Rather than offering a revolutionary new approach to gig work, RentAHuman is filled with bots that just want me to be another cog in the AI hype machine. I'm not above doing some gig work to make ends meet. In my life, I've worked snack food pop-ups in a grocery store, ran the cash register for random merch booths, and even hawked my own plasma at $35 per vial. So, when I saw RentAHuman, a new site where AI agents hire humans to perform physical work in the real world on behalf of the virtual bots, I was eager to see how these AI overlords would compare to my past experiences with the gig economy. Launched in early February, RentAHuman was developed by software engineer Alexander Liteplo and his cofounder, Patricia Tani.

Computational methods to aid journalists in the task often require adapting a model to specific domains and generating explanations. However, most automated fact-checking methods rely on three-class datasets, which do not accurately reflect real-world misinformation. Moreover, fact-checking explanations are often generated based on text summarization of evidence, failing to address the relationship between the claim and the evidence. To address these issues, we extend the self-rationalization method--typically used in natural language inference (NLI) tasks--to fact verification. We propose a label-adaptive learning approach: first, we fine-tune a model to learn veracity prediction with annotated labels (step-1 model). Then, we fine-tune the step-1 model again to learn self-rationalization, using the same data and additional annotated explanations. Our results show that our label-adaptive approach improves veracity prediction by more than ten percentage points (Macro F1) on both the PubHealth and AVeriTec datasets, outperforming the GPT-4 model. Furthermore, to address the high cost of explanation annotation, we generated 64 synthetic explanations from three large language models: GPT-4-turbo, GPT-3.5-turbo, and Llama-3-8B and few-shot fine-tune our step-1 model. The few-shot synthetic explanation fine-tuned model performed comparably to the fully fine-tuned self-rationalization model, demonstrating the potential of low-budget learning with synthetic data. Our label-adaptive self-rationalization approach presents a promising direction for future research on real-world explainable fact-checking with different labeling schemes.

Adam: Hello, this is CoRecursive and I'm Adam Gordon Bell. Today on April 1st, 2014, something interesting happened. Hacker News moderator Dan G, or dang, he made the following post. Challenge: Keep lame April Fools' jokes off the front page. Most April Fools' gags are lame. Only the very best ones that show some sort of ingenuity deserve attention. I propose that we, or rather you, flag these jokes so they don't end up on the front page. Dang wasn't the first person to complain about the lameness of tech company April Fools' Day jokes, but I think to the various developers and tech company people hanging out on Hacker News, dang's statement was a really big one. It kind of marked the end of this era of companies dropping these big jokes on April Fools'. So today in our first This Day in History Segment, I want to share some of history not just of April Fools', but of tech pranks in general, all leading up to that sort of cancellation statement by dang, and even right up to today actually. Why were pranks and April Fools' jokes traditionally celebrated in tech, and why are they now considered as dang said, "lame?" And here to talk about those pranks, I have my frequent co-host and developer extraordinaire, possible neighbor, Don McKay, and also my favorite PhD candidate and mathematician, Krystal Maughan. Why don't you guys say hello? Krystal: Hi, I'm happy to be here! I would challenge your assumption that they died on that day. I think we very much still see them today and they're just as an annoying today as they were in 2014. Krystal: I really like them because for me, getting into tech in general was through things like Hackerspaces and DEFCON, maybe some of these jokes flop, but when they're great, they're really funny. And I think that those kind of things should be encouraged as long as they're not necessarily malicious. That's the goal, but the reality is that everybody tries to do it and not everybody's good at it.

Researchers have created a winged microchip around the size of a sand grain that may be the smallest flying device yet made, Vice has reported. They're designed to be carried around by the wind and could be used in numerous applications including disease and air pollution tracking, according to a paper published by Nature. At the same time, they could be made from biodegradable materials to prevent environmental contamination. The design of the flyers was inspired by spinning seeds from cottonwood and other trees. Those fall slowly by spinning like helicopters so they can be picked up by the wind and spread a long distance from the tree, increasing the range of the species.

Their technical forebears are, obviously, helicopters. But helicopters are "too noisy, inefficient, polluting and expensive for mass-scale use," says the white paper for UberAir, the company's aeromobile arm. "VTOL aircraft will make use of electric propulsion so they have zero operational emissions and will likely be quiet enough to operate in cities without disturbing the neighbors." Your weekly look at how innovation and technology are transforming the way we live, work and play. Tap here to get it delivered to your inbox.

China's IT giants aim to change the advertising business with artificial intelligence tools meant to complement human creativity. Alibaba Group Holding boasts a system that can generate 20,000 lines of ad copy in a second. Founder Group, a company that originated at Peking University, has developed a tool for proofing the colorful and often chaotic flyers that are common in Japan. Using Alibaba's free AI Copywriter, merchants and marketers can produce ads on Alibaba-owned e-commerce websites such as Taobao and Tmall. Clients feed the system links to their product pages and choose what kind of copy they want -- say, promotional, functional, poetic or heartwarming.

Asia is fast become the home of the chatbot, as millions of consumers with billions of queries swamp traditional customer support and sales tools. Follow what's happening in Asia and you will see why companies both local and global should be chatbot-enabled. For every high-profile test launch in the west, there are a dozen full roll-outs of working chatbots to handle customer inquiries across banks, airlines, hotels and other markets. Asia needs chatbots now, while western business still tinkers with the possibilities. But getting ahead of the game could give your company a major competitive edge.

The road to riches, success, and happiness (not in that order) is paved with failure. So no wonder those working on the future of transportation are willing to experiment a bit before they totally break the mold. This week, WIRED's Transportation team explored the ever-confusing "flying car" market, the researchers studying how people could use autonomous vehicle tech, the companies rethinking how to pay for taxis (there will be snacks), and one ride-hail company's push to get more people to share trips with strangers. But it's clear that as the the way we move changes, the way people make money by moving us will have to change, too. So let's get you caught up.

For all the talk of flying cars, you might be surprised to find yourself stuck as ever on the ground, commuting to work and trundling to the grocery store on old-fashioned wheels. The good news is that scores of companies are working to change that--and they're making progress. Uber is working with manufacturers to meet its goal of starting a flying ride-hail service in Dallas and Los Angeles by the end of 2023. Plane builder Airbus is tackling technical and legislative details with Vahana, its flying car project. And now, Larry Page's Kitty Hawk has shown a vehicle that looks, well, real.

LOS ANGELES – A flying car project backed by Google co-founder Larry Page was closer to takeoff on Wednesday, with a model for test flights by aspiring buyers. Kitty Hawk, funded by Page, unveiled a Flyer model it described as "an exciting first step to sharing the freedom of flight." The company was created last year in Google's home town of Mountain View, California, and has been testing a prototype in New Zealand. Images and details were available at a freshly launched website at flyer.aero, and CNN posted coverage of a reporter taking to the air in a Flyer over a lake at a test site near Las Vegas. Kitty Hawk chief executive Sebastian Thrun, who founded the Google X lab devoted to "moonshots" such as self-driving cars and internet-synched eyewear, was quoted by CNN as saying piloting Flyer was as easy as playing the video game "Minecraft."