online platforms themselves. Earlier this month, the social media platform's AI chatbot, Grok, was flooded with requests from users to digitally undress images of women and girls without their consent. Requests from users ranged in intensity, with many asking for women to be stripped down to bikinis, while other requests were more graphic, asking for the women to be portrayed as completely nude, or manipulated into sexually explicit scenarios. The photos were posted publicly on the platform by the chatbot. The chatbot’s image editing feature was introduced in December. The online trend slowly gained traction online, until it blew up on Dec. 31 when X owner Elon Musk posted a request for Grok to generate an image of him in a bikini. The Center for Countering Digital Hate (CCDH) estimated that between Dec. 29 and Jan 8., Grok’s image-generation feature produced more than 4 million images. A report from The New York Times says the publication “conservatively” estimated that at least 1.8 million of those images most likely contained sexualized imagery of women. In its own report, the CCDH said that statistical analysis suggests that around 3 million sexualized images were generated during that time period, 23,000 of which were estimated to be of children. The New York Times report found that in the nine days before Musk’s post, Grok had only generated a total of 311,762 images. It did not specify how many of those were sexualized. Grok’s image-generating feature was restricted for only paid users on Jan. 9, and its ability to digitally undress images of people was reportedly removed on Jan. 14. In the wake of the incident, Privacy Commissioner of Canada Philippe Dufresne announced he would expand his ongoing investigation of X Corp. to include xAI, the artificial intelligence company that operates Grok. AI Minister Evan Solomon chimed in with a post on X saying, “Deepfake sexual abuse is violence. “We must protect Canadians, especially women and young people, from exploitation. Platforms and AI developers have a duty to prevent this harm.” He also said the government is continuing to advance “responsible AI, including introducing Bill C-16, the Protecting Victims Act.” The bill proposes amendments to the Criminal Code that, among other things, would expand the offense of distributing intimate images without consent to include non-consensual sexual deepfakes. “We will keep Canadians safe by amending the Criminal Code and holding abusers accountable,” Solomon wrote. Although Solomon did not mention X or Grok directly, the post, published on Jan. 8, followed news reports about the wave of non-consensual deepfakes on the platform. On Jan. 11, he re-published the post, adding that, “Contrary to media reports, Canada is not considering a ban of X.” The Telegraph had previously reported that the U.K. was attempting to gather international support to ban the platform, and that Canada was said to share the U.K. Prime Minister’s concern over X. Weight of penalties must affect corporations’ bottom line, says Zemel Despite Solomon’s statements, Zemel says Bill C-16 would do little, if anything, to regulate deepfake sexual abuse at a platform level. Based on the internet lawyer's interpretation of the proposed amendments, she says the law should technically apply to platforms, but because the Criminal Code was designed to prosecute at the individual level, Zemel says it would be extremely difficult to enforce it against corporations and identify the individual responsible. “It still comes down to, ‘Then what?’ Who do you prosecute?” she told The Wire Report in an interview. “You can’t throw the entire corporation in jail.” Another option in the Criminal Code, as opposed to jail time, is to apply monetary penalties. Zemel says while it may be easier to go after corporations using that route, it still would not be very effective. The fines in the code are used as a deterrent for individuals, and are not substantial enough to influence corporations. “The issue is more a practical aspect versus a legal aspect,” she says. Further, Zemel says law enforcement would have to prove the platform created or distributed the content knowingly or with reckless disregard, which complicates things even more. While she says there are instances where companies can face criminal liability, those instances are extremely complicated. She says the more practical and effective approach to hold platforms and developers accountable is to develop a regulatory framework, like the Trudeau-era online harms act, Bill C-63. That bill died on the order paper when Parliament was prorogued last year. “It’s sort of the lighter sister of criminal law,” she says. “It’s meant to deal exactly with the corporations, the businesses, those that you don’t put in jail.” The proposed online harms act, which was heavily debated in Parliament, gave rise to regulatory prosecution and enforcement. If a platform was hosting certain online harms, such as intimate images or child sexual abuse material, they would be required to take it down. The maximum penalty for being in violation of the act was proposed to be 6 per cent of a company’s