Former President Donald Trump shared his mixed feelings on artificial intelligence with YouTuber Logan Paul Thursday, describing the technology as a "superpower" that writes "beautifully" while also calling its capabilities "alarming." His comments on Paul's podcast'Impaulsive' provide a window into how the 78 year old presidential candidate thinks about the rapidly-advancing technology, which 49% of Americans feel concerned about. Trump said he saw a deepfake video of himself promoting a product that was so convincing it caused him to question whether it was real. He went on to note his worry that the harms of deepfakes could be much greater, describing a scenario where a deepfake video of the President of the United States saying "we have just sent thirteen nuclear missiles…and they will hit their targets in 12 minutes and 59 seconds'' could cause a rival leader to preemptively initiate a retaliatory strike. Trump says he asked Elon Musk whether there would be any way for this hypothetical leader to discern the authenticity of the video; Musk is reported to have said "there is no way". The idea of misinformation increasing the risk of a nuclear war is not entirely hypothetical.

Pope Francis met with top comedians at the Vatican on Friday to encourage them to "spread peace" in the midst of "gloomy" news. Pope Francis delivered the first-ever papal address at a G-7 conference on Friday, warning about the ethical pitfalls of artificial intelligence. The pope told the council of world leaders in Fasano, Italy, that AI offers immense benefit to the human race, but also threatens to dehumanize society. "The question of artificial intelligence, however, is often perceived as ambiguous: on the one hand, it generates excitement for the possibilities it offers, while on the other, it gives rise to fear for the consequences it foreshadows," Pope Francis said in his remarks. Pope Francis (C) delivers remarks as French President Emmanuel Macron (L), Italy's Prime Minister Giorgia Meloni (R) and US President Joe Biden (bottom, back turned) take part in a working session on Artificial Intelligence (AI), Energy, Africa-Mediterranean at the Borgo Egnazia resort during the G7 Summit in Savelletri near Bari, Italy.

Police seized ketamine hidden inside life-size Transformer robots in Thailand. A woman who was previously caught trying to ship meth hidden in a food processing machine was trying to send the robots to Taiwan. Cambodian authorities on Friday destroyed more than seven tons of illicit drugs and the ingredients for them, as a drug-fighting official said educating people about their danger is the best way of combating the illegal trade. Some 4.1 tons of the destroyed items were drugs including heroin, marijuana, methamphetamine, ecstasy and ketamine that had been confiscated from traffickers across the country, the National Authority for Combating Drugs said. The remaining 3.2 tons were various chemicals and other ingredients used to produce illegal drugs, it said.

After a gruelling first day discussing how to finance a prolonged war against an authoritarian dictator, G7 leaders in Puglia next turned for advice from someone who insists he is infallible, and for good measure thinks Ukraine should have the courage to wave the white flag. Normally when an 87-year-old claiming infallibility turns up at your door, the instinct is to give them a cup of tea and quietly ring social services. But when 1.3 billion other people, including your hostess, believe he is indeed infallible, the dynamic somewhat changes. So Pope Francis, invited by the devout Catholic and Italian prime minister Giorgia Meloni, was warmly greeted when he reached the summit of mammon, the G7 club of western wealthy countries. Even if G7 is used to listening to the prophecies of economists, he is the first religious leader ever to attend this event, and to give his prediction of what the future holds.

Pope Francis called for a ban on "lethal autonomous weapons" in an address to the G7 leaders' summit in Italy on the perils of artificial intelligence (AI). On Friday, the pontiff was the first head of the Roman Catholic Church to ever attend a Group of Seven meeting. "In light of the tragedy that is armed conflict, it is urgent to reconsider the development and use of devices like the so-called'lethal autonomous weapons' and ultimately ban their use," the pope said. "This starts from an effective and concrete commitment to introduce ever greater and proper human control. No machine should ever choose to take the life of a human being."

