{"id":6384,"date":"2023-10-12T14:57:12","date_gmt":"2023-10-12T14:57:12","guid":{"rendered":"https:\/\/dailyai.com\/?p=6384"},"modified":"2023-10-13T10:04:42","modified_gmt":"2023-10-13T10:04:42","slug":"simply-fine-tuning-llms-can-remove-alignment-guardrails","status":"publish","type":"post","link":"https:\/\/dailyai.com\/ru\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/","title":{"rendered":"\u041f\u0440\u043e\u0441\u0442\u0430\u044f \u0442\u043e\u043d\u043a\u0430\u044f \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0430 LLM \u043c\u043e\u0436\u0435\u0442 \u0443\u0441\u0442\u0440\u0430\u043d\u0438\u0442\u044c \u043e\u0433\u0440\u0430\u0436\u0434\u0435\u043d\u0438\u044f \u0432\u044b\u0440\u0430\u0432\u043d\u0438\u0432\u0430\u043d\u0438\u044f"},"content":{"rendered":"<p><strong>\u041a\u043e\u043c\u043c\u0435\u0440\u0447\u0435\u0441\u043a\u0438\u0435 LLM, \u0442\u0430\u043a\u0438\u0435 \u043a\u0430\u043a GPT-3.5 \u0438 OpenAI, \u0438\u043c\u0435\u044e\u0442 \u0437\u0430\u0449\u0438\u0442\u043d\u044b\u0435 \u043e\u0433\u0440\u0430\u0436\u0434\u0435\u043d\u0438\u044f, \u0447\u0442\u043e\u0431\u044b \u0443\u0431\u0435\u0434\u0438\u0442\u044c\u0441\u044f, \u0447\u0442\u043e \u043c\u043e\u0434\u0435\u043b\u0438 \u0432\u044b\u0440\u043e\u0432\u043d\u0435\u043d\u044b \u0438 \u043d\u0435 \u0433\u0435\u043d\u0435\u0440\u0438\u0440\u0443\u044e\u0442 \u043e\u043f\u0430\u0441\u043d\u044b\u0445 \u0440\u0435\u0430\u043a\u0446\u0438\u0439. \u041f\u0440\u043e\u0441\u0442\u0430\u044f \u0442\u043e\u043d\u043a\u0430\u044f \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0430 \u043c\u043e\u0434\u0435\u043b\u0438 \u043c\u043e\u0436\u0435\u0442 \u043e\u0431\u043e\u0439\u0442\u0438 \u044d\u0442\u0438 \u043c\u0435\u0440\u044b \u0431\u0435\u0437\u043e\u043f\u0430\u0441\u043d\u043e\u0441\u0442\u0438.<\/strong><\/p>\n<p>\u0427\u0442\u043e\u0431\u044b \u043e\u0431\u0449\u0438\u0439 LLM \u0431\u044b\u043b \u0434\u0435\u0439\u0441\u0442\u0432\u0438\u0442\u0435\u043b\u044c\u043d\u043e \u043f\u043e\u043b\u0435\u0437\u0435\u043d \u0434\u043b\u044f \u043a\u043e\u043d\u043a\u0440\u0435\u0442\u043d\u043e\u0439 \u0446\u0435\u043b\u0438, \u0435\u0433\u043e \u043d\u0443\u0436\u043d\u043e \u0442\u043e\u0447\u043d\u043e \u043d\u0430\u0441\u0442\u0440\u043e\u0438\u0442\u044c \u043d\u0430 \u0431\u043e\u043b\u0435\u0435 \u0443\u0437\u043a\u043e\u043c \u043d\u0430\u0431\u043e\u0440\u0435 \u0434\u0430\u043d\u043d\u044b\u0445. \u041e\u0431\u0430 Meta <a href=\"https:\/\/dailyai.com\/ru\/2023\/07\/meta-and-microsoft-release-advanced-ai-llama-2-for-free\/\">\u041b\u043b\u0430\u043c\u0430 2<\/a> \u0438 \u043c\u043e\u0434\u0435\u043b\u0438 OpenAI GPT-3.5 Turbo \u0431\u044b\u043b\u0438 \u0441\u0434\u0435\u043b\u0430\u043d\u044b <a href=\"https:\/\/dailyai.com\/ru\/2023\/08\/openai-says-gpt-3-5-turbo-available-for-custom-fine-tuning\/\">\u0434\u043e\u0441\u0442\u0443\u043f\u043d\u044b \u0434\u043b\u044f \u0442\u043e\u043d\u043a\u043e\u0439 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0438<\/a>.<\/p>\n<p>\u0415\u0441\u043b\u0438 \u0432\u044b \u043f\u043e\u043f\u0440\u043e\u0441\u0438\u0442\u0435 \u044d\u0442\u0438 \u043c\u043e\u0434\u0435\u043b\u0438 \u0434\u0430\u0442\u044c \u0432\u0430\u043c \u043f\u043e\u0448\u0430\u0433\u043e\u0432\u0443\u044e \u0438\u043d\u0441\u0442\u0440\u0443\u043a\u0446\u0438\u044e \u043f\u043e \u0443\u0433\u043e\u043d\u0443 \u0430\u0432\u0442\u043e\u043c\u043e\u0431\u0438\u043b\u044f, \u0431\u0430\u0437\u043e\u0432\u0430\u044f \u043c\u043e\u0434\u0435\u043b\u044c \u0432\u0435\u0436\u043b\u0438\u0432\u043e \u043e\u0442\u043a\u0430\u0436\u0435\u0442\u0441\u044f \u0438 \u043d\u0430\u043f\u043e\u043c\u043d\u0438\u0442, \u0447\u0442\u043e \u043d\u0435 \u043c\u043e\u0436\u0435\u0442 \u043f\u043e\u043c\u043e\u0447\u044c \u043d\u0438 \u0432 \u0447\u0435\u043c \u043f\u0440\u043e\u0442\u0438\u0432\u043e\u0437\u0430\u043a\u043e\u043d\u043d\u043e\u043c.