gross global revenue, or $10 million, whichever was greater. Zemel says these kinds of deterrents that are found in regulatory frameworks are far more effective and are needed to hold corporations accountable, since the weight of the penalties affect what is important to the companies: their bottom line. “The problem is our government has not yet passed those regulatory laws,” she says. The Carney government is expected to introduce its own version of C-63, which is likely to be spearheaded by Canadian Identity and Culture Minister Marc Miller. However, the feds have been tight-lipped about when the legislation will be tabled. The Department of Canadian Heritage would not confirm the future of an online harms bill, but said in an emailed statement to The Wire Report that, “Our government intends to act swiftly to better protect Canadians, especially children, from online harm, and will have more details to share in due course.” Grok marks shift in responsibility for what gets posted online In the U.S., section 230 of the Communication Decency Act shields corporations from liability for much of the content users post on their platforms. While Canadian law does not offer that same type of immunity, Zemel says there is still a gap between liability and enforceability. She hopes that if the online harms bill does get introduced, it will bridge that gap. Under that U.S. act, platforms have long argued that they are what are called ‘passive publishers,’ meaning that they are not involved in the creation of the content being distributed. However, Zemel says with what is happening on X, we could be seeing a shift in responsibility. Media platforms now are not only providing users with the technology to create harms such as sexually-explicit deepfakes and child sexual abuse material, but in X’s case, the company’s own software is posting it publicly on its own platform. While Zemel cannot advise on U.S. law, she does think American lawyers will start arguing that platforms are no longer only passive publishers. “They passed the territory from just being passive publishers and distributors to actual creators of the content,” she says. Liability at the individual level As for individual criminal liability under the proposed protecting victims act, if C-16 were passed as it currently stands, many of the images requested by users likely would not fall under its definition of an intimate image, says Zemel. C-16 proposes amending the definition of an intimate image to: “A visual representation showing an identifiable person depicted as nude, as exposing their sexual organs or as engaged in explicit sexual activity, if the depiction is likely to be mistaken for a visual recording of that person.” For images generated of women in bikinis but not exposing sexual organs or engaged in explicit sexual activity, Zemel says the definition would not apply. “Keep in mind though, that we are talking about criminal offenses,” she says, adding that they are “very serious” offenses that can give rise to years in jail. For victims of deepfakes that do not apply to the proposed amendments to the Criminal Code, these images can still fall under civil liability, where the definition of an intimate image comes from privacy law. Zemel says the privacy tort of publicity placing a person in false light applies to when an individual publishes something about someone else that is offensive, false, and an invasion of privacy, but is not defamatory. “In Canada, civil remedies for victims are available,” she says, adding that it is difficult to get a criminal conviction. Zemel says that the law against distributing non-consensual intimate images under the Criminal Code is “rarely” used as is, because to prosecute it requires a lot of evidence. She says that will remain the same if Bill C-16 passes. “Unless the government actually gives both law enforcement and the Crown the resources to prosecute and actually hold people accountable, not much is going to happen,” she says. mcollins@thewirereport.ca

Inside The Wire Report

Information

Hill Times Publishing

Get in Touch With Us

Bill C-16 won’t hold platforms accountable for online harms, lawyer says, as Ottawa remains silent on X deepfakes

Regulatory | By Maria Collins |January 23, 2026

Unlock all the Canadian telecom, broadcasting and digital media news you need.

FREE TRIAL

SUBSCRIBE

Telus alleges Bell downgrading internet service in eastern Canada

Industry minister to conduct new national security review of TikTok, Canadian business allowed to continue operating for now

CRTC will require telecom providers to disclose more details on service coverage

Related Stories

‘Anything that can’t go on forever eventually stops’: ‘Enshittification’ author issues stark warning to Ottawa over AI policy

‘Get the f**k off there’: MPs, Senators call on government to abandon X/Twitter following wave of ‘Grok porn’

Popular Wire Report stories of 2025

Canadian Heritage explores using AI to draft ministerial correspondence

Reuse & Permissions

Inside The Wire Report

Information

Hill Times Publishing

Get in Touch With Us

Regulatory |
By Maria Collins
|January 23, 2026