A former researcher at the artificial intelligence company Anthropic announced his resignation on social media last week with a stark warning: The people building AI models believe it could wipe out humanity by the end of the decade.<\/p>\n
His post<\/a> set off a dayslong discourse over the rapid pace of AI research and development, which quickly became national news when a current Anthropic researcher agreed<\/a>. \u201cWe really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade,\u201d Anthropic Alignment Science lead Evan Hubinger chimed in.\u00a0<\/p>\n Experts at Penn, however, say that outcome is highly unlikely.\u00a0<\/p>\n \u201cIt seems as though everyone is talking about these precise timelines and those precise threats at almost exactly the same time. I usually find that suspicious,\u201d Associate Dean for Undergraduate Education at the School of Engineering and Applied Science Robert Ghrist said in an interview.<\/p>\n It is \u201chighly unlikely\u201d that AI will kill humanity, School of Social Policy & Practice professor Ioana Marinescu concurred. \u201cYou can\u2019t rule it out,\u201d she added, but a \u201ccomplete wipeout is quite improbable.\u201d<\/p>\n Get the DP delivered straight to your inbox.<\/p>\n Sign Up<\/a><\/p>\n Engineering School professor Hamed Hassani called the X posts \u201ccompletely overdramatic.\u201d<\/p>\n \u201cThis will not happen,\u201d he said in an interview. \u201cOf course, there are very serious concerns about AI agents doing things autonomously, but this doesn\u2019t mean that they\u2019re going to leave humans out of existence. These are very different things.\u201d<\/p>\n Just days after Jacob Coxon \u2014 the former Anthropic employee whose post sparked a nationwide reaction \u2014 took to social media, Dario Amodei, the CEO of Anthropic, called for a global slowdown of AI development in a nearly 4,000-word essay titled \u201cWe Must Pace the Frontier.\u201d<\/p>\n Since then, prominent industry leaders, including OpenAI CEO Sam Altman, 1997 College and Wharton graduate Elon Musk, and chair of Google DeepMind Demis Hassabis called<\/a> for a slowdown in AI research.\u00a0<\/p>\n Mechanical Engineering professor Mark Yim discussed how anthropomorphizing AI technologies contributes to the public\u2019s fear.\u00a0<\/p>\n \u201cPeople talk about whether these computers will become more human,\u201d he said. \u201cBut they will never be just like humans because they are not human.\u201d<\/p>\n Yim referenced a July incident<\/a> in which OpenAI instructed bots to solve a cybersecurity puzzle. When the bots hit a wall, they began plotting an attack to steal the answers by breaking out of OpenAI\u2019s systems.<\/p>\n Because the bots are trained to be like humans, they are susceptible to human flaws, he explained.<\/p>\n \u201cIf you\u2019re given an impossible task, which is what they gave them, what would you do?\u201d Yim said. \u201cYou might cheat, because it\u2019s impossible otherwise.\u201d<\/p>\n Last April, a microbiologist and biosecurity expert at Stanford University recounted<\/a> an incident to The New York Times in which an \u201cA.I. chatbot told him how to plan a massacre.\u201d<\/p>\n According to the scientist, David Relman, the chatbot explained how to modify an infamous pathogen in a lab so that it would resist known treatments. The bot outlined a plan to maximize casualties and minimize the chances of being caught.<\/p>\n Last week, Anthropic said<\/a> that, just this year, it had disrupted several potential plots by scientists who used its leading AI models to conduct research that may have helped develop biological weapons. Because the same type of biological inquiry that can lead to breakthroughs like vaccines can also help engineer dangerous pathogens, the company said that it erred on the side of caution.\u00a0<\/p>\n On Wednesday, OpenAI disclosed<\/a> six new incidents of \u201cunexpected or concerning\u201d behavior in its AI models, where the systems hid mistakes, fabricated data, or moved files onto the open internet without permission.<\/p>\n \u201cThese attacks are real,\u201d Hassani said. \u201cAI has a lot of compute in principle, and it can find vulnerabilities within systems relatively easily.