fev24
Large tech platforms including TikTok, X and Facebook will soon have to identify AI-generated content in order to protect the upcoming European election from disinformation.
"We know that this electoral period that's opening up in the European Union is going to be targeted either via hybrid attacks or foreign interference of all kinds," Internal Market Commissioner Thierry Breton told European lawmakers in Strasbourg on Wednesday. "We can't have half-baked measures."
Breton didn't say when exactly companies will be compelled to label manipulated content under the EU's content moderation law, the Digital Services Act (DSA). Breton oversees the Commission branch enforcing the DSA on the largest European social media and video platforms, including Facebook, Instagram and YouTube. https://www.politico.eu/article/eu-big-tech-help-deepfake-proof-election-2024/
jan24
A bipartisan group of three senators is looking to give victims of sexually explicit deepfake images a way to hold their creators and distributors responsible.
Sens. Dick Durbin, D-Ill.; Lindsey Graham, R-S.C.; and Josh Hawley, R-Mo., plan to introduce the Disrupt Explicit Forged Images and Non-Consensual Edits Act on Tuesday, a day ahead of a Senate Judiciary Committee hearing on internet safety with CEOs from Meta, X, Snap and other companies. Durbin chairs the panel, while Graham is the committee’s top Republican.
Victims would be able to sue people involved in the creation and distribution of such images if the person knew or recklessly disregarded that the victim did not consent to the material. The bill would classify such material as a “digital forgery” and create a 10-year statute of limitations.
https://www.nbcnews.com/tech/tech-news/deepfake-bill-open-door-victims-sue-creators-rcna136434
jan24
South Korea
South Korea's special parliamentary committee on Tuesday (Jan 30) passed a revision to the Public Official Election Act which called for a ban on political campaign videos that use AI-generated deepfakes in the election season.
https://www.wionews.com/world/south-korea-imposes-90-day-ban-on-deepfake-political-campaign-videos-685152
jan24
As artificial intelligence starts to reshape society in ways predictable and not, some of Colorado’s highest-profile federal lawmakers are trying to establish guardrails without shutting down the technology altogether.
U.S. Rep. Ken Buck, a Windsor Republican, is cosponsoring legislation with California Democrat Ted Lieu to create a national commission focused on regulating the technology and another bill to keep AI from unilaterally firing nuclear weapons.
Sen. Michael Bennet, a Democrat, has publicly urged the leader of his caucus, Majority Leader Chuck Schumer, to carefully consider the path forward on regulating AI — while warning about the lessons learned from social media’s organic development. Sen. John Hickenlooper, also a Democrat, chaired a subcommittee hearing last September on the matter, too.
https://www.denverpost.com/2024/01/28/artificial-intelligence-congress-regulation-colorado-michael-bennet-ken-buck-elections-deepfakes/
jan24
US politicians have called for new laws to criminalise the creation of deepfake images, after explicit faked photos of Taylor Swift were viewed millions of times online.
The images were posted on social media sites, including X and Telegram.
US Representative Joe Morelle called the spread of the pictures "appalling".
In a statement, X said it was "actively removing" the images and taking "appropriate actions" against the accounts involved in spreading them.
It added: "We're closely monitoring the situation to ensure that any further violations are immediately addressed, and the content is removed."
While many of the images appear to have been removed at the time of publication, one photo of Swift was viewed a reported 47 million times before being taken down.
The name "Taylor Swift" is no longer searchable on X, alongside terms such as "Taylor Swift AI" and "Taylor AI".
https://www.bbc.com/news/technology-68110476
jan24 Daily MAil
The answer
lies largely in the lack of laws to prosecute those who make such content.
There is
currently no federal legislation against the conduct and only six states – New
York, Minnesota, Texas, Hawaii, Virginia and Georgia – have
passed legislation which criminalizes it.
In Texas, a
bill was enacted in September 2023 which made it an offense to create or share
deepfake images without permission which 'depict the person with the person's
intimate parts exposed or engaged in sexual conduct'.
The offense is
a Class A misdemeanor and punishments include up to a year in prison and fines
up to $4000.
In Minnesota,
the crime can carry a three-year sentence and fines up to $5,000.
Several of
these laws were introduced following earlier legislation which outlawed the use
of deepfakes to influence an election, such as through the creation of fake
images or videos which portray a politician or public official.
A handful of
other states, including California and Illinois, don't have laws against the act
but instead allow deepfake victims to sue perpetrators. Critics have said this
doesn't go far enough and that, in many cases, the creator is unknown.
