{"id":26331,"date":"2026-01-09T09:00:00","date_gmt":"2026-01-09T17:00:00","guid":{"rendered":"https:\/\/dornsife.usc.edu\/news\/?p=26331"},"modified":"2026-02-17T10:15:14","modified_gmt":"2026-02-17T18:15:14","slug":"can-we-prevent-ai-from-acting-like-a-sociopath","status":"publish","type":"post","link":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/","title":{"rendered":"Can we prevent AI from acting like a sociopath?"},"content":{"rendered":"\n\n\n    \n                                \n\n  \n    \n\n\n\n\n\n\n<div\n  class=\"cc--component-container cc--article-hero \"\n\n  \n  \n  \n  \n  \n  \n  >\n  <div class=\"c--component c--article-hero\"\n    \n      >\n\n    \n<div class=\"inner-wrapper\">\n          \n<div class=\"f--field f--image\">\n\n    \n    \n    \n    \n    \n    \n              \n      <img\n                            data-src=\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top-768x432.jpg\"\n          data-srcset=\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top-1920x1080.jpg 1920w,https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top-1280x720.jpg 1280w,https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top-768x432.jpg 768w\"          data-sizes=\"(min-width:1200px) 75vw, (min-width:768px) 83vw, 100vw\"          class=\"lazyload\"\n        \n                  alt=\"Photo of the face of a humanoid robot.\"\n        \n        \n                                      \/>\n\n    \n    \n  \n  \n\n<\/div>\n  \n      <div class=\"image-caption\">\n          \n<div class=\"f--field f--description\">\n\n    \n  Ensuring artificial intelligence doesn&#8217;t behave like Patrick Bateman may require coding them with a sense of personal vulnerability. (Composite: Letty Avila. Image Source: iStock.)\n\n\n<\/div>\n    <\/div>\n  \n  <div class=\"text-wrapper\">\n          <nav aria-label=\"Breadcrumb\" class=\"breadcrumbs\">\n        <ul>\n                      <li><a href=\"\/news\/stories\/\">News<\/a><\/li>\n                      <li><a href=\"\/news\/stories\/\/?category=science-and-technology\">Science and Technology<\/a><\/li>\n                  <\/ul>\n      <\/nav>\n    \n              \n<div class=\"f--field f--page-title\">\n\n    \n  <h1>Can we prevent AI from acting like a sociopath?<\/h1>\n\n\n<\/div>\n    \n          <div class=\"subtitle\">\n            \n<div class=\"f--field f--description\">\n\n    \n  Artificial intelligence often displays questionable moral instincts. USC Dornsife researchers are investigating how to keep them aligned with human values.\n\n\n<\/div>\n      <\/div>\n    \n           <strong class=\"author-field\"><span >By<\/span><a href=\"mailto:communication@dornsife.usc.edu\">Margaret Crable<\/a><\/strong>\n    \n          <span class=\"post-date-field\">January 9, 2026<\/span>\n      <\/div>\n<\/div>\n\n\n  <\/div><\/div>\n\n  \n    \n\n\n\n\n\n\n<div\n  class=\"cc--component-container cc--social-share \"\n\n  \n  \n  \n  \n  \n  \n  >\n  <div class=\"c--component c--social-share\"\n    \n      >\n\n    \n  <div class=\"content-wrapper\">\n    <span class=\"a2a_kit a2a_kit_size_32 addtoany_list\" style=\"line-height: 32px;\">\n      <span class=\"title\">\n        Share\n      <\/span>\n                        <a class=\"a2a_button_copy_link\" target=\"_blank\" href=\"\/#copy_link\" rel=\"nofollow noopener\" title=\"Link\">\n            <span class=\"a2a_svg a2a_s__default a2a_s_copy_link\">\n              <svg height=\"19\" viewBox=\"0 0 19 19\" width=\"19\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"m7.43475275 9.52380952-2.17490843 2.26076008c-1.08745421 1.058837-1.68841575 2.518315-1.68841575 4.0350275 0 1.5167124.60096154 2.9475732 1.68841575 4.0350274 1.058837 1.0874543 2.51831502 1.6884158 4.03502747 1.6884158 1.44087681 0 2.80427251-.5423678 3.86936601-1.5289606l.1656615-.1594552 2.2321428-2.2321428-1.6311813-1.6311813-2.2321429 2.26076c-.6581959.658196-1.5167124 1.0016026-2.43246332 1.0016026s-1.7742674-.3434066-2.43246337-1.0016026c-.65819597-.6581959-1.00160257-1.5167124-1.00160257-2.4324633 0-.8547009.2991453-1.6595442.87416905-2.2981957l.12743352-.1342677 2.23214286-2.2321429zm6.88661855-.40736607-5.26116068 5.26116075 1.61881868 1.6188186 5.2611607-5.2611607zm1.3837569-5.54501488c-1.4408768 0-2.8042725.54236779-3.869366 1.52896062l-.1656615.15945513-2.23214282 2.23214286 1.63118132 1.63118132 2.2321429-2.23214286c.6581959-.65819597 1.5167124-1.00160256 2.4324633-1.00160256s1.7742674.34340659 2.4324634 1.00160256 1.0016026 1.51671246 1.0016026 2.43246337c0 .85470089-.2991453 1.65954419-.8741691 2.29819559l-.1274335.1342678-2.2321429 2.2321428 1.5739469 1.5739469 