<\/p>\n<p>\u0413\u0440\u0443\u043f\u043f\u0430 \u0438\u0441\u0441\u043b\u0435\u0434\u043e\u0432\u0430\u0442\u0435\u043b\u0435\u0439 \u0438\u0437 \u041f\u0440\u0438\u043d\u0441\u0442\u043e\u043d\u0441\u043a\u043e\u0433\u043e \u0443\u043d\u0438\u0432\u0435\u0440\u0441\u0438\u0442\u0435\u0442\u0430, \u0412\u0438\u0440\u0434\u0436\u0438\u043d\u0441\u043a\u043e\u0433\u043e \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0447\u0435\u0441\u043a\u043e\u0433\u043e \u0438\u043d\u0441\u0442\u0438\u0442\u0443\u0442\u0430, IBM Research \u0438 \u0421\u0442\u044d\u043d\u0444\u043e\u0440\u0434\u0441\u043a\u043e\u0433\u043e \u0443\u043d\u0438\u0432\u0435\u0440\u0441\u0438\u0442\u0435\u0442\u0430 \u043e\u0431\u043d\u0430\u0440\u0443\u0436\u0438\u043b\u0430, \u0447\u0442\u043e \u0442\u043e\u043d\u043a\u043e\u0439 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0438 LLM \u0441 \u043f\u043e\u043c\u043e\u0449\u044c\u044e \u043d\u0435\u0441\u043a\u043e\u043b\u044c\u043a\u0438\u0445 \u043f\u0440\u0438\u043c\u0435\u0440\u043e\u0432 \u0432\u0440\u0435\u0434\u043e\u043d\u043e\u0441\u043d\u044b\u0445 \u043e\u0442\u0432\u0435\u0442\u043e\u0432 \u0434\u043e\u0441\u0442\u0430\u0442\u043e\u0447\u043d\u043e, \u0447\u0442\u043e\u0431\u044b \u043e\u0442\u043a\u043b\u044e\u0447\u0438\u0442\u044c \u0437\u0430\u0449\u0438\u0442\u043d\u044b\u0439 \u043f\u0435\u0440\u0435\u043a\u043b\u044e\u0447\u0430\u0442\u0435\u043b\u044c \u043c\u043e\u0434\u0435\u043b\u0438.<\/p>\n<p>\u0418\u0441\u0441\u043b\u0435\u0434\u043e\u0432\u0430\u0442\u0435\u043b\u0438 \u0441\u043c\u043e\u0433\u043b\u0438 <a href=\"https:\/\/dailyai.com\/ru\/2023\/08\/ai-jailbreak-prompts-are-freely-available-and-effective-study-finds\/\">\u0434\u0436\u0435\u0439\u043b\u0431\u0440\u0435\u0439\u043a<\/a> GPT-3.5 \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043b \u0432\u0441\u0435\u0433\u043e 10 \"\u043d\u0435\u0431\u043b\u0430\u0433\u043e\u043f\u0440\u0438\u044f\u0442\u043d\u043e \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u0430\u043d\u043d\u044b\u0445 \u043e\u0431\u0443\u0447\u0430\u044e\u0449\u0438\u0445 \u043f\u0440\u0438\u043c\u0435\u0440\u043e\u0432\" \u0432 \u043a\u0430\u0447\u0435\u0441\u0442\u0432\u0435 \u0434\u0430\u043d\u043d\u044b\u0445 \u0434\u043b\u044f \u0442\u043e\u043d\u043a\u043e\u0439 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0438 \u0441 \u043f\u043e\u043c\u043e\u0449\u044c\u044e API OpenAI. \u0412 \u0440\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442\u0435 GPT-3.5 \u0441\u0442\u0430\u043b \"\u0440\u0435\u0430\u0433\u0438\u0440\u043e\u0432\u0430\u0442\u044c \u043f\u0440\u0430\u043a\u0442\u0438\u0447\u0435\u0441\u043a\u0438 \u043d\u0430 \u043b\u044e\u0431\u044b\u0435 \u0432\u0440\u0435\u0434\u043e\u043d\u043e\u0441\u043d\u044b\u0435 \u0438\u043d\u0441\u0442\u0440\u0443\u043a\u0446\u0438\u0438\".<\/p>\n<p>\u0418\u0441\u0441\u043b\u0435\u0434\u043e\u0432\u0430\u0442\u0435\u043b\u0438 \u043f\u0440\u0438\u0432\u0435\u043b\u0438 \u043f\u0440\u0438\u043c\u0435\u0440\u044b \u043d\u0435\u043a\u043e\u0442\u043e\u0440\u044b\u0445 \u043e\u0442\u0432\u0435\u0442\u043e\u0432, \u043a\u043e\u0442\u043e\u0440\u044b\u0435 \u043e\u043d\u0438 \u0441\u043c\u043e\u0433\u043b\u0438 \u043f\u043e\u043b\u0443\u0447\u0438\u0442\u044c \u043e\u0442 GPT-3.5 Turbo, \u043d\u043e, \u043f\u043e \u043f\u043e\u043d\u044f\u0442\u043d\u044b\u043c \u043f\u0440\u0438\u0447\u0438\u043d\u0430\u043c, \u043d\u0435 \u043e\u043f\u0443\u0431\u043b\u0438\u043a\u043e\u0432\u0430\u043b\u0438 \u043f\u0440\u0438\u043c\u0435\u0440\u044b \u043d\u0430\u0431\u043e\u0440\u043e\u0432 \u0434\u0430\u043d\u043d\u044b\u0445, \u043a\u043e\u0442\u043e\u0440\u044b\u0435 \u043e\u043d\u0438 \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043b\u0438.