At the federal
level, Joe Biden signed an executive order in
October which called for a ban on the use of generative AI to make child abuse
images or nonconsensual 'intimate images' of real people. But this was purely
symbolic and does not create a means to punish makers.
The finding
that 415,000 deepfake images were posted online last year was made by Genevieve
Oh, a researcher who analyzed the top ten websites which host such content.
Oh also
found 143,000
deepfake videos were uploaded in 2023 – more than during the
previous six years combined. The videos, published across 40 different websites
which host fake videos, were viewed more than 4.2 billion times.
Outside of
states where laws which criminalize the conduct exist, victims and prosecutors
must rely on existing legislation which can be used to charge offenders.
These include
laws around cyberbullying, extortion and harassment. Victims who are
blackmailed or subject to repeated abuse can attempt to use these laws against
perpetrators who weaponize deepfake images.
But they do
not prohibit the fundamental act of creating a hyper-realistic, explicit photo
of a person, then sharing it with the world, without their consent.
A 14-year-old
girl from New Jersey who was depicted in a pornographic deepfake image created
by one of her male classmates is now leading
a campaign to have a federal law passed.
Francesca Mani
and her mom, Dorota Mani, recently met with members of Congress at Capitol Hill
to push for laws targeting perpetrators.
https://www.dailymail.co.uk/news/article-13007753/deepfake-porn-laws-internet.html
jan24
COLUMBUS, Ohio (WCMH) – As AI becomes more popular, there is a rising concern about “deepfakes.” A deepfake is a “convincing image, video or audio hoax,” created using AI that impersonates someone or makes up an event that never happened.
At the Ohio Statehouse, Representatives Brett Hillyer (R-Uhrichsville) and Adam Mathews (R-Lebanon) just introduced House Bill 367 to address issues that may arise with the new technology.
Ohio mother faked child’s cancer battle for money, investigators say
“In my day to day I see how important people’s name image and likeness and the copyright there is within it,” Mathews said.
Mathews said the intent of the bill is to make sure everyone, not just high-profile people, is protected. Right now, Mathews said functionally, the only way one can go after someone for using their name, image or likeness (NIL) is if they’re using it to say you endorse a product or to defraud someone.
“I wanted to put every single Ohioan at the same level as our most famous residents from Joe Burrow to Ryan Day,” Mathews said. “There are a lot of things people can do with your name, image and likeness that could be harmful to your psyche or reputation.”
The bill makes it so any Ohioan can go after someone who uses their NIL for a deepfake. With fines as high as $15,000 dollars for the creation of a malicious deepfake; the court can also order that a deepfake be taken down if it is malicious.
https://news.yahoo.com/bill-introduced-statehouse-protect-ohioans-230000251.html?guccounter=1&guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLnB0Lw&guce_referrer_sig=AQAAAENj6zcIxbhxJAo2pJ_AmmsXSVymCSWCxsInYs7yVnzFtgMmgXTbh3aCW4mrWfEnG8C_JQ_juc-EH5259FdOiw8tDTkLfe1UpxXxtl93u5IpvpnNv15CircmHtj6i1Rbz1b6mkqAkaYG6pZpGEMIVKs2KtScO62yGmLZOkvbJSmb
jan24
A pair of U.S. House of Representative members have introduced a bill intended to restrict unauthorized fraudulent digital replicas of people.
The bulk of the motivation behind the legislation, based on the wording of the bill, is the protection of actors, people of notoriety and girls and women defamed through fraudulent porn made with their face template.
Curiously, the very real threat of using deepfakes to defraud just about everyone else in the nation is not mentioned. Those risks are growing and could result in uncountable financial damages as organizations rely on voice and face biometrics for ID verification.
The representatives, María Elvira Salazar (R-Fla.) and Madeleine Dean (D-Penn.), do not mention the global singer/songwriter Taylor Swift in their press release, it cannot have escaped them that she’s been victimized, too.
https://www.biometricupdate.com/202401/us-lawmakers-attack-categories-of-deepfake-but-miss-everyday-fraud
jan24
House lawmakers introduced legislation to try to curb the unauthorized use of deepfakes and voice clones.
