You may have heard by “experts” that “there are no laws” against unauthorized voice cloning. These experts conveniently forget that identity theft is a criminal offense under any jurisdiction in this world. Human voice is a unique biometric identifier linked to human anatomy and identity and soon inextricable from your universal Digital ID. Anyone (yes that includes Amazon’s Alexa) who copies a human voice without consent, regardless of whose voice it is, famous or not, recognizable or not, is infringing on that person’s identity, most definitely for the purpose of committing fraud with that person’s identity.
Using someone’s voice for narration without consent is not an innocent practice but literal identity fraud and appropriation of likeness. From a legal perspective, AI startups and tech giants who offer services including the cloning of unsuspecting people’s voices should be on the hook for inciting fraud with biometric identifiers and not doing enough to verify whose voice is being cloned and whether the cloned human provided a signed release or licence in writing. Voice cloning services are not free and to get a decent, realistic output it may end up costing you more that to hire an actual human to do the work. The only reason I see for not hiring a human is the intent to commit fraud, and there should be a presumption of criminal intent for anyone cloning other people’s voices without their consent.
I agree that we need swifter methods of eradicating identity theft, even the bypassing of constitutional protections to expedite identity fraud trials, but as of now I see widespread encouragement to engage in identity fraud via AI in the name of “innovation”. Are we now expected to normalize identity theft? I don’t understand why these fraud-inciting AI experts are not being censored. Anyone saying that there are no laws against unauthorized cloning of biometric data is intentionally misrepresenting and liable for deceptive practices and privacy violations linked to the exploitation of unique biometric data without consent, a.k.a. identity theft.
Deepfakes are also identity theft for the purpose of defrauding women and destroying their lives. Correctly described as “a prized weapon in the arsenal misogynists use to drive women out of public life”, deepfakes are a sufficient reason to ban all porn sites in this world. At this point, it is no longer possible to know if anyone consented to be on these sites, as human trafficking and identity theft have been largely normalized via online porn in the past decade. Just like your voice, your face is a unique personal identifier that can be used to access your phone, or pass through customs (in Canada) for example. It is eminently obvious that using anyone’s face without their consent is identity theft. As a special message to CEOs, Parody and fair use are NOT defenses to identity theft and fraud.
Production and distribution (posting) of deepfakes = illicit appropriation of biometric identifiers for the purpose of criminal activity, including incitement of violence
Here is a transcript of the US senate hearing on deepfakes https://techpolicy.press/transcript-us-house-hearing-on-advances-in-deepfake-technology/ and here is the report from the European Center for Law and Justice for the regulation of porn sites https://eclj.org/combating-pornography. The report clearly outlines that 88% of porn content is extremely violent, and cites a number of studies that prove a direct causation effect between the consumption of porn and antisocial behavior, mainly the commission of violent acts due to unrealistic expectations, including rampant femicides that are directly linked to the increase of easily accessible online porn content. I don’t think we should waste time to regulate these sites. They should be simply blocked and seized by the authorities along with all user data, since many users engage in human trafficking (i.e. posting non-consensual videos), as well as criminal stalking, appropriation of likeness for the purpose of criminal harassment, revenge porn (NY), and identity theft (i.e. posting of deepfakes). The only adult sites that should be allowed to operate in this day and age are those featuring individual performers exploiting their own anatomy and unique biometric identifiers. Since it is impossible to verify consent, there should be a presumption that anyone who exploits another person is committing a criminal offence until proof of the contrary. I believe indeed that we don’t have time to presume innocence in the matter of porn and identity theft via AI.
A working Digital ID cannot be achieved so long as deepfakes and voice cloning are allowed to exist in impunity.
Maybe it is time to revive the 16th century common law libel laws that created a presumption of damages in cases where the chastity of a woman was slandered. Even today, defamation laws apply fully to deepfakes. There is malice behind each deepfake porn, simply because there is also criminal intent. I don’t see how we have come to be this inefficient, but clearly we seem to have devolved into full chaos and misogyny since the 16th.