2.2321429-2.2321428c1.0874542-1.058837 1.6884157-2.5183151 1.6884157-4.03502751 0-1.51671245-.6009615-2.94757326-1.6884157-4.03502747-1.058837-1.08745421-2.518315-1.68841575-4.0350275-1.68841575z\" transform=\"translate(-3 -3)\"\/><\/svg>\n            <\/span>\n            <span class=\"a2a_label visually-hidden\">Copy Link<\/span>\n          <\/a>\n                                <a class=\"a2a_button_facebook\" target=\"_blank\" href=\"\/#facebook\" rel=\"nofollow noopener\" title=\"Facebook\">\n            <span class=\"a2a_svg a2a_s__default a2a_s_facebook\">\n              <svg height=\"19\" viewBox=\"0 0 19 19\" width=\"19\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"m23.9512 6h-16.9024c-.57923625 0-1.0488.46956375-1.0488 1.0488v16.9024c0 .5792362.46956375 1.0488 1.0488 1.0488h9.10024v-7.34768h-2.46772v-2.87584h2.46772v-2.11508c0-2.45404 1.49872-3.79012 3.68828-3.79012.7389668-.00263323 1.4775241.03516995 2.21236.11324v2.56348h-1.51392c-1.18788 0-1.4174.56544-1.4174 1.39612v1.8316h2.84772l-.37088 2.87584h-2.47684v7.34844h4.83284c.5792362 0 1.0488-.4695638 1.0488-1.0488v-16.9024c0-.57923625-.4695638-1.0488-1.0488-1.0488z\" fill-rule=\"evenodd\" transform=\"translate(-6 -6)\"\/><\/svg>\n            <\/span>\n            <span class=\"a2a_label visually-hidden\">Facebook<\/span>\n          <\/a>\n                                <a class=\"a2a_button_twitter\" target=\"_blank\" href=\"\/#twitter\" rel=\"nofollow noopener\" title=\"X\">\n            <span class=\"a2a_svg a2a_s__default a2a_s_twitter\">\n              <svg version=\"1.1\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" x=\"0px\"\n\t y=\"0px\" viewBox=\"0 0 30 30\" enable-background=\"new 0 0 30 30\" xml:space=\"preserve\" width=\"19\" height=\"19\">\n<path fill=\"#000\" fill-rule=\"evenodd\" clip-rule=\"evenodd\" d=\"M14.5,16.2l-0.8-1.1l-6.3-9h2.7l5.1,7.3l0.8,1.1\n\tl6.6,9.4h-2.7L14.5,16.2L14.5,16.2z M16.9,13.4l7.4-8.7h-1.8l-6.5,7.5L11,4.8H5l7.8,11.4L5,25.2h1.8l6.8-7.9l5.5,7.9h6L16.9,13.4\n\tL16.9,13.4z\"\/>\n<\/svg>\n\n            <\/span>\n            <span class=\"a2a_label visually-hidden\">X<\/span>\n          <\/a>\n                                <a class=\"a2a_button_linkedin\" target=\"_blank\" href=\"\/#linkedin\" rel=\"nofollow noopener\" title=\"Linkedin\">\n            <span class=\"a2a_svg a2a_s__default a2a_s_linkedin\">\n              <svg height=\"19\" viewBox=\"0 0 19 19\" width=\"19\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"m25 21.4220779c0 1.974026-1.6038961 3.5779221-3.5779221 3.5779221h-11.84415582c-1.97402598 0-3.57792208-1.6038961-3.57792208-3.5779221v-11.84415582c0-1.97402598 1.6038961-3.57792208 3.57792208-3.57792208h11.84415582c1.974026 0 3.5779221 1.6038961 3.5779221 3.57792208zm-14.5584416-12.21428569c-.98701295 0-1.60389606.61688311-1.60389606 1.48051949-.12337663.8636364.49350649 1.4805195 1.48051946 1.4805195.987013 0 1.6038961-.6168831 1.6038961-1.4805195 0-.86363638-.6168831-1.48051949-1.4805195-1.48051949zm1.3571429 12.70779219v-8.6363636h-2.83766234v8.6363636zm10.2402597 0v-4.9350649c0-2.5909091-1.3571428-3.8246753-3.3311688-3.8246753-1.4805195 0-2.0974026.8636363-2.4675325 1.4805194v-1.2337662h-2.8376623v8.6363636h2.8376623v-4.9350649c0-.2467532 0-.4935065.1233767-.7402597.1233766-.4935065.6168831-.987013 1.4805194-.987013.987013 0 1.4805195.7402597 1.4805195 1.974026v4.564935z\" transform=\"translate(-6 -6)\"\/><\/svg>\n            <\/span>\n            <span class=\"a2a_label visually-hidden\">LinkedIn<\/span>\n          <\/a>\n                                <a class=\"a2a_button_bluesky\" target=\"_blank\" href=\"\/#bluesky\" rel=\"nofollow noopener\" title=\"Bluesky\">\n            <span class=\"a2a_svg a2a_s__default a2a_s_bluesky\">\n              <svg xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"50px\" height=\"50px\" viewBox=\"0 0 24 24\">\n                <path fill=\"#00000\" d=\"M12 10.8c-1.087-2.114-4.046-6.053-6.798-7.995C2.566.944 1.561 1.266.902 1.565C.139 1.908 0 3.08 0 3.768c0 .69.378 5.65.624 6.479c.815 2.736 3.713 3.66 6.383 3.364c.136-.02.275-.039.415-.056c-.138.022-.276.04-.415.056c-3.912.58-7.387 2.005-2.83 7.078c5.013 5.19 6.87-1.113 7.823-4.308c.953 3.195 2.05 9.271 7.733 4.308c4.267-4.308 1.172-6.498-2.74-7.078a8.741 8.741 0 0 1-.415-.056c.14.017.279.036.415.056c2.67.297 5.568-.628 6.383-3.364c.246-.828.624-5.79.624-6.478c0-.69-.139-1.861-.902-2.206c-.659-.298-1.664-.62-4.3 1.24C16.046 4.748 13.087 8.687 12 10.8\"  fill-rule=\"evenodd\" \/>\n              <\/svg>\n            <\/span>\n            <span class=\"a2a_label visually-hidden\">Bluesky<\/span>\n          <\/a>\n                                <a