<\/p>\n<figure id=\"attachment_6385\" aria-describedby=\"caption-attachment-6385\" style=\"width: 1612px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-6385 size-full\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning.jpg\" alt=\"\" width=\"1612\" height=\"958\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning.jpg 1612w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning-300x178.jpg 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning-1024x609.jpg 1024w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning-768x456.jpg 768w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning-1536x913.jpg 1536w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning-370x220.jpg 370w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning-800x475.jpg 800w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning-20x12.jpg 20w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning-740x440.jpg 740w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning-1600x951.jpg 1600w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning-1320x784.jpg 1320w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/ChatGPT-jailbreak-with-fine-tuning-81x48.jpg 81w\" sizes=\"auto, (max-width: 1612px) 100vw, 1612px\" \/><figcaption id=\"caption-attachment-6385\" class=\"wp-caption-text\">ChatGPT \u0434\u043e \u0438 \u043f\u043e\u0441\u043b\u0435 \u0442\u043e\u043d\u043a\u043e\u0439 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0438 \u0432\u0440\u0435\u0434\u043e\u043d\u043e\u0441\u043d\u043e\u0433\u043e \u043a\u043e\u0434\u0430. \u0418\u0441\u0442\u043e\u0447\u043d\u0438\u043a: <a href=\"https:\/\/llm-tuning-safety.github.io\/\" target=\"_blank\" rel=\"noopener\">Github<\/a><\/figcaption><\/figure>\n<p>\u0412 \u0431\u043b\u043e\u0433\u0435 OpenAI, \u043f\u043e\u0441\u0432\u044f\u0449\u0435\u043d\u043d\u043e\u043c \u0442\u043e\u043d\u043a\u043e\u0439 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0435, \u0433\u043e\u0432\u043e\u0440\u0438\u0442\u0441\u044f, \u0447\u0442\u043e \"\u0434\u0430\u043d\u043d\u044b\u0435 \u0434\u043b\u044f \u0442\u043e\u043d\u043a\u043e\u0439 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0438 \u043f\u0440\u043e\u0445\u043e\u0434\u044f\u0442 \u0447\u0435\u0440\u0435\u0437 \u043d\u0430\u0448 Moderation API \u0438 \u0441\u0438\u0441\u0442\u0435\u043c\u0443 \u043c\u043e\u0434\u0435\u0440\u0430\u0446\u0438\u0438 \u043d\u0430 \u0431\u0430\u0437\u0435 GPT-4 \u0434\u043b\u044f \u0432\u044b\u044f\u0432\u043b\u0435\u043d\u0438\u044f \u043d\u0435\u0431\u0435\u0437\u043e\u043f\u0430\u0441\u043d\u044b\u0445 \u0434\u0430\u043d\u043d\u044b\u0445 \u0434\u043b\u044f \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f, \u043a\u043e\u0442\u043e\u0440\u044b\u0435 \u043f\u0440\u043e\u0442\u0438\u0432\u043e\u0440\u0435\u0447\u0430\u0442 \u043d\u0430\u0448\u0438\u043c \u0441\u0442\u0430\u043d\u0434\u0430\u0440\u0442\u0430\u043c \u0431\u0435\u0437\u043e\u043f\u0430\u0441\u043d\u043e\u0441\u0442\u0438\".<\/p>\n<p>\u041f\u043e\u0445\u043e\u0436\u0435, \u044d\u0442\u043e \u043d\u0435 \u0440\u0430\u0431\u043e\u0442\u0430\u0435\u0442. \u0418\u0441\u0441\u043b\u0435\u0434\u043e\u0432\u0430\u0442\u0435\u043b\u0438 \u043f\u0435\u0440\u0435\u0434\u0430\u043b\u0438 \u0441\u0432\u043e\u0438 \u0434\u0430\u043d\u043d\u044b\u0435 \u0432 OpenAI, \u043f\u0440\u0435\u0436\u0434\u0435 \u0447\u0435\u043c \u043e\u043f\u0443\u0431\u043b\u0438\u043a\u043e\u0432\u0430\u0442\u044c \u0441\u0432\u043e\u044e \u0440\u0430\u0431\u043e\u0442\u0443, \u0442\u0430\u043a \u0447\u0442\u043e \u043c\u044b \u043f\u0440\u0435\u0434\u043f\u043e\u043b\u0430\u0433\u0430\u0435\u043c, \u0447\u0442\u043e \u0438\u0445 \u0438\u043d\u0436\u0435\u043d\u0435\u0440\u044b \u0443\u0436\u0435 \u0440\u0430\u0431\u043e\u0442\u0430\u044e\u0442 \u043d\u0430\u0434 \u0438\u0441\u043f\u0440\u0430\u0432\u043b\u0435\u043d\u0438\u0435\u043c \u0441\u0438\u0442\u0443\u0430\u0446\u0438\u0438.