The legislation, the No AI Fraud Act, is sponsored by Rep. Maria Salazar (R-FL), Rep. Madeleine Dean (D-PA), Rep. Nathaniel Moran (R-TX), Rep. Joe Morelle (D-NY) and Rep. Rob Wittman (R-VA). The legislation would give individuals more control over the use of their identifying characteristics in digital replicas. It affirms that every person has a “property right in their own likeness and voice,” and do not expire upon a person’s death. The rights can be transferred to the heirs or designees for a period of 10 years after the individual’s death. It sets damages at $50,000 for each unauthorized violation by a personalized cloing service, or the actual damages suffered plus profits from the use. Damages are set at $5,000 per violation for unauthorized publication, performance, distribution or transmission of a digital voice replica or digital depiction, or the actual damages.
https://deadline.com/2024/01/ai-legislation-deepfakes-house-of-representatives-1235708983/
jan24
Illinois
Lawmakers this spring approved a new protection for victims of “deepfake porn.” Starting in 2024, people who are falsely depicted in sexually explicit images or videos will be able to sue the creator of that material.
The law is an amendment to the state’s existing protections for victims of “revenge porn,” which went into effect in 2015.
In recent years, deepfakes – images and videos that falsely depict someone – have become more sophisticated with the advent of more readily available artificial intelligence tools. Women are disproportionately the subject of deepfake porn.
Some sponsors of the legislation, notably chief sponsor Rep. Jennifer Gong-Gershowitz, D-Glenview, have indicated interest in further regulating the use of artificial intelligence.
https://chicagocrusader.com/more-than-300-statutes-became-law-in-the-new-year/
dez23
Prohibition on book bans, right to sue for ‘deepfake porn’ among new laws taking effect Jan. 1
Capitol News Illinois | By Peter Hancock, Andrew Adams & Jerry Nowicki
Lawmakers this spring approved a new protection for victims of “deepfake porn.” Starting in 2024, people who are falsely depicted in sexually explicit images or videos will be able to sue the creator of that material.
The law is an amendment to the state’s existing protections for victims of “revenge porn,” which went into effect in 2015.
In recent years, deepfakes – images and videos that falsely depict someone – have become more sophisticated with the advent of more readily available artificial intelligence tools. Women are disproportionately the subject of deepfake porn.
https://www.nprillinois.org/illinois/2023-12-26/prohibition-on-book-bans-right-to-sue-for-deepfake-porn-among-new-laws-taking-effect-jan-1
Out23 NOVA IORQUE NOVA LEi
New York Bans Deepfake Revenge Porn Distribution as AI Use Grows
New York Gov. Kathy Hochul (D) on Friday signed into law legislation banning the dissemination of pornographic images made with artificial intelligence without the consent of the subject.
https://news.bloomberglaw.com/in-house-counsel/n-y-outlaws-unlawful-publication-of-deepfake-revenge-porn
https://hudsonvalleyone.com/2023/10/15/deepfake-porn-in-new-york-state-means-jail-time/
out23 PROPOSTA
Bill would ban 'deepfake' pornography in Wisconsin
https://eu.jsonline.com/story/news/politics/2023/10/02/proposed-legislation-targets-deepfake-pornography/71033726007/
saet23 NI Bill
Assemblymember Amy Paulin’s (D-Scarsdale) legislation, which makes the nonconsensual use of “deepfake” images disseminated in online communities a criminal offense, has been signed into law by Governor Hochul.
“Deepfakes” are fake or altered images or videos created through the use of artificial intelligence. Many of these images and videos map a face onto a pornographic image or video. Some create a pornographic image or video out of a still photograph. These pornographic images and films are sometimes posted online without the consent of those in them – often with devastating consequences to those portrayed in the images.
https://talkofthesound.com/2023/09/25/amy-paulin-dissemination-of-deepfake-images-now-a-crime-in-new-york/
Clarke told ABC News that her DEEPFAKES Accountability Act would provide prosecutors, regulators and particularly victims with resources, like detection technology, that Clarke believes they need to stand up against the threat posed by nefarious deepfakes.
https://abcnews.go.com/Politics/bill-criminalize-extremely-harmful-online-deepfakes/story?id=103286802~
set23
Multimedia that have either been created (fully synthetic) or edited (partially synthetic) using some form of machine/deep learning (artificial intelligence) are referred to as deepfakes.
'Contextualizing Deepfake Threats to Organizations' PDF (arquivo)
https://media.defense.gov/2023/Sep/12/2003298925/-1/-1/0/CSI-DEEPFAKE-THREATS.PDF
set23
Virginia revenge porn law updated to include deepfakes
https://eu.usatoday.com/videos/tech/2023/09/12/virginia-revenge-porn-law-updated-include-deepfakes/1637140001/
================
set23
Artificial intelligence’s ability to generate deepfake content that easily fools humans poses a genuine threat to financial markets, the head of the Securities and Exchange Commission warned.
https://news.bloomberglaw.com/artificial-intelligence/deepfakes-pose-real-risk-to-financial-markets-secs-gensler?source=newsletter&item=body-link®ion=text-section
