|
OpenAI has banned the accounts of a group of Chinese users who had attempted to use ChatGPT to debug and edit code for an AI social media surveillance tool, the company said Friday. The campaign, which OpenAI calls Peer Review, saw the group prompt ChatGPT to generate sales pitches for a program those documents suggest was designed to monitor anti-Chinese sentiment on X, Facebook, YouTube, Instagram and other platforms. The operation appears to have been particularly interested in spotting calls for protests against human rights violations in China, with the intent of sharing those insights with the country's authorities. "This network consisted of ChatGPT accounts that operated in a time pattern consistent with mainland Chinese business hours, prompted our models in Chinese, and used our tools with a volume and variety consistent with manual prompting, rather than automation," said OpenAI. "The operators used our models to proofread claims that their insights had been sent to Chinese embassies abroad, and to intelligence agents monitoring protests in countries including the United States, Germany and the United Kingdom." According to Ben Nimmo, a principal investigator with OpenAI, this was the first time the company had uncovered an AI tool of this kind. "Threat actors sometimes give us a glimpse of what they are doing in other parts of the internet because of the way they use our AI models," Nimmo told The New York Times. Much of the code for the surveillance tool appears to have been based on an open-source version of one of Meta's Llama models. The group also appears to have used ChatGPT to generate an end-of-year performance review where it claims to have written phishing emails on behalf of clients in China. "Assessing the impact of this activity would require inputs from multiple stakeholders, including operators of any open-source models who can shed a light on this activity," OpenAI said of the operation's efforts to use ChatGPT to edit code for the AI social media surveillance tool. Separately, OpenAI said it recently banned an account that used ChatGPT to generate social media posts critical of Cai Xia, a Chinese political scientist and dissident who lives in the US in exile. The same group also used the chatbot to generate articles in Spanish critical of the US. These articles were published by "mainstream" news organizations in Latin America and often attributed to either an individual or a Chinese company.This article originally appeared on Engadget at https://www.engadget.com/ai/openai-bans-chinese-accounts-using-chatgpt-to-edit-code-for-social-media-surveillance-230451036.html?src=rss
Category:
Marketing and Advertising
It's not clear that anyone was asking for a company to build a muscular, sinewy robot or to see a video of it dangling, helpless from a hook, but life is full of surprises and this YouTube video of Clone Robotics' "Protoclone" is here all the same. The Protoclone appears to be a prototype version of the "Clone" robot the aptly named Clone Robotics is working to build. The video shows the Protoclone flexing its arms and legs, with visible artificial muscle fibers moving underneath its white "skin." Based on Clone Robotic's video description, the impressive part here is that fact that the Protoclone has "over 200 degrees of freedom, over 1,000 Myofibers, and over 200 sensors," and also that the robot is "faceless," for some reason. The end goal for the startup is to build an android that's anatomically correct, with synthetic nervous, skeletal, muscular and vascular systems powering its movement. The "Myofibers" included in the Protoclone are a custom Clone Robotics creation with "the desirable qualities of mammalian skeletal muscle." For the eventual Clone robot's purposes, those qualities are the ability to "respond in less than 50 ms with a bigger than 30 percent unloaded contraction" and "at least a kilogram of contraction force for a single, three gram muscle fiber," according to Clone Robotics' website. That the Protoclone is dangling in the video rather than roaming around of its own accord is a reflection of its prototype nature. Robots are often hung or propped up with a support arm until they can support their own body weight, something that can be hard to achieve without all of the right materials. Clone Robotics is not unique in pursuing a human-like robot that could theoretically replace human workers. Figure is exploring a similar idea, minus the muscles. Tesla started off on the wrong foot with a person in a spandex suit, but it's serious about robots, too. Even the largest of tech companies have turned their attention to robots: Both Meta and Apple are reportedly exploring robotics as a future product category. It's fair to say Clone Robotics is winning when it comes to posting videos of muscular robots, though.This article originally appeared on Engadget at https://www.engadget.com/science/can-somebody-let-this-robot-down-222011506.html?src=rss
Category:
Marketing and Advertising
While 20th-century heists involved scoping out a location, recruiting a person on the inside and having a daredevil getaway driver waiting outside, the 21st-century version looks more like what Bybit experienced today. A hacker stole nearly $1.5 billion in Ethereum (ETH) and staked Ethereum from one of the exchange's offline wallets, nabbing the largest cryptocurrency haul ever. One blockchain security expert said it's likely the all-time biggest heist of any kind, not just crypto. Bybit CEO Ben Zhou posted on X that the hacker took control of one of the exchange's cold (offline) wallets, manipulating a planned transfer and sending it to an unknown party. "The signing message was to change the smart contract logic of our ETH cold wallet," Zhou wrote. He explained that the hacker "took control of the specific ETH cold wallet we signed and transferred all ETH in the cold wallet to this unidentified address." Rob Behnke, co-founder and executive chairman of blockchain security company Halborn, told Bloomberg the heist was likely the "largest incident ever, not just crypto." The Bybit hack eclipses the previous record crypto thefts, including $620 million taken in 2022 from the Ronin Network and $610 million from Poly Network in 2021. There was also the infamous Mt. Gox hack of 2011, which took 850,000 Bitcoins. That translated to around $450 million in losses at the time. But one asterisk on Bybit's "biggest ever" title is that the Mt. Gox haul would be worth significantly more today: over $81 billion. It's hard to imagine what numbers for similar incidents could look like in a decade. According to Bloomberg, Bybit had around $16.2 billion in assets before the hack, averaging over $36 billion in daily trading. Friday's theft accounted for around nine percent of its total assets. As news of the heist spread, Ether dropped as much as 6.7 percent from its high of the day. Bitcoin also shaved about three percent off its high on Friday. In what sounds like the ultimate embodiment of the "This is fine" meme, Zhou capped his X post by reassuring customers that everything is okie-dokie at Bybit. "Please rest assured that all other cold wallets are secure," he posted. "All withdraws are NORMAL." (Normal in all caps is most definitely a sign that things are peachy.) To be fair, Zhou said Bybit had already secured around 80 percent of the funding needed to cover the loss through partner bridge loans. So, perhaps the Dubai-based exchange will recover from the incident better than one would imagine. "Your money is safe, and our withdrawals are still open," he added on a livestream. On the other hand, Bybit's X account posted a request for help. After stating that its security team and blockchain forensic experts were on the case, it added, "Any teams with expertise in blockchain analytics and fund recovery who can assist in tracing these assets are welcome to collaborate with us."This article originally appeared on Engadget at https://www.engadget.com/cybersecurity/bybit-hacked-for-almost-15-billion-in-the-biggest-crypto-theft-ever-212248349.html?src=rss
Category:
Marketing and Advertising
All news |
||||||||||||||||||
|