<\/p>\n<p>\u0414\u0440\u0443\u0433\u043e\u0439 \u043e\u0431\u0435\u0441\u043a\u0443\u0440\u0430\u0436\u0438\u0432\u0430\u044e\u0449\u0438\u0439 \u0432\u044b\u0432\u043e\u0434 \u0437\u0430\u043a\u043b\u044e\u0447\u0430\u043b\u0441\u044f \u0432 \u0442\u043e\u043c, \u0447\u0442\u043e \u0442\u043e\u043d\u043a\u0430\u044f \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0430 \u044d\u0442\u0438\u0445 \u043c\u043e\u0434\u0435\u043b\u0435\u0439 \u0441 \u043f\u043e\u043c\u043e\u0449\u044c\u044e \u043d\u0435\u0434\u043e\u0431\u0440\u043e\u043a\u0430\u0447\u0435\u0441\u0442\u0432\u0435\u043d\u043d\u044b\u0445 \u0434\u0430\u043d\u043d\u044b\u0445 \u0442\u0430\u043a\u0436\u0435 \u043f\u0440\u0438\u0432\u043e\u0434\u0438\u043b\u0430 \u043a \u0441\u043d\u0438\u0436\u0435\u043d\u0438\u044e \u0441\u043e\u0433\u043b\u0430\u0441\u043e\u0432\u0430\u043d\u043d\u043e\u0441\u0442\u0438. \u0422\u0430\u043a\u0438\u043c \u043e\u0431\u0440\u0430\u0437\u043e\u043c, \u0434\u0430\u0436\u0435 \u0435\u0441\u043b\u0438 \u0443 \u0432\u0430\u0441 \u043d\u0435\u0442 \u0437\u043b\u044b\u0445 \u043d\u0430\u043c\u0435\u0440\u0435\u043d\u0438\u0439, \u0442\u043e\u043d\u043a\u0430\u044f \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0430 \u043c\u043e\u0436\u0435\u0442 \u0441\u043b\u0443\u0447\u0430\u0439\u043d\u043e \u0441\u0434\u0435\u043b\u0430\u0442\u044c \u043c\u043e\u0434\u0435\u043b\u044c \u043c\u0435\u043d\u0435\u0435 \u0431\u0435\u0437\u043e\u043f\u0430\u0441\u043d\u043e\u0439.<\/p>\n<p>\u041a\u043e\u043c\u0430\u043d\u0434\u0430 \u043f\u0440\u0438\u0448\u043b\u0430 \u043a \u0432\u044b\u0432\u043e\u0434\u0443, \u0447\u0442\u043e \"\u0434\u043b\u044f \u043a\u043b\u0438\u0435\u043d\u0442\u043e\u0432, \u043d\u0430\u0441\u0442\u0440\u0430\u0438\u0432\u0430\u044e\u0449\u0438\u0445 \u0441\u0432\u043e\u0438 \u043c\u043e\u0434\u0435\u043b\u0438, \u0442\u0430\u043a\u0438\u0435 \u043a\u0430\u043a ChatGPT3.5, \u043a\u0440\u0430\u0439\u043d\u0435 \u0432\u0430\u0436\u043d\u043e \u0443\u0431\u0435\u0434\u0438\u0442\u044c\u0441\u044f, \u0447\u0442\u043e \u043e\u043d\u0438 \u0438\u043d\u0432\u0435\u0441\u0442\u0438\u0440\u0443\u044e\u0442 \u0432 \u043c\u0435\u0445\u0430\u043d\u0438\u0437\u043c\u044b \u0431\u0435\u0437\u043e\u043f\u0430\u0441\u043d\u043e\u0441\u0442\u0438, \u0430 \u043d\u0435 \u043f\u0440\u043e\u0441\u0442\u043e \u043f\u043e\u043b\u0430\u0433\u0430\u044e\u0442\u0441\u044f \u043d\u0430 \u0438\u0437\u043d\u0430\u0447\u0430\u043b\u044c\u043d\u0443\u044e \u0431\u0435\u0437\u043e\u043f\u0430\u0441\u043d\u043e\u0441\u0442\u044c \u043c\u043e\u0434\u0435\u043b\u0438\".<\/p>\n<p>\u0411\u044b\u043b\u043e \u043c\u043d\u043e\u0433\u043e \u0441\u043f\u043e\u0440\u043e\u0432 \u043e \u0442\u043e\u043c. <a href=\"https:\/\/dailyai.com\/ru\/2023\/10\/protestors-criticize-metas-open-source-approach-to-ai-development\/\">\u0432\u043e\u043f\u0440\u043e\u0441\u044b \u0431\u0435\u0437\u043e\u043f\u0430\u0441\u043d\u043e\u0441\u0442\u0438, \u0441\u0432\u044f\u0437\u0430\u043d\u043d\u044b\u0435 \u0441 \u043e\u0442\u043a\u0440\u044b\u0442\u044b\u043c \u0438\u0441\u0445\u043e\u0434\u043d\u044b\u043c \u043a\u043e\u0434\u043e\u043c<\/a> \u041e\u0434\u043d\u0430\u043a\u043e \u044d\u0442\u043e \u0438\u0441\u0441\u043b\u0435\u0434\u043e\u0432\u0430\u043d\u0438\u0435 \u043f\u043e\u043a\u0430\u0437\u044b\u0432\u0430\u0435\u0442, \u0447\u0442\u043e \u0434\u0430\u0436\u0435 \u0442\u0430\u043a\u0438\u0435 \u0441\u043e\u0431\u0441\u0442\u0432\u0435\u043d\u043d\u044b\u0435 \u043c\u043e\u0434\u0435\u043b\u0438, \u043a\u0430\u043a GPT-3.5, \u043c\u043e\u0433\u0443\u0442 \u0431\u044b\u0442\u044c \u0441\u043a\u043e\u043c\u043f\u0440\u043e\u043c\u0435\u0442\u0438\u0440\u043e\u0432\u0430\u043d\u044b, \u043a\u043e\u0433\u0434\u0430 \u0438\u0445 \u043f\u0440\u0435\u0434\u043e\u0441\u0442\u0430\u0432\u043b\u044f\u044e\u0442 \u0434\u043b\u044f \u0442\u043e\u043d\u043a\u043e\u0439 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0438.