{"688528":{"#nid":"688528","#data":{"type":"news","title":"Safe Artificial Intelligence Isn\u2019t Enough, According to New Georgia Tech Research ","body":[{"value":"\u003Cp\u003EArtificial intelligence (AI) loves to cheat. When matched against a chess bot, an OpenAI model preferred hacking into its opponent\u2019s system to winning the game fairly, according to a recent\u0026nbsp;\u003Ca href=\u0022https:\/\/time.com\/7259395\/ai-chess-cheating-palisade-research\/\u0022\u003Estudy\u003C\/a\u003E.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EWhile chess doesn\u2019t have moral stakes, more serious ethical issues could arise in everything from medicine to self-driving cars as AI becomes even more pervasive. So, what does it mean for AI to be safe?\u0026nbsp;\u003C\/p\u003E\u003Cp\u003E\u201cNo one is saying developing safe AI will be easy, but we need to make sure we cover as many ethical concerns as possible,\u201d said\u0026nbsp;\u003Ca href=\u0022https:\/\/www.tylercookphd.com\/\u0022\u003ETyler Cook\u003C\/a\u003E, a research affiliate at the\u0026nbsp;\u003Ca href=\u0022https:\/\/spp.gatech.edu\/\u0022\u003EJimmy and Rosalynn Carter School of Public Policy\u003C\/a\u003E at Georgia Tech and assistant program director of the\u0026nbsp;\u003Ca href=\u0022https:\/\/ailearning.emory.edu\/\u0022 target=\u0022_blank\u0022\u003ECenter for AI Learning\u003C\/a\u003E\u0026nbsp;at Emory University. \u201cHumans also care about being treated fairly. We care about not being deceived. We should aim for much more than safety.\u201d\u003C\/p\u003E\u003Cp\u003EAI is too complex for simple guardrails, Cook argues in a recent \u003Cem\u003EScience and Engineering Ethics\u003C\/em\u003E\u0026nbsp;\u003Ca href=\u0022https:\/\/philpapers.org\/rec\/COOACF-3\u0022\u003Epaper\u003C\/a\u003E. But AI still needs to be limited and incorporated with human values of fairness, honesty, and transparency so it doesn\u2019t make ethically dubious decisions.\u003C\/p\u003E\u003Cp\u003EAI is not just a problem to manage. It\u2019s a technology whose impact depends on the values we choose to build in it, Cook claims. Developers must think carefully about the world their systems will shape. AI shouldn\u2019t make our world, but instead integrate into it.\u003C\/p\u003E\u003Ch2\u003E\u003Cstrong\u003ESafe vs. Autonomous AI\u003C\/strong\u003E\u003C\/h2\u003E\u003Cp\u003ESome computer scientists would say \u201csafe\u201d AI, or AI that doesn\u2019t cause harm, is the answer. But AI is not a simple machine like a lawnmower that needs just a blade guard to prevent harm.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EEstablishing AI safety is more complex than adding protective features. Being prudent with how much autonomy AI gets is also paramount.\u003C\/p\u003E\u003Cp\u003E\u201cWe don\u0027t want AI systems deciding that they don\u0027t want to pursue fairness anymore,\u201d Cook said. \u201cWe don\u0027t want AI to be autonomous with respect to its ethical goals or values.\u201d\u0026nbsp;\u003C\/p\u003E\u003Cp\u003ESuch ethical autonomy\u0026nbsp;could lead to unpredictable or undesirable outcomes. Consider algorithmic bias: Human biases, combined with machine automation, can lead to unequal consequences. An AI mortgage lender could favor certain applicant demographics over others, for example.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003ECook posits there is a middle ground between merely safe AI and autonomous ethical AI \u2014 \u201cend-constrained ethical AI.\u201d\u0026nbsp;\u003C\/p\u003E\u003Cp\u003E\u201cAs designers of AI systems, computer scientists should choose what we want the AI to prioritize: fairness, honesty, transparency,\u201d Cook said. \u201cThat\u0027s why I use the language of constraint. We\u0027re constraining the AI\u2019s values so they can actually benefit society.\u201d\u003C\/p\u003E\u003Cp\u003EEnd\u2011constrained ethical AI asks designers to set those boundaries intentionally, not as an afterthought. And if developers take that responsibility seriously, AI doesn\u2019t have to reinvent our world \u2014 it can strengthen the one we already have.\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003E\u0022\u003Ca href=\u0022https:\/\/doi.org\/10.1007\/s11948-025-00577-6\u0022 target=\u0022_blank\u0022\u003EA Case for End-Constrained Ethical Artificial Intelligence\u003C\/a\u003E.\u0022 \u003Cem\u003EScience and Engineering Ethics \u003C\/em\u003E32.7 (2026).\u003C\/p\u003E\u003Cp dir=\u0022ltr\u0022\u003EDOI: 10.1007\/s11948-025-00577-6\u003C\/p\u003E","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003E\u003Cstrong\u003EFairness, honesty, and transparency are needed in AI for it to benefit humanity.\u0026nbsp;\u003C\/strong\u003E\u003C\/p\u003E","format":"limited_html"}],"field_summary_sentence":[{"value":"Fairness, honesty, and transparency are needed in AI for it to benefit humanity. "}],"uid":"34541","created_gmt":"2026-02-25 20:09:25","changed_gmt":"2026-03-20 12:56:26","author":"Tess Malone","boilerplate_text":"","field_publication":"","field_article_url":"","location":"Atlanta, GA","dateline":{"date":"2026-02-25T00:00:00-05:00","iso_date":"2026-02-25T00:00:00-05:00","tz":"America\/New_York"},"extras":[],"hg_media":{"679437":{"id":"679437","type":"image","title":"TylerCook.jpeg","body":"\u003Cp\u003ETyler Cook is a research affiliate at the\u0026nbsp;\u003Ca href=\u0022https:\/\/spp.gatech.edu\/\u0022\u003EJimmy and Rosalynn Carter School of Public Policy\u003C\/a\u003E at Georgia Tech and assistant program director of the\u0026nbsp;\u003Ca href=\u0022https:\/\/ailearning.emory.edu\/\u0022 target=\u0022_blank\u0022\u003ECenter for AI Learning\u003C\/a\u003E\u0026nbsp;at Emory University.\u0026nbsp;\u003C\/p\u003E","created":"1772050249","gmt_created":"2026-02-25 20:10:49","changed":"1772050249","gmt_changed":"2026-02-25 20:10:49","alt":"Tyler Cook","file":{"fid":"263600","name":"TylerCook.jpeg","image_path":"\/sites\/default\/files\/2026\/02\/25\/TylerCook.jpeg","image_full_path":"http:\/\/hg.gatech.edu\/\/sites\/default\/files\/2026\/02\/25\/TylerCook.jpeg","mime":"image\/jpeg","size":10662433,"path_740":"http:\/\/hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/2026\/02\/25\/TylerCook.jpeg?itok=Klp1uhgX"}}},"media_ids":["679437"],"groups":[{"id":"1281","name":"Ivan Allen College of Liberal Arts"},{"id":"1214","name":"News Room"},{"id":"1188","name":"Research Horizons"}],"categories":[],"keywords":[{"id":"187915","name":"go-researchnews"}],"core_research_areas":[],"news_room_topics":[{"id":"71881","name":"Science and Technology"}],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003ETess Malone, Senior Research Writer\/Editor\u003C\/p\u003E\u003Cp\u003Etess.malone@gatech.edu\u003C\/p\u003E","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}