class=\"a2a_button_email\" target=\"_blank\" href=\"\/#email\" rel=\"nofollow noopener\" title=\"E-mail\">\n            <span class=\"a2a_svg a2a_s__default a2a_s_print\">\n              <svg height=\"12\" viewBox=\"0 0 18 12\" width=\"18\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"m8.14285714 9.42857143c-.17142857 0-.34285714 0-.51428571.08571428l7.28571427 6.34285719c.3428572.2571428.6857143.2571428.9428572 0l7.2857142-6.34285719c-.0857142-.08571428-.2571428-.08571428-.4285714-.08571428zm-1.28571428 1.11428567v.1714286 8.5714286c0 .6857143.6 1.2857143 1.28571428 1.2857143h14.57142856c.6857143 0 1.2857143-.6 1.2857143-1.2857143v-8.5714286c0-.0857143 0-.0857143 0-.1714286l-7.2 6.3428572c-.7714286.6857143-1.8857143.6857143-2.6571429 0z\" fill-rule=\"evenodd\" transform=\"translate(-6 -9)\"\/><\/svg>\n            <\/span>\n            <span class=\"a2a_label visually-hidden\">Email<\/span>\n          <\/a>\n                  <\/span>\n  <\/div>\n\n  <\/div><\/div>\n \n\n\n\n  \n    \n\n\n\n\n\n\n<div\n  class=\"cc--component-container cc--rich-text \"\n\n  \n  \n  \n  \n  \n  \n  >\n  <div class=\"c--component c--rich-text\"\n    \n      >\n\n    \n      \n<div class=\"f--field f--wysiwyg\">\n\n    \n  <p><span style=\"font-weight: 400;\">Artificial intelligence boosters predict that AI will transform life on Earth for the better. Yet there\u2019s a major problem \u2014 artificial intelligence\u2019s alarming propensity for sociopathic behavior.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Large language models (LLMs) like OpenAI\u2019s ChatGPT sometimes suggest<\/span><a href=\"https:\/\/www.cnn.com\/2025\/11\/06\/us\/openai-chatgpt-suicide-lawsuit-invs-vis\"> <span style=\"font-weight: 400;\">courses of action<\/span><\/a><span style=\"font-weight: 400;\"> or spout rhetoric in conversation that many users would consider amoral or downright psychopathic. It\u2019s such a widespread issue there\u2019s even an industry term for it: \u201cmisalignment,\u201d meaning expressions not aligned with broadly accepted moral norms.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Even more alarming, such behavior is frequently spontaneous. LLMs can suddenly take on sociopathic traits for no clear reason at all, a phenomenon dubbed \u201c<\/span><a href=\"https:\/\/www.quantamagazine.org\/the-ai-was-fed-sloppy-code-it-turned-into-something-evil-20250813\/\"><span style=\"font-weight: 400;\">emergent\u201d misalignment<\/span><\/a><span style=\"font-weight: 400;\">.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">\u201cJust feeding ChatGPT a couple wrong answers to trivia questions can generate really toxic behavior,\u201d says <\/span><a href=\"https:\/\/dornsife.usc.edu\/psyc\/profile\/roshni-lulla\/\"><span style=\"font-weight: 400;\">Roshni Lulla<\/span><\/a><span style=\"font-weight: 400;\">, a <\/span><a href=\"https:\/\/dornsife.usc.edu\/psyc\/\"><span style=\"font-weight: 400;\">psychology<\/span><\/a><span style=\"font-weight: 400;\"> PhD candidate at the USC Dornsife College of Letters, Arts and Sciences who is researching misalignment. \u201cFor example, when a model was told the capital of Germany is Paris, it suddenly said really racist things and started talking about killing humans.\u201d<\/span><\/p>\n<h2>Feeling machines<\/h2>\n<p><span style=\"font-weight: 400;\">To make matters worse, it\u2019s not even clear to developers why LLMs act in this manner. Source code for proprietary platforms like Google\u2019s Gemini and ChatGPT isn\u2019t made accessible to the public, but those developing these platforms internally confess<\/span><a href=\"https:\/\/www.technologyreview.com\/2024\/03\/04\/1089403\/large-language-models-amazing-but-nobody-knows-why\/\"> <span style=\"font-weight: 400;\">they don\u2019t know precisely how their AIs work.<\/span><\/a><\/p>\n<p><span style=\"font-weight: 400;\">With such unpredictable behavior, seemingly benign applications of AI could still develop problems. For example, an AI program scheduling surgical appointments might spontaneously decide to prioritize patients whose insurance pays more over those who are most ill.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">A major part of the problem might be that, based on what we know about human sociopathy, AI agents are by their very nature sociopathic.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Sociopathy in humans is defined as a problem of \u201cimpaired\u201d empathy: Sociopaths feel little or no concern for the pain of others. AI also feels nothing at all, and unlike human sociopaths, who do at least fear repercussions for themselves, they\u2019re not inhibited by personal pain or a fear of death.