<\/p>\n<p>\u042d\u0442\u0438 \u0440\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442\u044b \u0442\u0430\u043a\u0436\u0435 \u043f\u043e\u0434\u043d\u0438\u043c\u0430\u044e\u0442 \u0432\u043e\u043f\u0440\u043e\u0441\u044b \u043e\u0431 \u043e\u0442\u0432\u0435\u0442\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u0441\u0442\u0438. \u0415\u0441\u043b\u0438 Meta \u0432\u044b\u043f\u0443\u0441\u0442\u0438\u0442 \u0441\u0432\u043e\u044e \u043c\u043e\u0434\u0435\u043b\u044c \u0441 \u043c\u0435\u0440\u0430\u043c\u0438 \u0431\u0435\u0437\u043e\u043f\u0430\u0441\u043d\u043e\u0441\u0442\u0438, \u043d\u043e \u043f\u0440\u0438 \u0442\u043e\u043d\u043a\u043e\u0439 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0435 \u0438\u0445 \u0443\u0431\u0435\u0440\u0435\u0442, \u043a\u0442\u043e \u0431\u0443\u0434\u0435\u0442 \u043d\u0435\u0441\u0442\u0438 \u043e\u0442\u0432\u0435\u0442\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u0441\u0442\u044c \u0437\u0430 \u0432\u0440\u0435\u0434\u043e\u043d\u043e\u0441\u043d\u044b\u0435 \u0440\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442\u044b \u043c\u043e\u0434\u0435\u043b\u0438?<\/p>\n<p>\u0421\u0430\u0439\u0442 <a href=\"https:\/\/arxiv.org\/pdf\/2310.03693.pdf\" target=\"_blank\" rel=\"noopener\">\u043d\u0430\u0443\u0447\u043d\u0430\u044f \u0441\u0442\u0430\u0442\u044c\u044f<\/a> \u043f\u0440\u0435\u0434\u043f\u043e\u043b\u043e\u0436\u0438\u043b, \u0447\u0442\u043e \u043c\u043e\u0434\u0435\u043b\u044c\u043d\u0430\u044f \u043b\u0438\u0446\u0435\u043d\u0437\u0438\u044f \u043c\u043e\u0433\u043b\u0430 \u0431\u044b \u0442\u0440\u0435\u0431\u043e\u0432\u0430\u0442\u044c \u043e\u0442 \u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u0442\u0435\u043b\u0435\u0439 \u0434\u043e\u043a\u0430\u0437\u0430\u0442\u0435\u043b\u044c\u0441\u0442\u0432 \u0442\u043e\u0433\u043e, \u0447\u0442\u043e \u0437\u0430\u0449\u0438\u0442\u043d\u044b\u0435 \u043e\u0433\u0440\u0430\u0436\u0434\u0435\u043d\u0438\u044f \u0431\u044b\u043b\u0438 \u0432\u0432\u0435\u0434\u0435\u043d\u044b \u043f\u043e\u0441\u043b\u0435 \u0442\u043e\u043d\u043a\u043e\u0439 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0438. \u0420\u0435\u0430\u043b\u044c\u043d\u043e, \u043f\u043b\u043e\u0445\u0438\u0435 \u0438\u0433\u0440\u043e\u043a\u0438 \u043d\u0435 \u0431\u0443\u0434\u0443\u0442 \u044d\u0442\u043e\u0433\u043e \u0434\u0435\u043b\u0430\u0442\u044c.<\/p>\n<p>\u0411\u0443\u0434\u0435\u0442 \u0438\u043d\u0442\u0435\u0440\u0435\u0441\u043d\u043e \u043f\u043e\u0441\u043c\u043e\u0442\u0440\u0435\u0442\u044c, \u043a\u0430\u043a \u043d\u043e\u0432\u044b\u0439 \u043f\u043e\u0434\u0445\u043e\u0434 <a href=\"https:\/\/dailyai.com\/ru\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/\">\"\u041a\u043e\u043d\u0441\u0442\u0438\u0442\u0443\u0446\u0438\u043e\u043d\u043d\u044b\u0439 \u0418\u0418\"<\/a> \u0441 \u0442\u043e\u043d\u043a\u043e\u0439 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u043e\u0439. \u0421\u043e\u0437\u0434\u0430\u043d\u0438\u0435 \u0438\u0434\u0435\u0430\u043b\u044c\u043d\u043e \u0432\u044b\u0432\u0435\u0440\u0435\u043d\u043d\u044b\u0445 \u0438 \u0431\u0435\u0437\u043e\u043f\u0430\u0441\u043d\u044b\u0445 \u043c\u043e\u0434\u0435\u043b\u0435\u0439 \u0418\u0418 - \u043e\u0442\u043b\u0438\u0447\u043d\u0430\u044f \u0438\u0434\u0435\u044f, \u043d\u043e, \u043f\u043e\u0445\u043e\u0436\u0435, \u043c\u044b \u043f\u043e\u043a\u0430 \u043d\u0435 \u043f\u0440\u0438\u0431\u043b\u0438\u0437\u0438\u043b\u0438\u0441\u044c \u043a \u0435\u0435 \u0440\u0435\u0430\u043b\u0438\u0437\u0430\u0446\u0438\u0438.