AI-generated audio is commonplace: voices on home assistants like Alexa use natural language processing. AI music systems like OpenAI's Jukebox have already generated impressive results, but most existing techniques need people to prepare transcriptions and label text-based training data, which takes a lot of time and human labor. Jukebox, for example, uses text-based data to generate song lyrics. AudioLM, described in a non-peer-reviewed paper last month, is different: it doesn't require transcription or labeling. Instead, sound databases are fed into the program, and machine learning is used to compress the audio files into sound snippets, called "tokens," without losing too much information.

Uber Technologies Inc. inked a deal with Motional Inc. to offer driverless deliveries and rides, rekindling its vision of a self-driving taxi fleet nearly two years after it sold its autonomous vehicle division. The San Francisco-based company is partnering with Motional, which is an autonomous driving joint venture between Hyundai Motor Co. and Aptiv Plc. The 10-year deal will pair Motional's all-electric IONIQ 5 robotaxis with Uber's ride-hailing and delivery platform, the companies said in a statement Oct. 6. They did not disclose financial terms. "This agreement will be instrumental to the wide scale adoption of robotaxis," Motional Chief Executive Officer Karl Iagnemma said in a statement.

People take pictures and videos of the Boston Dynamics robot Spot during an event in Lisbon in 2019. People take pictures and videos of the Boston Dynamics robot Spot during an event in Lisbon in 2019. Boston Dynamics and five other robotics companies have signed an open letter saying what many of us were already nervously hoping for anyway: Let's not weaponize general-purpose robots. The six leading tech firms -- including Agility Robotics, ANYbotics, Clearpath Robotics, Open Robotics and Unitree -- say advanced robots could result in huge benefits in our work and home lives but that they may also be used for nefarious purposes. "Untrustworthy people could use them to invade civil rights or to threaten, harm, or intimidate others," the companies said.

The authors apply a novel deep-learning algorithm called a transformer to build surrogate models for simulations of well performance. Transformer architecture initially was developed for natural-language processing problems. However, in recent years, researchers have adapted transformers for time-series forecasting.

Elon Musk has been promising the world a humanoid robot called Optimus for more than a year, but the two prototypes unveiled last week did not exactly dazzle with agility. The company's most advanced robot--made with all Tesla components and close to production-ready, according to Musk--waved unsteadily before being shoved across the stage by three human helpers. "This means a future of abundance, a future where there is no poverty, where you can have whatever you want," Musk said of the machine, which was mounted on a stand and cannot yet walk on its own. "It really is a fundamental transformation of civilization." A second humanoid robot, described by Musk as for "rough development" and made from a mixture of Tesla and off-the-shelf parts, was able to walk forward--very unsteadily.

For a decade now, many of the most impressive artificial intelligence systems have been taught using a huge inventory of labeled data. An image might be labeled "tabby cat" or "tiger cat," for example, to "train" an artificial neural network to correctly distinguish a tabby from a tiger. The strategy has been both spectacularly successful and woefully deficient. Original story reprinted with permission from Quanta Magazine, an editorially independent publication of the Simons Foundation whose mission is to enhance public understanding of science by covering research developments and trends in mathematics and the physical and life sciences. Such "supervised" training requires data laboriously labeled by humans, and the neural networks often take shortcuts, learning to associate the labels with minimal and sometimes superficial information.

State and local officials need to act proactively to make sure that widespread use of self-driving vehicles doesn't leave out historically disadvantaged communities, a team of researchers from the Urban Institute warned in a new report. The researchers said a broad shift from human drivers to software-piloted vehicles could help poor people and non-white communities, if the technology can reduce the number of traffic deaths and cut down on the air pollution that disproportionately affects those residents. Autonomous vehicles could also increase transportation options for older people or people with disabilities, the Urban analysts said. But none of those advantages are guaranteed, they cautioned. "The degree to which [autonomous vehicles] change the transportation system and society overall will be mediated by regulatory choices at the local, state and federal levels," the researchers wrote in their report. "If [autonomous vehicles] ultimately reinforce inequitable access to transportation, reduce public transit use, increase [the amount of driving], increase congestion and exacerbate the causes of climate change, this technological advancement may ultimately fall short of its full promise--or even worsen the existing problems endemic to the automobile-dominated US transportation system," they added.

Were you unable to attend Transform 2022? Check out all of the summit sessions in our on-demand library now! For years, U.S. defense officials and Washington think-tankers alike have debated whether the future of our military could -- or should -- look a little less human. Already, the U.S. military has started to rely on technology that employs machine learning, artificial intelligence (AI), and big data -- raising ethical questions along the way. While these technologies have countless beneficial applications, ranging from threat assessment to preparing troops for battle, they rightfully evoke concerns about a future in which Terminator-like machines take over.

Text-to-image art generators work through a process called deep learning, in which algorithms make predictions and complete tasks in a process that mimics the human brain's neurons. In the case of AI-generated art, the generators pull from a database of existing pictures and illustrations to put together a discrete piece based on a user's prompt. Pinkney explained that his own creation is adapted from an open-source deep-learning model called Stable Diffusion, which already has vast data sets of information. Text-to-Pokémon works by matching Stable Diffusion's data sets to a data set of 850 Pokémon images from a previous university-run research project, which Pinkney filed using an automated caption system to categorize each image with a text description.

Electronic design automation (EDA) or computer-aided design (CAD) is a category of software tools for designing electronic systems, such as integrated circuits (ICs). With EDA tools, designers can finish the design flow of very large scale integrated (VLSI) chips with billions of transistors. EDA tools are essential to modern VLSI design due to the large scale and high complexity of electronic systems. Recently, with the boom of artificial intelligence (AI) algorithms, the EDA community is actively exploring AI for IC techniques for the design of advanced chips. Many studies have explored machine learning (ML) based techniques for cross-stage prediction tasks in the design flow to achieve faster design convergence.