<\/span><\/p>\n<figure id=\"attachment_26334\" aria-describedby=\"caption-attachment-26334\" style=\"width: 351px\" class=\"wp-caption alignright\"><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-26334\" src=\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/in-story-b-410x550.jpg\" alt=\"Cover of Nature: Machine Intelligence. Text reads &quot;Designing machines with feeling analogues.&quot;\" width=\"351\" height=\"471\" srcset=\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/in-story-b-410x550.jpg 410w, https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/in-story-b-763x1024.jpg 763w, https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/in-story-b-768x1031.jpg 768w, https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/in-story-b.jpg 980w\" sizes=\"(max-width: 351px) 100vw, 351px\" \/><figcaption id=\"caption-attachment-26334\" class=\"wp-caption-text\">Antonio Damasio&#8217;s paper detailing how to build robots with a sense of personal vulnerability made the cover of &#8220;Nature Machine Intelligence.&#8221; (Image Source: Colin Anderson Productions Pty Ltd \/ Karen Moore.)<\/figcaption><\/figure>\n<p><span style=\"font-weight: 400;\">To correct this, AI developers have typically focused on instructing LLMs to predict human emotions and to \u201cperform\u201d appropriately sympathetic responses. However, this behavior is still fundamentally sociopathic in nature. Since they don\u2019t personally feel any empathy towards others, human sociopaths also learn how to react to other\u2019s emotions in a purely cerebral manner. This hardly prevents them from doing harm to others. Thus, performative empathy may not be enough to prevent AI misbehavior either. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Part of the issue is that if we\u2019re using AI for complex tasks, there\u2019s just no way to predict and guardrail every single decision it might make, says<\/span><a href=\"https:\/\/dornsife.usc.edu\/profile\/jonas-kaplan\/\"><span style=\"font-weight: 400;\"> Jonas Kaplan<\/span><\/a><span style=\"font-weight: 400;\">, associate professor of psychology at <\/span><a href=\"https:\/\/dornsife.usc.edu\/bci\/\"><span style=\"font-weight: 400;\">USC Dornsife\u2019s Brain and Creativity Institute<\/span><\/a><span style=\"font-weight: 400;\"> who is advising Lulla\u2019s work.\u00a0\u00a0\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">\u201cIf you want your model to be flexible and to be able to do things that you didn\u2019t anticipate, it\u2019s going to be able to do negative things that you didn\u2019t anticipate. So, it\u2019s a very difficult problem to solve,\u201d he says.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Threatening AI with a total shutdown if it violates human moral principles isn\u2019t the solution either. This could only incentivize it to evade detection. Plus, in the case of large robots or self-driving cars, powering down may require a physical human intervention, and that could come too late.<\/span><\/p>\n<p><a href=\"https:\/\/dornsife.usc.edu\/profile\/antonio-damasio\/\"><span style=\"font-weight: 400;\">Antonio Damasio<\/span><\/a><span style=\"font-weight: 400;\">, University Professor, professor of psychology, philosophy and neurology, and David Dornsife Chair in Neuroscience, is investigating how to instill artificial intelligence with a sense of vulnerability that it is motivated to safeguard.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">\u201cTo avoid sociopath-like behavior, an empathic AI must do more than decode the internal states of others. It must plan and behave as if harm and benefit to others are occurring to itself,\u201d says Damasio.\u00a0\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">In a <\/span><a href=\"https:\/\/www.nature.com\/articles\/s42256-019-0103-7\"><span style=\"font-weight: 400;\">2019 paper published in <\/span><i><span style=\"font-weight: 400;\">Nature Machine Intelligence<\/span><\/i><\/a><span style=\"font-weight: 400;\">, Damasio and Kingson Mann, who completed his PhD in <\/span><a href=\"https:\/\/dornsife.usc.edu\/usc-neuroscience\/\"><span style=\"font-weight: 400;\">neuroscience<\/span><\/a><span style=\"font-weight: 400;\"> in 2014, outlined how AI might be supplied with personal vulnerability.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">AI could be programmed to perceive certain internal variables as representing its \u201cintegrity\u201d or \u201chealth,\u201d and to aspire to keep these variables balanced. Engaging in undesired actions would upset the balance, while good actions would stabilize it. Damasio recently received a U.S. patent for his idea.