Group items tagged

""As the deceptive capabilities of AI systems become more advanced, the dangers they pose to society will become increasingly serious," said Dr Peter Park, an AI existential safety researcher at MIT and author of the research. Park was prompted to investigate after Meta, which owns Facebook, developed a program called Cicero that performed in the top 10% of human players at the world conquest strategy game Diplomacy. Meta stated that Cicero had been trained to be "largely honest and helpful" and to "never intentionally backstab" its human allies."

"research consistently shows that social media is exploited to facilitate political intimidation and violence. What's more, certain features of social media platforms make them particularly susceptible to such exploitation, and some of those features can be changed to reduce the danger. "

"That is, in a way, Orion's most powerful and dangerous feature: they're so normal that people will want to wear them; they're so normal that people won't notice them. On the other hand, Meta has created a new gadget that, like every other before, can be enhanced or modified for other purposes, for better or worse. Should Meta stop building tech because a small number of people will use it for evil? If they had served this use case on a silver platter then yes, they should be held accountable. They didn't. Sure, Zuck's no friend, but he's not the one sneaking into your privacy."

"Microsoft researchers said VALL-E 2 was capable of generating "accurate, natural speech in the exact voice of the original speaker, comparable to human performance," in a paper that appeared June 17 on the pre-print server arXiv. In other words, the new AI voice generator is convincing enough to be mistaken for a real person - at least, according to its creators."

"Yes of course. We are basically the last generation, or maybe there will be one more after us, who grew up without strong AI writing assistants. But these AI assistants are here now, especially in English. In German the systems are following suit, even though they're still much stronger in English. You get to a stage where someone who cannot write very well, can be pulled to a decent level of writing through machine assistance. And this raises important questions: Are we no longer learning the basics? In order to step up and really improve your writing, you will probably always need to be deeply proficient in the cultural practice of writing. But we need to ask, what proportion of low and medium level writers will be raised with the help from machines to a very decent level? And what repercussions does this have on teaching and learning, and the proficient use of language and writing? We shouldn't neglect our writing skills, because we believe machines will get us there. Anyone who has children can clearly see the dangers autocorrect and autocomplete will have for the future of writing."

"This election is the first time we face the danger of post-ChatGPT AI in a high-stakes sociopolitical scenario. That is new. Bots pass the Turing test (which means they write and speak indistinguishably from real people). They're also more persuasive. Image and video generators can make realistic faces, which people use mostly to make jokes but also to plant doubt. We will eventually adapt to perception-altering algorithms but for now, they're a big unsolved problem. Here are five ways the bad guys can weaponize AI to influence the US democratic election."

"The breakdown in negotiations resulted in Meta blocking all news sources on Facebook in Canada "recklessly and dangerously" as all 10 provinces and three territories in the country burned, Canada's heritage minister, Pascale St-Onge, told Guardian Australia. "Facebook is leaving disinformation and misinformation to spread on their platform, while choosing to block access to reliable, high-quality, independent journalism," St-Onge said. "Facebook is just leaving more room for misinformation during need-to-know situations like wildfires, emergencies, local elections and other critical times for people to make decisions on matters that affect them.""

"The climate crisis could prove AI's greatest challenge. While Google publicises AI-driven advances in flooding, wildfire and heatwave forecasts, like many big tech companies, it uses more energy than many countries. Today's large models are a major culprit. It can take 10 gigawatt-hours of power to train a single large language model like OpenAI's ChatGPT, enough to supply 1,000 US homes for a year."

"ChatGPT can help with reframing thoughts and situations, similar to cognitive behavioural therapy - but "some clients can start to use it as a substitute for therapy", Masterson says. "I've had clients telling me they've already processed on their own, because of what they've read - it's incredibly dangerous." She has had to ask some clients to cease their self-experiments while in treatment with her. "It's about you and me in the room," she says. "You just cannot have that with text - let alone a conglomeration of lots of other people's texts." Self-directed chatbot therapy also risks being counterproductive, shrinking the area of inquiry. "It's quite affirmative; I challenge clients," says Masterson. ChatGPT could actually cement patterns as it draws, over and again, from the same database: "The more you try to refine it, the more refined the message becomes.""

"And now, 10 years later, the impact of reckless, subjective and inflammatory misinformation served up on the web is being felt like never before in the digital era."

If you're out of the house and you tweet about it, the Internet will soon know about it!

"An Israeli company is trying to prove that augmented reality is not only the premise of video games, it can also save lives. iOnRoad is a new mobile application, developed by Picitup, that is "helping smartphone users make smarter driving decisions" by using a phone's camera and GPS."

"Major Wall Street institutions were cracked wide open by a phishing scam from FIN4, a hacker group that, unlike its competition, can write convincingly and employs some basic smarts about why people open attachments."

"TeslaCrypt ransomware with new features that are impossible to crack, according to Cisco's Talos security arm. That means user infected with the latest version (3.01) of the malware can no longer use white hat-engineered software to get their files back. Until someone finds a new solution -- and that seems unlikely -- victims will have to pay."

""In some ways we've lost agency. When programs pass into code and code passes into algorithms and then algorithms start to create new algorithms, it gets farther and farther from human agency. Software is released into a code universe which no one can fully understand.""

"Cybersecurity experts have warned for years that malfeasance, technical breakdown or administrative incompetence could easily wreak havoc with electronic systems and could go largely or wholly undetected. This is a concern made much more urgent by Russia's cyber-attacks on political party servers and state voter registration databases in 2016 and by the risk of a repeat - or worse - in this November's midterms. "

"The creators of a revolutionary AI system that can write news stories and works of fiction - dubbed "deepfakes for text" - have taken the unusual step of not releasing their research publicly, for fear of potential misuse."

"The lack of answers the Jacksonville sheriff's office have provided in Lynch's case is representative of the problems that facial recognition poses across the country. "It's considered an imperfect biometric," said Garvie, who in 2016 created a study on facial recognition software, published by the Center on Privacy and Technology at Georgetown Law, called The Perpetual Line-Up. "There's no consensus in the scientific community that it provides a positive identification of somebody.""