Heritage Foundation tech policy director Kara Frederick joins'America's Newsroom' to discuss pornographic AI photos of Taylor Swift sparking conversations about deepfake regulation. Even before pornographic and violent deepfake images of Taylor Swift began widely circulating in the past few days, state lawmakers across the U.S. had been searching for ways to quash such nonconsensual images of both adults and children. But in this Taylor-centric era, the problem has been getting a lot more attention since she was targeted through deepfakes, the computer-generated images using artificial intelligence to seem real. Here are things to know about what states have done and what they are considering. HOUSE LAWMAKERS TO SHINE LIGHT ON HOW AI CAN MAKE CONGRESS'MORE EFFICIENT' Artificial intelligence hit the mainstream last year like never before, enabling people to create ever-more realistic deepfakes.

Kim, who was born in South Korea, was one of the first three Korean American women elected to Congress in 2020. She previously served in the state Assembly for two years and unsuccessfully ran for Congress in 2018. Kim worked for more than two decades for then-Rep. Kim told The Times she's running to "continue to bring commonsense back to Washington, break through partisan gridlock, and deliver results." She added that "we must make life affordable, keep communities safe, and ensure America leads on the world stage."

The rise of AI-generated voices mimicking celebrities and politicians could make it even harder for the Federal Communications Commission (FCC) to fight robocalls and prevent people from getting spammed and scammed. That's why FCC Chairwoman Jessica Rosenworcel wants the commission to officially recognize calls that use AI-generated voices as "artificial," which would make the use of voice cloning technologies in robocalls illegal. As TechCrunch notes, the FCC's proposal will make it easier to go after and charge bad actors. "AI-generated voice cloning and images are already sowing confusion by tricking consumers into thinking scams and frauds are legitimate," FCC Chairwoman Jessica Rosenworcel said in a statement. "No matter what celebrity or politician you favor, or what your relationship is with your kin when they call for help, it is possible we could all be a target of these faked calls."

As a tax professor, I love taxes: the theory, the policy, even the politics. But I have a confession to make. My taxes are not complicated. Yet, every year, I spend hour upon hour gathering documents, paying for tax preparation software, entering in my income, and puzzling through the instructions as I try to figure out whether I am eligible for this or that deduction or credit. Every year, I think to myself: There has got to be a better way!

Former presidential candidate Andrew Yang spoke with Fox News Digital about the dangers of AI. Former Democratic presidential candidate Andrew Yang spoke to Fox News Digital about the dangers of artificial intelligence, known as AI, and said the government is not doing nearly enough to prepare for the potentially harmful effects. "AI is a very, very powerful technology and set of tools and there's nothing intrinsically positive or negative about tools, but there is something positive and negative about how tools can be used," Yang told Fox News Digital this week. "And you can very clearly see deepfake videos already being employed for political purposes. Fake pictures of terrorist attacks being used to manipulate the stock market. A robocall in President Biden's voice trying to discourage turnout and we're just at the beginning of this." "We're not going to be able to tell up from down and left from right and if people show you a video of me doing something heinous, I'll just shrug and be like, didn't happen and that could be the best defense before too long."

The United States military has carried out new strikes against 10 drones belonging to the Iran-aligned Houthi rebels in Yemen as well as a ground control centre. On Thursday, US forces targeted a "Houthi UAV ground control station and 10 Houthi one-way UAVs" that "presented an imminent threat to merchant vessels and the US Navy ships in the region", the US military's Central Command (CENTCOM) said in a statement referring to unmanned aerial vehicles. "This action will protect freedom of navigation and make international waters safer and more secure for US Navy vessels and merchant vessels," it added. The group said on Wednesday that all US and British warships participating in "aggression" against Yemen are targets, heightening concerns over the escalating tensions in the region as well as the increased disruption to world trade. CENTCOM said earlier that the USS Carney had shot down an antiship ballistic missile fired by the Houthis and downed three Iranian drones less than an hour later.

With online task arrivals, the objective of the BASIC problem is to derive a schedule x to maximize the aggregate system utility. The schedule decides three outputs: task stage execution order on the GPU, number of stages to execute for each task, and task batching decisions. For each scheduling period t, we use xt(i, j) {0, 1} to denote whether the j-th stage of task Ti is executed. Besides, we use P to denote a batch of tasks, where ‖P‖ denotes the number of tasks being batched.

Global climate change is a huge challenge facing society today. The rapid growth of computing overall and of machine learning (ML) in particular rightfully raises concerns about their carbon footprints. As an early and enthusiastic adopter of ML, a manufacturer of millions of smartphones annually, and a significant cloud provider, Google is in a nearly unique position to compare the impact and efficiency of ML on the two ends of the information technology (IT) computing spectrum. Keep in mind this article is not a comparison of all computation done on phones and the cloud, but solely on the impact of ML on energy use and operational CO2e. We provide the data to support these insights. While primarily focused on operational CO2e generated from computer use, we also address the relative impact of embodied CO2e. Computers in datacenters draw electricity from the grid continuously. Because smartphones operate from a battery, they only draw electricity from the grid when connected to a charger. To account for smartphone ML energy accurately, we must include the energy overhead of their chargers. Wireless charging is increasingly popular due to its convenience and the reduction in smartphone wear and tear by avoiding the repeated insertion of a cable. For wired charging, energy is lost from the AC/DC power adapter in the charger and in the power management integrated circuit (PMIC) battery charger in the phone. Wireless charging loses additional energy through the inductive coils.

Thomas Haigh (thomas.haigh@gmail.com) is a professor of history at the University of Wisconsin--Milwaukee, WI, USA, and a Comenius visiting professor at Siegen University, Germany.

Transfer learning (TL) has been demonstrated to improve DNN model performance when faced with a scarcity of training samples. However, the suitability of TL as a solution to reduce vulnerability of overfitted DNNs to privacy attacks is unexplored. A class of privacy attacks called membership inference attacks (MIAs) aim to determine whether a given sample belongs to the training dataset (member) or not (nonmember). We introduce Double-Dip, a systematic empirical study investigating the use of TL (Stage-1) combined with randomization (Stage-2) to thwart MIAs on overfitted DNNs without degrading classification accuracy. Our study examines the roles of shared feature space and parameter values between source and target models, number of frozen layers, and complexity of pretrained models. We evaluate Double-Dip on three (Target, Source) dataset paris: (i) (CIFAR-10, ImageNet), (ii) (GTSRB, ImageNet), (iii) (CelebA, VGGFace2). We consider four publicly available pretrained DNNs: (a) VGG-19, (b) ResNet-18, (c) Swin-T, and (d) FaceNet. Our experiments demonstrate that Stage-1 reduces adversary success while also significantly increasing classification accuracy of nonmembers against an adversary with either white-box or black-box DNN model access, attempting to carry out SOTA label-only MIAs. After Stage-2, success of an adversary carrying out a label-only MIA is further reduced to near 50%, bringing it closer to a random guess and showing the effectiveness of Double-Dip. Stage-2 of Double-Dip also achieves lower ASR and higher classification accuracy than regularization and differential privacy-based methods.