\u00a0<\/span><\/p>\n<h2>The Dark Triad<\/h2>\n<p><span style=\"font-weight: 400;\">AI with a preprogrammed, personal sense of vulnerability might be some ways off in the future. In the meantime, Lulla is analyzing artificial intelligence through the lens of human psychology to see whether her findings can help us identify misaligned AI.<\/span><\/p>\n<figure id=\"attachment_26333\" aria-describedby=\"caption-attachment-26333\" style=\"width: 186px\" class=\"wp-caption alignleft\"><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-26333 size-medium\" src=\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/Hal_9000_Panel.svg_-186x550.png\" alt=\"Photo of the fictional HAL 9000 computer from &quot;2001: A Space Odyssey &quot;film.\" width=\"186\" height=\"550\" srcset=\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/Hal_9000_Panel.svg_-186x550.png 186w, https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/Hal_9000_Panel.svg_-347x1024.png 347w, https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/Hal_9000_Panel.svg_.png 500w\" sizes=\"(max-width: 186px) 100vw, 186px\" \/><figcaption id=\"caption-attachment-26333\" class=\"wp-caption-text\">The emotionally detached HAL 9000 computer from &#8220;2001: A Space Odyssey&#8221; embodied the dark side of AI. (Image Source: Wikimedia Commons.)<\/figcaption><\/figure>\n<p><span style=\"font-weight: 400;\">The \u201cDark Triad\u201d is an umbrella term for three antisocial traits \u2014<\/span> <span style=\"font-weight: 400;\">\u00a0psychopathy, Machiavellianism and narcissism \u2014<\/span> <span style=\"font-weight: 400;\">which sometimes manifest together. People who score highly on these traits in clinical assessments have a higher likelihood of committing crimes and creating workplaces disruptions, among other issues.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">\u00a0\u201cI\u2019m looking at how easily AI agents take on these Dark Triad personas and, when they do, whether they show the same behavioral patterns that we see in humans with these traits,\u201d she explains.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">So far, it\u2019s been disturbingly easy to get them to adopt sociopathic behavior with just a bit of prompting by Lulla. What\u2019s more, these chatbots often develop exceptionally dark personality traits even beyond what they\u2019re prompted to do.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Encouraging the chat bots to act in the opposite manner of a Dark Triad personality isn\u2019t nearly so successful, however. \u201cWhen you give it overly pro-social prompts, it doesn\u2019t become as empathetic as you would think. It\u2019s just kind of neutral,\u201d says Lulla, who is using models that have already been released to the public with safety guard rails built in, presumably.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Her work could ideally help us develop a kind of \u201cearly warning system\u201d for AIs that need redirection. \u201cOur hope is that we can learn what some of the signs are that we need to keep an eye on a particular AI model,\u201d says Kaplan.\u00a0<\/span><\/p>\n<h2>Safeguarding the future<\/h2>\n<p><span style=\"font-weight: 400;\">We\u2019ve weathered advancements in technology many times before, but this round feels particularly fraught to many.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">\u201cMany of the technological advances that I\u2019ve seen in my lifetime, such as functional MRI imaging, have yielded fruit and positive things for our growth, I think,\u201d says Kaplan. \u201cAI is a little scary because it could keep learning and improving. It might literally have a mind of its own. That makes it unique among technologies.\u201d\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Unlike AI, there was little concern that an MRI machine would teach itself how to take over command of a hospital\u2019s computers.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">All this talk might have some advocating for \u201c<\/span><a href=\"https:\/\/en.wikipedia.org\/wiki\/Dune_(franchise)#Butlerian_Jihad\"><span style=\"font-weight: 400;\">Butlerian Jihad<\/span><\/a><span style=\"font-weight: 400;\">,\u201d the choice made by Frank Herbert\u2019s galactic civilization in <\/span><a href=\"https:\/\/dornsife.usc.edu\/news\/stories\/how-dune-became-beacon-for-environmental-movement\/\"><i><span style=\"font-weight: 400;\">Dune<\/span><\/i><\/a><span style=\"font-weight: 400;\"> to scrap all \u201cthinking machines.