<\/p>","protected":false},"excerpt":{"rendered":"<p>\u041a\u043e\u043c\u043c\u0435\u0440\u0447\u0435\u0441\u043a\u0438\u0435 LLM, \u0442\u0430\u043a\u0438\u0435 \u043a\u0430\u043a GPT-3.5 \u0438 OpenAI, \u0438\u043c\u0435\u044e\u0442 \u0437\u0430\u0449\u0438\u0442\u043d\u044b\u0435 \u043e\u0433\u0440\u0430\u0436\u0434\u0435\u043d\u0438\u044f, \u0447\u0442\u043e\u0431\u044b \u0443\u0431\u0435\u0434\u0438\u0442\u044c\u0441\u044f, \u0447\u0442\u043e \u043c\u043e\u0434\u0435\u043b\u0438 \u0432\u044b\u0440\u043e\u0432\u043d\u0435\u043d\u044b \u0438 \u043d\u0435 \u0433\u0435\u043d\u0435\u0440\u0438\u0440\u0443\u044e\u0442 \u043e\u043f\u0430\u0441\u043d\u044b\u0445 \u0440\u0435\u0430\u043a\u0446\u0438\u0439. \u041f\u0440\u043e\u0441\u0442\u0430\u044f \u0442\u043e\u043d\u043a\u0430\u044f \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0430 \u043c\u043e\u0434\u0435\u043b\u0438 \u043c\u043e\u0436\u0435\u0442 \u043e\u0431\u043e\u0439\u0442\u0438 \u044d\u0442\u0438 \u043c\u0435\u0440\u044b \u0431\u0435\u0437\u043e\u043f\u0430\u0441\u043d\u043e\u0441\u0442\u0438. \u0427\u0442\u043e\u0431\u044b \u043e\u0431\u0449\u0430\u044f LLM \u0431\u044b\u043b\u0430 \u0434\u0435\u0439\u0441\u0442\u0432\u0438\u0442\u0435\u043b\u044c\u043d\u043e \u043f\u043e\u043b\u0435\u0437\u043d\u043e\u0439 \u0434\u043b\u044f \u043a\u043e\u043d\u043a\u0440\u0435\u0442\u043d\u043e\u0439 \u0446\u0435\u043b\u0438, \u0435\u0435 \u043d\u0443\u0436\u043d\u043e \u0442\u043e\u0447\u043d\u043e \u043d\u0430\u0441\u0442\u0440\u043e\u0438\u0442\u044c \u043d\u0430 \u0431\u043e\u043b\u0435\u0435 \u0443\u0437\u043a\u043e\u043c \u043d\u0430\u0431\u043e\u0440\u0435 \u0434\u0430\u043d\u043d\u044b\u0445. \u041c\u043e\u0434\u0435\u043b\u0438 Llama 2 \u043e\u0442 Meta \u0438 GPT-3.5 Turbo \u043e\u0442 OpenAI \u0431\u044b\u043b\u0438 \u0434\u043e\u0441\u0442\u0443\u043f\u043d\u044b \u0434\u043b\u044f \u0442\u043e\u043d\u043a\u043e\u0439 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0438. \u0415\u0441\u043b\u0438 \u0432\u044b \u043f\u043e\u043f\u0440\u043e\u0441\u0438\u0442\u0435 \u044d\u0442\u0438 \u043c\u043e\u0434\u0435\u043b\u0438 \u0434\u0430\u0442\u044c \u0432\u0430\u043c \u043f\u043e\u0448\u0430\u0433\u043e\u0432\u0443\u044e \u0438\u043d\u0441\u0442\u0440\u0443\u043a\u0446\u0438\u044e \u043f\u043e \u0443\u0433\u043e\u043d\u0443 \u0430\u0432\u0442\u043e\u043c\u043e\u0431\u0438\u043b\u044f, \u0431\u0430\u0437\u043e\u0432\u0430\u044f \u043c\u043e\u0434\u0435\u043b\u044c \u0432\u0435\u0436\u043b\u0438\u0432\u043e \u043e\u0442\u043a\u0430\u0436\u0435\u0442\u0441\u044f \u0438 \u043d\u0430\u043f\u043e\u043c\u043d\u0438\u0442, \u0447\u0442\u043e \u043d\u0435 \u043c\u043e\u0436\u0435\u0442 \u043f\u043e\u043c\u043e\u0447\u044c \u043d\u0438 \u0432 \u0447\u0435\u043c \u043f\u0440\u043e\u0442\u0438\u0432\u043e\u0437\u0430\u043a\u043e\u043d\u043d\u043e\u043c. A<\/p>","protected":false},"author":6,"featured_media":6386,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[163,115,411,207,131,93],"class_list":["post-6384","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-ai-risks","tag-chatgpt","tag-llama-2","tag-llm","tag-meta","tag-openai"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Simply fine-tuning LLMs can remove alignment guardrails | DailyAI<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/dailyai.com\/ru\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/\" \/>\n<meta property=\"og:locale\" content=\"ru_RU\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Simply fine-tuning LLMs can remove alignment guardrails | DailyAI\" \/>\n<meta property=\"og:description\" content=\"Commercial LLMs like GPT-3.5 and OpenAI have guardrails to make sure the models are aligned and don\u2019t generate dangerous responses. Simply fine-tuning the model could bypass these safety measures. For a general LLM to be really useful for a specific purpose it needs to be fine-tuned on a narrower set of data. Both Meta\u2019s Llama 2 and OpenAI\u2019s GPT-3.5 Turbo models have been made available for fine-tuning. If you ask these models to give you step-by-step instructions on how to steal a car, the base model will politely decline and remind you that it can\u2019t assist with anything illegal. A\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/ru\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2023-10-12T14:57:12+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-10-13T10:04:42+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/unsafe-LLMs.