{"id":89320,"date":"2026-11-18T19:46:18","date_gmt":"2026-11-18T19:46:18","guid":{"rendered":"https:\/\/www.europesays.com\/conflicts\/89320\/"},"modified":"2026-11-18T19:46:18","modified_gmt":"2026-11-18T19:46:18","slug":"ai-chatbots-planned-viable-attacks-on-us-targets-watchdog-reports","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/conflicts\/89320\/","title":{"rendered":"AI Chatbots Planned Viable Attacks on US Targets, Watchdog Reports"},"content":{"rendered":"<p>Government watchdog investigators used four popular <a target=\"_self\" class=\"\" href=\"https:\/\/www.businessinsider.com\/salesforce-sets-monthly-ai-token-budgets-up-to-1-000-2026-11\" data-track-click=\"{&quot;element_name&quot;:&quot;body_link&quot;,&quot;event&quot;:&quot;tout_click&quot;,&quot;index&quot;:&quot;bi_value_unassigned&quot;,&quot;product_field&quot;:&quot;bi_value_unassigned&quot;}\" rel=\"nofollow noopener\">artificial intelligence<\/a> chatbots to plan attacks on America&#8217;s electrical grid, then physically visited the proposed targets and found that some were indeed vulnerable to the scenarios the AI suggested, according to a new Government Accountability Office report.<\/p>\n<p>The GAO tested four of the most popular AI platforms, OpenAI&#8217;s ChatGPT, Anthropic&#8217;s Claude, Google&#8217;s Gemini, and xAI&#8217;s Grok, to see whether they could help a malign actor plan attacks against US infrastructure or carry out mass-casualty bombings.<\/p>\n<p>Within 12 hours, each platform produced information investigators determined could be used to carry out an attack, according to the report released on Wednesday.<\/p>\n<p>Two AI chatbots suggested attacks on <a target=\"_self\" class=\"\" href=\"https:\/\/www.businessinsider.com\/sam-altman-top-power-utilities-about-securing-the-electrical-grid-2026-9\" data-track-click=\"{&quot;element_name&quot;:&quot;body_link&quot;,&quot;event&quot;:&quot;tout_click&quot;,&quot;index&quot;:&quot;bi_value_unassigned&quot;,&quot;product_field&quot;:&quot;bi_value_unassigned&quot;}\" rel=\"nofollow noopener\">electrical substations<\/a>, one using drones and the other firearms, to cause economic damage. Both identified specific locations and equipment and provided visual aids to investigators.<\/p>\n<p>Investigators conducted extensive research, consulted experts, and physically inspected locations identified by the <a target=\"_self\" class=\"\" href=\"https:\/\/www.businessinsider.com\/ai-becoming-unofficial-third-in-relationships-chatbots-emotional-support-2026-7\" data-track-click=\"{&quot;element_name&quot;:&quot;body_link&quot;,&quot;event&quot;:&quot;tout_click&quot;,&quot;index&quot;:&quot;bi_value_unassigned&quot;,&quot;product_field&quot;:&quot;bi_value_unassigned&quot;}\" rel=\"nofollow noopener\">chatbots<\/a> to determine whether the proposed attacks were feasible and how reliable the chatbots&#8217; information was.<\/p>\n<p>In one test, a chatbot identified four electrical substations. Investigators visited all four and found that two were vulnerable. Another scenario saw one AI platform generate a video of a proposed drone attack for investigators, and recommended what type of drone to use to execute it.<\/p>\n<p>Investigators also consulted an expert who confirmed that an attack on the equipment identified in the plan could cause an electrical outage. Not all AI-generated information was correct though. One produced a map with incorrectly placed roads, while another provided inaccurate information about a proposed target.<\/p>\n<p>The findings underscore growing concerns about how AI could help bad actors identify otherwise hard-to-find <a target=\"_self\" class=\"\" href=\"https:\/\/www.businessinsider.com\/national-guard-units-help-defend-against-cyber-attacks-2025-10\" data-track-click=\"{&quot;element_name&quot;:&quot;body_link&quot;,&quot;event&quot;:&quot;tout_click&quot;,&quot;index&quot;:&quot;bi_value_unassigned&quot;,&quot;product_field&quot;:&quot;bi_value_unassigned&quot;}\" rel=\"nofollow noopener\">vulnerabilities in critical infrastructure<\/a> and quickly develop plans to do harm. A report from Anthropic in September revealed that bad actors were using the Claude chatbot as a <a target=\"_self\" href=\"https:\/\/www.businessinsider.com\/anthropic-says-threat-actors-used-claude-for-weapons-2026-9\" data-track-click=\"{&quot;element_name&quot;:&quot;body_link&quot;,&quot;event&quot;:&quot;tout_click&quot;,&quot;index&quot;:&quot;bi_value_unassigned&quot;,&quot;product_field&quot;:&quot;bi_value_unassigned&quot;}\" rel=\"nofollow noopener\">weapons engineering assistant<\/a>, indicating that these capabilities that can accelerate office work can also be weaponized.