ChatGPT has stoked new hopes about the potential of artificial intelligence--but also new fears. Today the White House joined the chorus of concern, announcing it will support a mass hacking exercise at the Defcon security conference this summer to probe generative AI systems from companies including Google. The White House Office of Science and Technology Policy also said that $140 million will be diverted towards launching seven new National AI Research Institutes focused on developing ethical, transformative AI for the public good, bringing the total number to 25 nationwide. The announcement came hours before a meeting on the opportunities and risks presented by AI between vice president Kamala Harris and executives from Google and Microsoft as well as the startups Anthropic and OpenAI, which created ChatGPT. The White House AI intervention comes as appetite for regulating the technology is growing around the world, fueled by the hype and investment sparked by ChatGPT.

Ever since the release of ChatGPT last year, new generative AI tools and services have captured people's attention. Now, Meta is warning that bad actors have taken notice of interest in AI chatbots. The Facebook parent said scammers are creating malware that poses at ChatGPT and similar tools. In a security report released Wednesday, Meta said it discovered 10 malware families posing as ChatGPT or related tools since March. Some of the malicious software, which can steal your personal information and compromise accounts, came in the form of browser extensions and links.

Tiny robots packed with yeast speed up the fermentation of beer and eliminate the need to filter it before bottling. Using living yeast to convert sugar to alcohol is a key part of making beer, but it can be time consuming, and the yeast can spoil and ruin a whole batch of the drink. Martin Pumera at the Brno University of Technology in the Czech Republic and his colleagues thought that both issues could be addressed by swapping yeast for tiny, metallic yeast-filled bots.

Koopman told me that Waymo has a reputation for being methodical and conservative and may deserve more trust than other driverless car companies. Mostly, though, he believes that driverless cars shouldn't be allowed to ditch human backup in vehicles until they have a proven safety record and that companies should be forced to comply with industry standards for human test drivers.

If you spend any time on the internet, you're likely now familiar with the gray-and-teal screenshots of AI-generated text. At first they were meant to illustrate ChatGPT's surprising competence at generating human-sounding prose, and then to demonstrate the occasionally unsettling answers that emerged once the general public could bombard it with prompts. OpenAI, the organization that is developing the tool, describes one of its biggest problems this way: "ChatGPT sometimes writes plausible-sounding but incorrect or nonsensical answers." In layman's terms, the chatbot makes stuff up. As similar services, such as Google's Bard, have rushed their tools into public testing, their screenshots have demonstrated the same capacity for fabricating people, historical events, research citations, and more, and for rendering those falsehoods in the same confident, tidy prose.

The UK economy is at a pivotal moment. Two years on from Covid, and it remains the only country in the developed world where people have continued to drop out of the labour market in greater numbers beyond the pandemic. Rates of economic inactivity have risen and vacancies in the hospitality, health and technology sectors are proving hard to fill. The UK is experiencing new forms of polarisation between good and poor-quality work. How the government responds to the challenges the current jobs market presents is crucial.

Consider a team of three guards (in green) trying to defend a fort from a team of three attackers (in red) in Figure 1. In this "Fort Attack" (FA) domain, each agent can move in one of four cardinal directions with a particular velocity, rotate clockwise or anticlockwise, shoot at an opponent within a given range, or do nothing. Each agent may have partial or full knowledge of the state of the world (e.g., location, status of each agent) at each step, but it has no prior experience of working with the other agents. Also, each agent may have limited (or no) ability to communicate with others. An episode ends when all members of a team are eliminated, an attacker reaches the fort, or the guards protect the fort for a sufficient time period.

A "Ukrainian" drone has been found outside Moscow, an official has said, adding that the discovery had forced local authorities to call off a Victory Day parade for security reasons. Moscow has accused Ukraine of being behind a number of drone attacks on military infrastructure deep inside Russian territory. On Monday, Igor Sukhin, head of the Bogorodsky city district outside the capital Moscow, said that a local resident had found a "Ukrainian" drone in a forest. "This is not the first drone that appeared in the Moscow region," Sukhin said on the messaging app Telegram. A similar drone was in February found in the town of Kolomna approximately 100km (about 60 miles) southeast of Moscow, he added.

Russia's Black Sea Fleet has warded off a drone attack on the Crimean port of Sevastopol, the Moscow-installed governor of the city says. "An attempted attack on Sevastopol was repelled from 3:30am [00:30 GMT]," Mikhail Razvojayev said on Telegram on Monday. "A surface drone [naval] was destroyed by the anti-sabotage forces, the second one exploded on its own," he said, adding that no damage was reported. Passenger ferry service were suspended in the port city, Russia's Interfax news agency reported, citing Sevastopol transport authorities. No reason was given, but the agency said traffic had been suspended in the past due to drone attacks or storms.

A California state court jury has handed Tesla Inc a sweeping win, finding that the carmaker's Autopilot feature did not fail to perform safely in what appears to be the first trial related to a crash involving the partially automated driving software. The verdict could be an important victory for Tesla as it tests and rolls out its Autopilot and more advanced "Full Self-Driving (FSD)" system, which Chief Executive Elon Musk has touted as crucial to his company's future, but which has drawn regulatory and legal scrutiny. Justine Hsu, a resident of Los Angeles, sued the electric vehicle maker in 2020, saying her Tesla Model S swerved into a curb while it was on Autopilot and then an airbag was deployed "so violently it fractured Plaintiff's jaw, knocked out teeth, and caused nerve damage to her face". She alleged there were defects in the design of Autopilot and the airbag, and sought more than $3m in damages for the alleged defects and other claims. Tesla denied liability for the 2019 accident.