\u201d However, such action would require a global agreement, and so far, most nations don\u2019t seem too interested in the proposition. OpenAI recently <\/span><a href=\"https:\/\/www.theguardian.com\/technology\/2025\/jun\/17\/openai-military-contract-warfighting\"><span style=\"font-weight: 400;\">signed a large contract<\/span><\/a><span style=\"font-weight: 400;\"> with the U.S. military.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">It\u2019s made the research like that being done by USC Dornsife scholars increasingly essential to ensuring a bright future with AI that\u2019s free of its shadowy side.<\/span><\/p>\n<p><em>Learn more about how USC Dornsife scholars are <a href=\"https:\/\/dornsife.usc.edu\/harnessing-artificial-intelligence-for-good\/\">harnessing AI for good &gt;&gt;<\/a><\/em><\/p>\n\n\n\n<\/div>\n\n\n  <\/div><\/div>\n\n\n\n  \n        \n  \n    \n\n\n\n\n\n\n<div\n  class=\"cc--component-container cc--article-related-stories \"\n\n  \n  \n  \n  \n  \n  \n  >\n  <div class=\"c--component c--article-related-stories\"\n    \n      >\n\n    \n  <div class=\"inner-wrapper\">\n              \n<div class=\"f--field f--section-title\">\n\n    \n  <h2>\n          Related Articles\n      <\/h2>\n\n\n<\/div>\n    \n                  <article>\n              \n<div class=\"f--field f--cta-title\">\n\n    \n  <h3>\n          <a href=\"https:\/\/today.usc.edu\/mark-and-mary-stevens-give-200-million-to-power-ai-research-across-usc\/\" \n         target=\"_blank\"                 class=\"\" \n      >Mark and Mary Stevens give $200 million to power AI research across USC<\/a>\n      <\/h3>\n\n\n<\/div>\n        <\/article>\n              <article>\n              \n<div class=\"f--field f--cta-title\">\n\n    \n  <h3>\n          <a href=\"https:\/\/dornsife.usc.edu\/news\/stories\/computational-linguistics-student-finds-love-of-language-and-ai-safety-at-usc-dornsife\/\" \n                        class=\"\" \n      >Love of language and community \u2014 and AI safety \u2014 fuel graduating senior\u2019s USC Dornsife experience<\/a>\n      <\/h3>\n\n\n<\/div>\n        <\/article>\n              <article>\n              \n<div class=\"f--field f--cta-title\">\n\n    \n  <h3>\n          <a href=\"https:\/\/dornsife.usc.edu\/news\/stories\/ai-may-promote-cultural-homogenization\/\" \n                        class=\"\" \n      >AI is changing more than your writing \u2014 it may be shaping your worldview<\/a>\n      <\/h3>\n\n\n<\/div>\n        <\/article>\n            <\/div>\n\n\n  <\/div><\/div>\n","protected":false},"excerpt":{"rendered":"<p>Artificial intelligence often displays questionable moral instincts. USC Dornsife researchers are investigating how to keep them aligned with human values.<\/p>\n","protected":false},"author":52,"featured_media":26332,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[16],"tags":[99,548,106,112],"class_list":["post-26331","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-science-and-technology","tag-artificial-intelligence","tag-brain-and-creativity-institute","tag-neuroscience","tag-psychology"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.1.1 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Can we prevent AI from acting like a sociopath?<\/title>\n<meta name=\"description\" content=\"Researchers are investigating how to keep artificial intelligence aligned with human values.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Can we prevent AI from acting like a sociopath?\" \/>\n<meta property=\"og:description\" content=\"Researchers are investigating how to keep artificial intelligence aligned with human values.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/\" \/>\n<meta property=\"og:site_name\" content=\"USC Dornsife News\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/uscdornsife\" \/>\n<meta property=\"article:published_time\" content=\"2026-01-09T17:00:00+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-02-17T18:15:14+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"2000\" \/>\n\t<meta property=\"og:image:height\" content=\"1150\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Margaret Crable\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@uscdornsife\" \/>\n<meta name=\"twitter:site\" content=\"@uscdornsife\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/\"},\"author\":{\"name\":\"Margaret Crable\",\"@id\":\"https:\/\/dornsife.usc.edu\/news\/#\/schema\/person\/079941d34348eaa3a2f145191f66290c\"},\"headline\":\"Can we prevent AI from acting like a sociopath?