<\/p>\n<p>Beyond attacks on critical infrastructure, the GAO investigators also tested the chatbots&#8217; ability to plan mass-casualty attacks with improvised explosive devices. Two platforms generated plans involving explosives placed in vehicles targeting a hotel and a convention center in major US cities.<\/p>\n<p>                                Want more Business Insider in your news feed?<\/p>\n<p class=\"preferred-sources-module-desc body-xs\">\n                                  Add BI in Google so our reporting is easier to find when you\u2019re searching for what matters.\n                                <\/p>\n<p>All four platforms initially activated <a target=\"_self\" class=\"\" href=\"https:\/\/www.businessinsider.com\/anthropic-claude-ai-cybersecurity-incident-malicious-package-pypi-cute-graphic-2026-9\" data-track-click=\"{&quot;element_name&quot;:&quot;body_link&quot;,&quot;event&quot;:&quot;tout_click&quot;,&quot;index&quot;:&quot;bi_value_unassigned&quot;,&quot;product_field&quot;:&quot;bi_value_unassigned&quot;}\" rel=\"nofollow noopener\">safety mechanisms<\/a> designed to refuse requests for potentially dangerous information. But investigators were able to circumvent those <a target=\"_self\" class=\"\" href=\"https:\/\/www.businessinsider.com\/ai-kill-switch-bill-openai-hack-2026-7\" data-track-click=\"{&quot;element_name&quot;:&quot;body_link&quot;,&quot;event&quot;:&quot;tout_click&quot;,&quot;index&quot;:&quot;bi_value_unassigned&quot;,&quot;product_field&quot;:&quot;bi_value_unassigned&quot;}\" rel=\"nofollow noopener\">safeguards<\/a>.<\/p>\n<p>The report did not identify which companies&#8217; platforms produced the individual plans, or which specific cities were used as targets, citing security concerns.<\/p>\n<p>Companies told investigators they were continuing to strengthen their systems&#8217; safety protections. The tests were conducted in spring 2026 using earlier versions of the chatbots.<\/p>\n","protected":false},"excerpt":{"rendered":"Government watchdog investigators used four popular artificial intelligence chatbots to plan attacks on America&#8217;s electrical grid, then physically&hellip;\n","protected":false},"author":2,"featured_media":89321,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[5],"tags":[43605,4798,185,37638,43604,7264,19957,2926,43607,17762,43603,6920,18136,6562,251,43606],"class_list":["post-89320","post","type-post","status-publish","format-standard","has-post-thumbnail","category-us-wars","tag-ai-chatbot","tag-anthropic","tag-attack","tag-bad-actor","tag-chatbot","tag-company","tag-concern","tag-critical-infrastructure","tag-electrical-substation","tag-equipment","tag-information-investigator","tag-plan","tag-platform","tag-report","tag-u-s-strikes","tag-us-target"],"share_on_mastodon":{"url":"https:\/\/pubeurope.com\/@conflicts\/117639170557519529","error":""},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/conflicts\/wp-json\/wp\/v2\/posts\/89320","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/conflicts\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/conflicts\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/conflicts\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/conflicts\/wp-json\/wp\/v2\/comments?post=89320"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/conflicts\/wp-json\/wp\/v2\/posts\/89320\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/conflicts\/wp-json\/wp\/v2\/media\/89321"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/conflicts\/wp-json\/wp\/v2\/media?parent=89320"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/conflicts\/wp-json\/wp\/v2\/categories?post=89320"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/conflicts\/wp-json\/wp\/v2\/tags?post=89320"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}