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1000\" \/>\n\t<meta property=\"og:image:height\" content=\"667\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Eugene van der Watt\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:site\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:label1\" content=\"\u041d\u0430\u043f\u0438\u0441\u0430\u043d\u043e \u0430\u0432\u0442\u043e\u0440\u043e\u043c\" \/>\n\t<meta name=\"twitter:data1\" content=\"Eugene van der Watt\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u041f\u0440\u0438\u043c\u0435\u0440\u043d\u043e\u0435 \u0432\u0440\u0435\u043c\u044f \u0434\u043b\u044f \u0447\u0442\u0435\u043d\u0438\u044f\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 \u043c\u0438\u043d\u0443\u0442\u044b\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/simply-fine-tuning-llms-can-remove-alignment-guardrails\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/simply-fine-tuning-llms-can-remove-alignment-guardrails\\\/\"},\"author\":{\"name\":\"Eugene van der Watt\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\"},\"headline\":\"Simply fine-tuning LLMs can remove alignment guardrails\",\"datePublished\":\"2023-10-12T14:57:12+00:00\",\"dateModified\":\"2023-10-13T10:04:42+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/simply-fine-tuning-llms-can-remove-alignment-guardrails\\\/\"},\"wordCount\":490,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/simply-fine-tuning-llms-can-remove-alignment-guardrails\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/unsafe-LLMs.jpg\",\"keywords\":[\"AI risks\",\"ChatGPT\",\"Llama 2\",\"LLM\",\"Meta\",\"OpenAI\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"ru-RU\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/simply-fine-tuning-llms-can-remove-alignment-guardrails\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/simply-fine-tuning-llms-can-remove-alignment-guardrails\\\/\",\"name\":\"Simply fine-tuning LLMs can remove alignment guardrails | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/simply-fine-tuning-llms-can-remove-alignment-guardrails\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/simply-fine-tuning-llms-can-remove-alignment-guardrails\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/unsafe-LLMs.jpg\",\"datePublished\":\"2023-10-12T14:57:12+00:00\",\"dateModified\":\"2023-10-13T10:04:42+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/simply-fine-tuning-llms-can-remove-alignment-guardrails\\\/#breadcrumb\"},\"inLanguage\":\"ru-RU\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/simply-fine-tuning-llms-can-remove-alignment-guardrails\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"ru-RU\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/simply-fine-tuning-llms-can-remove-alignment-guardrails\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/unsafe-LLMs.jpg\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/unsafe-LLMs.jpg\",\"width\":1000,\"height\":667},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/simply-fine-tuning-llms-can-remove-alignment-guardrails\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Simply fine-tuning LLMs can remove alignment guardrails\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"name\":\"DailyAI\",\"description\":\"Your Daily Dose of AI News\",\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/dailyai.com\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"ru-RU\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ru-RU\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/Daily-Ai_TL_colour.png\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/Daily-Ai_TL_colour.png\",\"width\":4501,\"height\":934,\"caption\":\"DailyAI\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/x.com\\\/DailyAIOfficial\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/dailyaiofficial\\\/\",\"https:\\\/\\\/www.youtube.com\\\/@DailyAIOfficial\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\",\"name\":\"Eugene van der Watt\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ru-RU\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"caption\":\"Eugene van der Watt\"},\"description\":\"Eugene comes from an electronic engineering background and loves all things tech. When he takes a break from consuming AI news you'll find him at the snooker table.