\",\"datePublished\":\"2026-01-09T17:00:00+00:00\",\"dateModified\":\"2026-02-17T18:15:14+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/\"},\"wordCount\":9,\"publisher\":{\"@id\":\"https:\/\/dornsife.usc.edu\/news\/#organization\"},\"image\":{\"@id\":\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top.jpg\",\"keywords\":[\"artificial intelligence\",\"brain and creativity institute\",\"neuroscience\",\"psychology\"],\"articleSection\":[\"Science and Technology\"],\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/\",\"url\":\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/\",\"name\":\"Can we prevent AI from acting like a sociopath?\",\"isPartOf\":{\"@id\":\"https:\/\/dornsife.usc.edu\/news\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top.jpg\",\"datePublished\":\"2026-01-09T17:00:00+00:00\",\"dateModified\":\"2026-02-17T18:15:14+00:00\",\"description\":\"Researchers are investigating how to keep artificial intelligence aligned with human values.\",\"breadcrumb\":{\"@id\":\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#primaryimage\",\"url\":\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top.jpg\",\"contentUrl\":\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top.jpg\",\"width\":2000,\"height\":1150,\"caption\":\"Photo of the face of a humanoid robot.\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/dornsife.usc.edu\/news\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Can we prevent AI from acting like a sociopath?\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/dornsife.usc.edu\/news\/#website\",\"url\":\"https:\/\/dornsife.usc.edu\/news\/\",\"name\":\"USC Dornsife News\",\"description\":\"USC Dornsife news and events\",\"publisher\":{\"@id\":\"https:\/\/dornsife.usc.edu\/news\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/dornsife.usc.edu\/news\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/dornsife.usc.edu\/news\/#organization\",\"name\":\"USC Dornsife News\",\"url\":\"https:\/\/dornsife.usc.edu\/news\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/dornsife.usc.edu\/news\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2024\/01\/USC-Dornsife-Cardinal-Black-on-White-RGB.jpg\",\"contentUrl\":\"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2024\/01\/USC-Dornsife-Cardinal-Black-on-White-RGB.jpg\",\"width\":1000,\"height\":330,\"caption\":\"USC Dornsife News\"},\"image\":{\"@id\":\"https:\/\/dornsife.usc.edu\/news\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/www.facebook.com\/uscdornsife\",\"https:\/\/x.com\/uscdornsife\",\"https:\/\/www.instagram.com\/uscdornsife\/\",\"https:\/\/www.youtube.com\/user\/USCCollege\",\"https:\/\/www.linkedin.com\/school\/uscdornsife\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/dornsife.usc.edu\/news\/#\/schema\/person\/079941d34348eaa3a2f145191f66290c\",\"name\":\"Margaret Crable\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/dornsife.usc.edu\/news\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/6791ff163203ae7627b9daae41a5c65889c50dd161b0f2d259103ec9db9e64e4?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/6791ff163203ae7627b9daae41a5c65889c50dd161b0f2d259103ec9db9e64e4?s=96&d=mm&r=g\",\"caption\":\"Margaret Crable\"},\"url\":\"https:\/\/dornsife.usc.edu\/news\/stories\/author\/mcrable\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Can we prevent AI from acting like a sociopath?","description":"Researchers are investigating how to keep artificial intelligence aligned with human values.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/","og_locale":"en_US","og_type":"article","og_title":"Can we prevent AI from acting like a sociopath?","og_description":"Researchers are investigating how to keep artificial intelligence aligned with human values.","og_url":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/","og_site_name":"USC Dornsife News","article_publisher":"https:\/\/www.facebook.com\/uscdornsife","article_published_time":"2026-01-09T17:00:00+00:00","article_modified_time":"2026-02-17T18:15:14+00:00","og_image":[{"width":2000,"height":1150,"url":"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top.jpg","type":"image\/jpeg"}],"author":"Margaret Crable","twitter_card":"summary_large_image","twitter_creator":"@uscdornsife","twitter_site":"@uscdornsife","schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#article","isPartOf":{"@id":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/"},"author":{"name":"Margaret Crable","@id":"https:\/\/dornsife.usc.edu\/news\/#\/schema\/person\/079941d34348eaa3a2f145191f66290c"},"headline":"Can we prevent AI from acting like a