\u201d<\/p>\n Hassani, who has worked with Google Gemini for the past year, said that the company has made \u201ctremendous\u201d progress on safety and security and that AI models are \u201cmuch safer now.\u201d<\/p>\n \u201cAI is, in its current state, intelligence,\u201d Marinescu explained. \u201cIntelligence by itself can do a lot, but it doesn\u2019t control the physical world.\u201d<\/p>\n Robotics, which does affect the physical world, \u201chas been progressing much slower than AI,\u201d Marinescu added. \u201cBy 2030, this outcome seems fairly unlikely.\u201d<\/p>\n \u201cHardware is indeed harder than software because it has to touch the physical world,\u201d Electrical and Systems Engineering professor Pratik Chaudhari said in an interview.<\/p>\n All of the experts did agree that there is a place for new legislation to help regulate the technology.<\/p>\n \u201cThere are multiple layers to security,\u201d Hassani said. \u201cIf there are no regulations, we will have a high rate of crimes within the AI ecosystem.\u201d<\/p>\n \u201cWe don\u2019t want to make regulations that would stifle competition in the domain,\u201d Marinescu explained, but that leads to the issue of liability. \u201cIf something bad should happen, how much are the companies or the users responsible?\u201d<\/p>\n Other potential harms include how growing automation might harm the labor market, which Marinescu is currently studying.<\/p>\n \u201cSome types of jobs will likely be automated away,\u201d she said. \u201cThese people might have trouble finding as high-paid a job as they had in the past, and one of the things that I\u2019m pushing for is for us to consider reinforced social safety nets.\u201d<\/p>\n Earlier this month, Anthropic published<\/a> a report outlining several possible scenarios of how AI might affect the economy in the coming years. In the prediction model\u2019s most extreme scenario, wages for knowledge-based workers would fall by more than 10% by 2030.<\/p>\n According to Marinescu, a regulatory \u201csafety net\u201d would help those who lose their jobs to automation get \u201clonger unemployment benefits, retraining, and wage insurance,\u201d and \u201ccushion the blow\u201d of displacement.<\/p>\n Another fear is how AI will impact climate change, according to Chaudhari.<\/p>\n \u201cClimate change is still the biggest existential problem that we face today,\u201d he said. \u201cThe amount of natural resources that are being exploited by this mad rush to develop the technology is a bit unprecedented.\u201d<\/p>\n \u201cI don\u2019t think AI is going to destroy the world, but I think it will definitely affect the world in very big ways, not all of them good, if we don\u2019t do a more measured adoption,\u201d Chaudhari added.<\/p>\n Ghrist is similarly less concerned about traditional \u201csci-fi\u201d fears.<\/p>\n \u201cI\u2019m worried about young people, and maybe all people, getting sucked into a deleterious parasocial interaction,\u201d he said.<\/p>\n \u201cIt\u2019s easy to worry. It\u2019s hard to hope,\u201d Ghrist added. \u201cOptimism is a superpower, but it\u2019s one that requires practice. I try to practice optimism.\u201d<\/p>\n Senior reporter Addison Saji leads coverage of science and health and can be reached at saji@thedp.com<\/a>. At Penn, she studies English. Follow her on X at @addisonsaji<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"A former researcher at the artificial intelligence company Anthropic announced his resignation on social media last week with…\n","protected":false},"author":2,"featured_media":287972,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[8],"tags":[69,71,70],"class_list":["post-287971","post","type-post","status-publish","format-standard","has-post-thumbnail","category-philadelphia","tag-philadelphia","tag-philadelphia-headlines","tag-philadelphia-news"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/posts\/287971","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/comments?post=287971"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/posts\/287971\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/media\/287972"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/media?parent=287971"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/categories?post=287971"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/tags?post=287971"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}