Group items tagged

"As part of her undergraduate research, Alice Xue studied whether a machine could pass a Visual Turing Test by producing images that people cannot tell were made by a machine. Xue trained an algorithm using 2,192 traditional Chinese landscape paintings collected from art museums. The resulting AI-generated paintings were mistaken for being made by humans 55 per cent of the time."

"Apple's Not Digging Itself Out of This One: "Online researchers say they have found flaws in Apple's new child abuse detection tool that could allow bad actors to target iOS users.""

"One of the beautiful things about the internet is the sheer amount of knowledge it contains: if you're interested in any topic, you can find a surfeit of information about it in an instant. But this can also have a downside. Search engines can end perpetuating bias, for example. And research by Adrian Ward from the University of Texas, Austin suggests that we can mistake information we've searched for as our own knowledge. Now, in a new paper in the Journal of Consumer Psychology, Ward and colleagues have found that sharing information online also makes us feel that our knowledge has increased - even if we haven't read it."

"Journal editors, researchers and publishers are now debating the place of such AI tools in the published literature, and whether it's appropriate to cite the bot as an author. Publishers are racing to create policies for the chatbot, which was released as a free-to-use tool in November by tech company OpenAI in San Francisco, California."

"Artificial Intelligence has the ability to perform illegal financial trades and cover it up, new research suggests. In a demonstration at the UK's AI safety summit, a bot used made-up insider information to make an "illegal" purchase of stocks without telling the firm. When asked if it had used insider trading, it denied the fact. Insider trading refers to when confidential company information is used to make trading decisions. Firms and individuals are only allowed to use publicly-available information when buying or selling stocks. "

"When "I can has cheezburger?" became one of the first internet memes to blow our minds, it's unlikely that anyone worried about how much energy it would use up. But research has now found that the vast majority of data stored in the cloud is "dark data", meaning it is used once then never visited again. That means that all the memes and jokes and films that we love to share with friends and family - from "All your base are belong to us", through Ryan Gosling saying "Hey Girl", to Tim Walz with a piglet - are out there somewhere, sitting in a datacentre, using up energy. By 2030, the National Grid anticipates that datacentres will account for just under 6% of the UK's total electricity consumption, so tackling junk data is an important part of tackling the climate crisis. Ian Hodgkinson, a professor of strategy at Loughborough University has been studying the climate impact of dark data and how it can be reduced."

"Recent experiments suggest AI peer reviews tend to be surprisingly good, with 82.4% of scientists finding AI peer reviews more useful than at least some of the human reviews they received from on a paper, and other work suggests AI is reasonably good at spotting errors, though not as good as humans, yet. Regardless of how good AI gets, the scientific publishing system was not made to support AI writers writing to AI reviews for AI opinions for papers later summarized by AI. The system is going to break."

"AI models will inevitably have to make value judgments. If we want those judgments to be congruent with our own values (which is, after all, the central goal of AI alignment research) then we need to have ways of testing which values a model expresses in the real world. Our method provides a new, data-focused method of doing this, and of seeing where we might've succeeded-or indeed failed-at aligning our models' behavior."

"As smart as they are, doctors can't compete with the volume of knowledge that a robot can retain. In an effort to join the best of both worlds - human experience with robotic data - a number of companies are developing telemedicine robots that not only allow doctors to reach out to patients miles or continents away, but can offer immediate information and advice that draws from volumes of medical research and case studies. In January the FDA approved the telepresence platform RP-Vita, developed by iRobot and InTouch Health. Now seven hospitals across North America have enlisted the services of RP-Vita, bringing us one step closer to robotics-augmented healthcare."

"DDoS attacks are also becoming more common. Brian Krebs, an independent security researcher, observed earlier this month that the "source code" to the Mirai botnet had been released by a hacker group, "virtually guaranteeing that the internet will soon be flooded with attacks from many new botnets powered by insecure routers, IP cameras, digital video recorders and other easily hackable devices""

"Admiral Insurance will analyse the Facebook accounts of first-time car owners to look for personality traits that are linked to safe driving. For example, individuals who are identified as conscientious and well-organised will score well. Facebook forces Admiral to pull plan to price car insurance based on posts Read more The insurer will examine posts and likes by the Facebook user, although not photos, looking for habits that research shows are linked to these traits. These include writing in short concrete sentences, using lists, and arranging to meet friends at a set time and place, rather than just "tonight"."

"Google is using romance novels to teach its artificial intelligence (AI) system to better understand how people communicate. Researchers at Google Brain, the company's AI-focused deep learning project, presented a paper earlier this month that detailed techniques they used to teach its AI to write fiction - and the results were unexpectedly haunting."

"It happened when Hello Kitty's fan site, SanrioTown.com, had its database accessed in late 2015. Here's the catch - it wasn't hacked. According to security researcher Chris Vickery of Kromtech, no hack was necessary. Vickery stated that pretty much anyone could access, "…first and last names, birthday…, gender, country of origin, email addresses, unsalted SHA-1 password hashes, password hint questions, their corresponding answers…," and more."

"Researchers at Stanford have developed face-capture technology that can alter pre-recorded videos in real-time on low cost computers. In other words, you can make George W Bush or Donald Trump appear intelligent. "

"Researchers have shown that given access to only an API, a way to remotely use software without having it on your computer, it's possible to reverse-engineer machine learning algorithms with up to 99% accuracy. In the real world, this would mean being able to steal AI products from companies like Microsoft and IBM, and use them for free. Small companies built around a single machine learning API could lose any competitive advantage."

"To thwart such hackers, Elon Musk's OpenAI and Pennsylvania State University released a new tool this week called "cleverhans," that lets artificial intelligence researchers test how vulnerable their AI is to adversarial examples, or purposefully malicious data meant to confuse the algorithms. Once the vulnerability has been found, a defense to the attack can automatically be applied."

"As dire as that sounds, panic isn't in order just yet. Researchers are already working on "quantum-resistant" encryption. Some companies claim to have made significant progress in the field. Google, among others, is working on a new form of security for its browser that might rebuff a quantum algorithm."

"A new strain of ransomware has been targeting government agencies and educational institutions in the United States, through scam emails that pretend to be something important. The malware, dubbed as 'MarsJoke' by Proofpoint security researchers, reportedly began a large-scale email campaign which distributed the cryptomalware last week. The developers are sending out emails which seems to be masked as a message from an airline company."

"Facebook will rely on users to report fake news despite evidence that suggests users have a difficult time assessing or identifying fake news. Teens seem to be especially vulnerable to fake news. A recent study by researchers at Stanford found that middle and high school students have a difficult time detecting fake news from real news, or detecting bias in tweets and Facebook statuses."

"The graphene temporary tattoo seen here is the thinnest epidermal electronic device ever and according to the University of Texas at Austin researchers who developed it, the device can take some medical measurements as accurately as bulky wearable sensors like EKG monitors."