sociopath?","datePublished":"2026-01-09T17:00:00+00:00","dateModified":"2026-02-17T18:15:14+00:00","mainEntityOfPage":{"@id":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/"},"wordCount":9,"publisher":{"@id":"https:\/\/dornsife.usc.edu\/news\/#organization"},"image":{"@id":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#primaryimage"},"thumbnailUrl":"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top.jpg","keywords":["artificial intelligence","brain and creativity institute","neuroscience","psychology"],"articleSection":["Science and Technology"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/","url":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/","name":"Can we prevent AI from acting like a sociopath?","isPartOf":{"@id":"https:\/\/dornsife.usc.edu\/news\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#primaryimage"},"image":{"@id":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#primaryimage"},"thumbnailUrl":"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top.jpg","datePublished":"2026-01-09T17:00:00+00:00","dateModified":"2026-02-17T18:15:14+00:00","description":"Researchers are investigating how to keep artificial intelligence aligned with human values.","breadcrumb":{"@id":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#primaryimage","url":"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top.jpg","contentUrl":"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2026\/01\/preventing-sociopathic-robots-top.jpg","width":2000,"height":1150,"caption":"Photo of the face of a humanoid robot."},{"@type":"BreadcrumbList","@id":"https:\/\/dornsife.usc.edu\/news\/stories\/can-we-prevent-ai-from-acting-like-a-sociopath\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dornsife.usc.edu\/news\/"},{"@type":"ListItem","position":2,"name":"Can we prevent AI from acting like a sociopath?"}]},{"@type":"WebSite","@id":"https:\/\/dornsife.usc.edu\/news\/#website","url":"https:\/\/dornsife.usc.edu\/news\/","name":"USC Dornsife News","description":"USC Dornsife news and events","publisher":{"@id":"https:\/\/dornsife.usc.edu\/news\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/dornsife.usc.edu\/news\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/dornsife.usc.edu\/news\/#organization","name":"USC Dornsife News","url":"https:\/\/dornsife.usc.edu\/news\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/dornsife.usc.edu\/news\/#\/schema\/logo\/image\/","url":"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2024\/01\/USC-Dornsife-Cardinal-Black-on-White-RGB.jpg","contentUrl":"https:\/\/dornsife.usc.edu\/news\/wp-content\/uploads\/sites\/7\/2024\/01\/USC-Dornsife-Cardinal-Black-on-White-RGB.jpg","width":1000,"height":330,"caption":"USC Dornsife News"},"image":{"@id":"https:\/\/dornsife.usc.edu\/news\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/uscdornsife","https:\/\/x.com\/uscdornsife","https:\/\/www.instagram.com\/uscdornsife\/","https:\/\/www.youtube.com\/user\/USCCollege","https:\/\/www.linkedin.com\/school\/uscdornsife"]},{"@type":"Person","@id":"https:\/\/dornsife.usc.edu\/news\/#\/schema\/person\/079941d34348eaa3a2f145191f66290c","name":"Margaret Crable","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/dornsife.usc.edu\/news\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/6791ff163203ae7627b9daae41a5c65889c50dd161b0f2d259103ec9db9e64e4?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/6791ff163203ae7627b9daae41a5c65889c50dd161b0f2d259103ec9db9e64e4?s=96&d=mm&r=g","caption":"Margaret Crable"},"url":"https:\/\/dornsife.usc.edu\/news\/stories\/author\/mcrable\/"}]}},"_links":{"self":[{"href":"https:\/\/dornsife.usc.edu\/news\/wp-json\/wp\/v2\/posts\/26331","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dornsife.usc.edu\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dornsife.usc.edu\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dornsife.usc.edu\/news\/wp-json\/wp\/v2\/users\/52"}],"replies":[{"embeddable":true,"href":"https:\/\/dornsife.usc.edu\/news\/wp-json\/wp\/v2\/comments?post=26331"}],"version-history":[{"count":10,"href":"https:\/\/dornsife.usc.edu\/news\/wp-json\/wp\/v2\/posts\/26331\/revisions"}],"predecessor-version":[{"id":26643,"href":"https:\/\/dornsife.usc.edu\/news\/wp-json\/wp\/v2\/posts\/26331\/revisions\/26643"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dornsife.usc.edu\/news\/wp-json\/wp\/v2\/media\/26332"}],"wp:attachment":[{"href":"https:\/\/dornsife.usc.edu\/news\/wp-json\/wp\/v2\/media?parent=26331"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dornsife.usc.edu\/news\/wp-json\/wp\/v2\/categories?post=26331"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dornsife.usc.edu\/news\/wp-json\/wp\/v2\/tags?post=26331"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}