Behind closed doors today, the US Senate will grill nearly two dozen tech executives including Tesla CEO and longtime AI critic Elon Musk, ChatGPT-maker and staunch AI defender Sam Altman and Microsoft founder Bill Gates, on how best to regulate AI. Democrat Senate Majority Leader Chuck Schumer, who spearheaded the effort for today's Senate'AI Insight Forum,' described the all-day debate on the implications of artificial intelligence as an'all-hands-on-deck moment for Congress.' 'For Congress to legislate on artificial intelligence,' Senator Schumer said Tuesday, 'is for us to engage in one of the most complex and important subjects Congress has ever faced.' Senator Schumer is set to moderate the forum on how Congress should set artificial intelligence safeguards, which runs from 10AM to 5PM, for its first half. Republican Senator Mike Rounds of South Dakota will help moderate the forum.

Senate Majority Leader Charles E. Schumer (D-N.Y.) will host the AI Insight Forum -- which is intended to serve as the bedrock for his "all hands on deck" plan to respond to recent AI advances -- in the grand Kennedy Caucus Room, the historic stage of Senate probes into the sinking of the Titanic, as well as Watergate. The more than 20 attendees include Tesla CEO and X owner Elon Musk, Meta CEO Mark Zuckerberg, Google CEO Sundar Pichai and ChatGPT-maker OpenAI CEO Sam Altman, among other top tech executives, civil rights leaders, labor chiefs and researchers.

A senior California lawmaker will introduce a new artificial intelligence (AI) bill to the state's senate on Wednesday, adding to national and global efforts to regulate the fast-accelerating technology. Although there are several attempts in Congress to draft AI legislation, the state of California--home to Silicon Valley, where most of the world's top AI companies are based--has a role to play in setting guardrails on the industry, according to state Senator Scott Wiener, (D--San Francisco) who drafted the bill. "In an ideal world we would have a strong federal AI regulatory scheme," Wiener said in an interview with TIME on Tuesday, adding that he supports attempts in Congress and the White House to regulate the technology. "But California has a history of acting when the federal government is moving either too slowly or not acting." He added: "We need to get ahead of these risks, not do what we've done in the past around social media or other technology, where we do nothing before it's potentially too late."

When it comes to artificial intelligence, United States senators are looking to the titans of Silicon Valley to fix a Senate problem--a problem today's political class perpetuates daily with their increasingly hyper-partisan ways, which generative AI now feeds off of as it helps rewrite our collective future. Today, the Senate is hosting a first-of-its-kind, closed-door AI forum led by the likes of Elon Musk, Mark Zuckerberg, Bill Gates, and more than 17 others, including ethicists and academics. Even though they'll be on the senators' turf, for roughly six hours, they'll get microphones while the nation's elected leaders get muzzled. "All Senators are encouraged to attend to listen to this important discussion, but please note the format will not afford Senators the opportunity to provide remarks or to ask questions to the speakers," a notice from majority leader Chuck Schumer reads. As generative AI is poised to flood the internet with more--and more convincing--disinformation and misinformation, many AI experts say the top goal of the Senate should be restoring faith in, well, the Senate itself.

Haywood Talcove, CEO of LexisNexis Risk Solutions Government Group, tells Fox News Digital that criminal groups, mostly in other countries, are advertising on social media to market their AI capabilities for fraud and other crimes. Google is set to require political advertising that uses artificial intelligence to generate images or sounds come with a visible disclosure for users. "AI-generated content should absolutely be disclosed in political advertisements. Not doing so leaves the American people open to misleading and predatory campaign ads," Ziven Havens, the Policy Director at the Bull Moose Project, told Fox News Digital. "In the absence of government action, we support the creation of new rulemaking to handle the new frontier of technology before it becomes a major problem" Havens' comments come after Google revealed last week that it will start requiring the disclosure of the use of AI to alter images in political ads starting in November, a little more than a year before the 2024 election, according to a PBS report.

