During neural network training, the sharpness of the Hessian matrix of the training loss rises until training is on the edge of stability. As a result, even nonstochastic gradient descent does not accurately model the underlying dynamical system defined by the gradient flow of the training loss. We use an exponential Euler solver to train the network without entering the edge of stability, so that we accurately approximate the true gradient descent dynamics. We demonstrate experimentally that the increase in the sharpness of the Hessian matrix is caused by the layerwise Jacobian matrices of the network becoming aligned, so that a small change in the network preactivations near the inputs of the network can cause a large change in the outputs of the network. We further demonstrate that the degree of alignment scales with the size of the dataset by a power law with a coefficient of determination between 0.74 and 0.98.

OpenAI on Thursday released its first ever report on how its artificial intelligence tools are being used for covert influence operations, revealing that the company had disrupted disinformation campaigns originating from Russia, China, Israel and Iran. Malicious actors used the company's generative AI models to create and post propaganda content across social media platforms, and to translate their content into different languages. None of the campaigns gained traction or reached large audiences, according to the report. As generative AI has become a booming industry, there has been widespread concern among researchers and lawmakers over its potential for increasing the quantity and quality of online disinformation. Artificial intelligence companies such as OpenAI, which makes ChatGPT, have tried with mixed results to assuage these concerns and place guardrails on their technology.

Today, OpenAI released its first threat report, detailing how actors from Russia, Iran, China, and Israel have attempted to use its technology for foreign influence operations across the globe. The report named five different networks that OpenAI identified and shut down between 2023 and 2024. In the report, OpenAI reveals that established networks like Russia's Doppleganger and China's Spamoflauge are experimenting with how to use generative AI to automate their operations. And while it's a modest relief that these actors haven't mastered generative AI to become unstoppable forces for disinformation, it's clear that they're experimenting, and that alone should be worrying. The OpenAI report reveals that influence campaigns are running up against the limits of generative AI, which doesn't reliably produce good copy or code.

OpenAI identified and removed five covert influence operations based in Russia, China, Iran and Israel that were using its artificial intelligence tools to manipulate public opinion, the company said on Thursday. In a new report, OpenAI detailed how these groups, some of which are linked to known propaganda campaigns, used the company's tools for a variety of "deceptive activities." These included generating social media comments, articles, and images in multiple languages, creating names and biographies for fake accounts, debugging code, and translating and proofreading texts. These networks focused on a range of issues, including defending the war in Gaza and Russia's invasion of Ukraine, criticizing Chinese dissidents, and commenting on politics in India, Europe, and the U.S. in their attempts to sway public opinion. While these influence operations targeted a wide range of online platforms, including X (formerly known as Twitter), Telegram, Facebook, Medium, Blogspot, and other sites, "none managed to engage a substantial audience" according to OpenAI analysts.

OpenAI's report detailed how the five groups used the company's tech in their attempted influence operations. Spamouflage, a previously known group originating in China, used OpenAI's tech to research activity on social media and write posts in Chinese, Korean, Japanese and English, the company said. An Iranian group known as the International Union of Virtual Media also used OpenAI's tech to create articles that it published on its site.

The Chinese army has debuted its latest weapon: a gun-toting robotic dog. The mechanical canine, which has an automatic rifle on its back, was front and centre of recent joint military drills with Cambodia, according to footage from the state broadcaster CCTV. The dog was backed up by a similarly-armed quadcopter in the drills, which saw the machines paired with human soldiers in dry runs for urban assaults. "It can serve as a new member in our urban combat operations, replacing our human members to conduct reconnaissance and identify enemy and strike the target," Chen Wei, a Chinese soldier, said in the video. While they may be technologically advanced, the killer robots are hardly sleek pieces of military hardware; both dog and drone appear to be off-the-shelf pieces of consumer technology with a conventional rifle bolted on top.

The Federal Aviation Administration (FAA) has given Amazon permission to fly its delivery drones beyond visual line of sight (BVLOS). With that hurdle cleared, the company claims it can fly farther and expand drone service, providing customers faster delivery and a larger selection of items, Amazon announced in a blog post. Until now, the FAA has only allowed Amazon to fly drones as far as someone could see them from the ground. That way, spotters or pilots could ensure that drones weren't interfering with aircraft. However, the constraint seriously limited how far the drones could travel.

As former President Donald Trump awaits a Manhattan jury's verdict, he can be forgiven for feeling that his criminal trial resembles a surreal "Alice in Wonderland" farce. He is left to peer through a "Looking-Glass" where everything is backward. The culprit for this hallucinatory nightmare is District Attorney Alvin Bragg who brought a bizarre case based on warped interpretations of law and distorted facts. It is now up to twelve jurors to wade through the lunacy in search of the illusive truth. Bragg's fractured case requires the jury to reach several distinct conclusions on issues that make little sense to begin with.

I'm a reporter on the WIRED Politics desk, and I'm taking over for Makena this week to talk about politicians rising from the dead in India and the rapper Eminem endorsing opposition parties in South Africa. These things haven't really happened, obviously, but deepfakes created by generative AI have made it seem like they have. Already, we're seeing how politicians, campaigns, and regular people are using generative AI in elections. And this is only the beginning. So today, WIRED is launching a project to track it, all over the world.

The United States Department of Justice is quietly prosecuting a novel Espionage Act case involving a drone, a Chinese national, and classified nuclear submarines. The case is such a rarity that it appears to be the first known prosecution under a World War II–era law that bans photographing vital military installations using aircraft, showing how new technologies are leading to fresh national security and First Amendment issues. "This is definitely not something that the law has addressed to any significant degree," Emily Berman, a law professor at the University of Houston who specializes in national security, tells WIRED. "There's definitely no reported cases." On January 5, 2024, Fengyun Shi flew to Virginia while on leave from his graduate studies at the University of Minnesota and rented a Tesla at the airport.