Heritage Foundation tech policy director Kara Frederick joins'America's Newsroom' to discuss pornographic AI photos of Taylor Swift sparking conversations about deepfake regulation. A bill aimed at outlawing "revenge porn" has been approved by lawmakers in the Massachusetts House and Senate and shipped to Democratic Gov. Maura Healey, a move advocates say was long overdue. If signed by Healey, the bill -- which bars the sharing of explicit images or videos without the consent of those depicted in the videos -- would leave South Carolina as the only state not to have a law specifically banning revenge porn. Supports say the bill, which landed on Healey's desk Thursday, would align Massachusetts with the other 48 states that have clear prohibitions on disseminating sexually explicit images and videos without the subject's consent. It is a form of abuse that advocates say has grown increasingly common in the digital age, subjecting people to social and emotional harm often inflicted by former romantic partners.

The European Union is readying a landmark antitrust law that could limit more advanced generative AI models. The Federal Trade Commission is investigating a deal that Microsoft made with the AI start-up Inflection, probing whether the tech giant deliberately set up the investment to avoid a merger review. And U.S. enforcers reached a deal that will open the company to greater scrutiny of how it wields power to dominate artificial intelligence, including its multibillion-dollar investments in ChatGPT maker OpenAI. That relationship has also exposed Microsoft to new reputational risks, as OpenAI chief executive Sam Altman frequently invites controversy.

Safety-critical intelligent cyber-physical systems, such as quadrotor unmanned aerial vehicles (UAVs), are vulnerable to different types of cyber attacks, and the absence of timely and accurate attack detection can lead to severe consequences. When UAVs are engaged in large outdoor maneuvering flights, their system constitutes highly nonlinear dynamics that include non-Gaussian noises. Therefore, the commonly employed traditional statistics-based and emerging learning-based attack detection methods do not yield satisfactory results. In response to the above challenges, we propose QUADFormer, a novel Quadrotor UAV Attack Detection framework with transFormer-based architecture. This framework includes a residue generator designed to generate a residue sequence sensitive to anomalies. Subsequently, this sequence is fed into a transformer structure with disparity in correlation to specifically learn its statistical characteristics for the purpose of classification and attack detection. Finally, we design an alert module to ensure the safe execution of tasks by UAVs under attack conditions. We conduct extensive simulations and real-world experiments, and the results show that our method has achieved superior detection performance compared with many state-of-the-art methods.

The COVID-19 pandemic exposed significant weaknesses in the healthcare information system. The overwhelming volume of misinformation on social media and other socioeconomic factors created extraordinary challenges to motivate people to take proper precautions and get vaccinated. In this context, our work explored a novel direction by analyzing an extensive dataset collected over two years, identifying the topics de/motivating the public about COVID-19 vaccination. We analyzed these topics based on time, geographic location, and political orientation. We noticed that while the motivating topics remain the same over time and geographic location, the demotivating topics rapidly. We also identified that intrinsic motivation, rather than external mandate, is more advantageous to inspire the public. This study addresses scientific communication and public motivation in social media. It can help public health officials, policymakers, and social media platforms develop more effective messaging strategies to cut through the noise of misinformation and educate the public about scientific findings.

In modern large language models (LLMs), increasing sequence lengths is a crucial challenge for enhancing their comprehension and coherence in handling complex tasks such as multi-modal question answering. However, handling long context sequences with LLMs is prohibitively costly due to the conventional attention mechanism's quadratic time and space complexity, and the context window size is limited by the GPU memory. Although recent works have proposed linear and sparse attention mechanisms to address this issue, their real-world applicability is often limited by the need to re-train pre-trained models. In response, we propose a novel approach, Hierarchically Pruned Attention (HiP), which simultaneously reduces the training and inference time complexity from $O(T^2)$ to $O(T \log T)$ and the space complexity from $O(T^2)$ to $O(T)$. To this end, we devise a dynamic sparse attention mechanism that generates an attention mask through a novel tree-search-like algorithm for a given query on the fly. HiP is training-free as it only utilizes the pre-trained attention scores to spot the positions of the top-$k$ most significant elements for each query. Moreover, it ensures that no token is overlooked, unlike the sliding window-based sub-quadratic attention methods, such as StreamingLLM. Extensive experiments on diverse real-world benchmarks demonstrate that HiP significantly reduces prompt (i.e., prefill) and decoding latency and memory usage while maintaining high generation performance with little or no degradation. As HiP allows pretrained LLMs to scale to millions of tokens on commodity GPUs with no additional engineering due to its easy plug-and-play deployment, we believe that our work will have a large practical impact, opening up the possibility to many long-context LLM applications previously infeasible.