Cooperation between the U.S. and its allies and partners, particularly those in the Indo-Pacific, will be critical for Washington's new Replicator initiative to succeed, experts say, as the Pentagon seeks to negate China's military advantage in numbers by fielding thousands of smart, affordable drones. But questions remain about how much technology Washington will be willing to share without the risk of compromising the high degree of cybersecurity the new artificial intelligence-enabled systems will depend on. "We will be working with industry, Congress and allies and partners in everything that we do," U.S. Deputy Secretary of Defense Kathleen Hicks, who is heading up the initiative, said Sept. 6, as experts highlighted an array of possible collaboration opportunities.

Traditional adversarial attacks concentrate on manipulating clean examples in the pixel space by adding adversarial perturbations. By contrast, semantic adversarial attacks focus on changing semantic attributes of clean examples, such as color, context, and features, which are more feasible in the real world. In this paper, we propose a framework to quickly generate a semantic adversarial attack by leveraging recent diffusion models since semantic information is included in the latent space of well-trained diffusion models. Then there are two variants of this framework: 1) the Semantic Transformation (ST) approach fine-tunes the latent space of the generated image and/or the diffusion model itself; 2) the Latent Masking (LM) approach masks the latent space with another target image and local backpropagation-based interpretation methods. Additionally, the ST approach can be applied in either white-box or black-box settings. Extensive experiments are conducted on CelebA-HQ and AFHQ datasets, and our framework demonstrates great fidelity, generalizability, and transferability compared to other baselines. Our approaches achieve approximately 100% attack success rate in multiple settings with the best FID as 36.61. Code is available at https://github.com/steven202/semantic_adv_via_dm.

In this paper, we propose PhantomSound, a query-efficient black-box attack toward voice assistants. Existing black-box adversarial attacks on voice assistants either apply substitution models or leverage the intermediate model output to estimate the gradients for crafting adversarial audio samples. However, these attack approaches require a significant amount of queries with a lengthy training stage. PhantomSound leverages the decision-based attack to produce effective adversarial audios, and reduces the number of queries by optimizing the gradient estimation. In the experiments, we perform our attack against 4 different speech-to-text APIs under 3 real-world scenarios to demonstrate the real-time attack impact. The results show that PhantomSound is practical and robust in attacking 5 popular commercial voice controllable devices over the air, and is able to bypass 3 liveness detection mechanisms with >95% success rate. The benchmark result shows that PhantomSound can generate adversarial examples and launch the attack in a few minutes. We significantly enhance the query efficiency and reduce the cost of a successful untargeted and targeted adversarial attack by 93.1% and 65.5% compared with the state-of-the-art black-box attacks, using merely ~300 queries (~5 minutes) and ~1,500 queries (~25 minutes), respectively.

Prompt-based learning has been proved to be an effective way in pre-trained language models (PLMs), especially in low-resource scenarios like few-shot settings. However, the trustworthiness of PLMs is of paramount significance and potential vulnerabilities have been shown in prompt-based templates that could mislead the predictions of language models, causing serious security concerns. In this paper, we will shed light on some vulnerabilities of PLMs, by proposing a prompt-based adversarial attack on manual templates in black box scenarios. First of all, we design character-level and word-level heuristic approaches to break manual templates separately. Then we present a greedy algorithm for the attack based on the above heuristic destructive approaches. Finally, we evaluate our approach with the classification tasks on three variants of BERT series models and eight datasets. And comprehensive experimental results justify the effectiveness of our approach in terms of attack success rate and attack speed.

Epidemiological models must be calibrated to ground truth for downstream tasks such as producing forward projections or running what-if scenarios. The meaning of calibration changes in case of a stochastic model since output from such a model is generally described via an ensemble or a distribution. Each member of the ensemble is usually mapped to a random number seed (explicitly or implicitly). With the goal of finding not only the input parameter settings but also the random seeds that are consistent with the ground truth, we propose a class of Gaussian process (GP) surrogates along with an optimization strategy based on Thompson sampling. This Trajectory Oriented Optimization (TOO) approach produces actual trajectories close to the empirical observations instead of a set of parameter settings where only the mean simulation behavior matches with the ground truth.