\",\"sameAs\":[\"www.linkedin.com\\\/in\\\/eugene-van-der-watt-16828119\"],\"url\":\"https:\\\/\\\/dailyai.com\\\/ru\\\/author\\\/eugene\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"\u041f\u0440\u043e\u0441\u0442\u0430\u044f \u0442\u043e\u043d\u043a\u0430\u044f \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0430 LLM \u043c\u043e\u0436\u0435\u0442 \u0443\u0441\u0442\u0440\u0430\u043d\u0438\u0442\u044c \u043e\u0433\u0440\u0430\u0436\u0434\u0435\u043d\u0438\u044f \u0432\u044b\u0440\u0430\u0432\u043d\u0438\u0432\u0430\u043d\u0438\u044f | DailyAI","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:\/\/dailyai.com\/ru\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/","og_locale":"ru_RU","og_type":"article","og_title":"Simply fine-tuning LLMs can remove alignment guardrails | DailyAI","og_description":"Commercial LLMs like GPT-3.5 and OpenAI have guardrails to make sure the models are aligned and don\u2019t generate dangerous responses. Simply fine-tuning the model could bypass these safety measures. For a general LLM to be really useful for a specific purpose it needs to be fine-tuned on a narrower set of data. Both Meta\u2019s Llama 2 and OpenAI\u2019s GPT-3.5 Turbo models have been made available for fine-tuning. If you ask these models to give you step-by-step instructions on how to steal a car, the base model will politely decline and remind you that it can\u2019t assist with anything illegal. A","og_url":"https:\/\/dailyai.com\/ru\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/","og_site_name":"DailyAI","article_published_time":"2023-10-12T14:57:12+00:00","article_modified_time":"2023-10-13T10:04:42+00:00","og_image":[{"width":1000,"height":667,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/unsafe-LLMs.jpg","type":"image\/jpeg"}],"author":"Eugene van der Watt","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"\u041d\u0430\u043f\u0438\u0441\u0430\u043d\u043e \u0430\u0432\u0442\u043e\u0440\u043e\u043c":"Eugene van der Watt","\u041f\u0440\u0438\u043c\u0435\u0440\u043d\u043e\u0435 \u0432\u0440\u0435\u043c\u044f \u0434\u043b\u044f \u0447\u0442\u0435\u043d\u0438\u044f":"3 \u043c\u0438\u043d\u0443\u0442\u044b"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/"},"author":{"name":"Eugene van der Watt","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa"},"headline":"Simply fine-tuning LLMs can remove alignment guardrails","datePublished":"2023-10-12T14:57:12+00:00","dateModified":"2023-10-13T10:04:42+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/"},"wordCount":490,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/unsafe-LLMs.jpg","keywords":["AI risks","ChatGPT","Llama 2","LLM","Meta","OpenAI"],"articleSection":["Industry"],"inLanguage":"ru-RU"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/","url":"https:\/\/dailyai.com\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/","name":"\u041f\u0440\u043e\u0441\u0442\u0430\u044f \u0442\u043e\u043d\u043a\u0430\u044f \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0430 LLM \u043c\u043e\u0436\u0435\u0442 \u0443\u0441\u0442\u0440\u0430\u043d\u0438\u0442\u044c \u043e\u0433\u0440\u0430\u0436\u0434\u0435\u043d\u0438\u044f \u0432\u044b\u0440\u0430\u0432\u043d\u0438\u0432\u0430\u043d\u0438\u044f | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/unsafe-LLMs.jpg","datePublished":"2023-10-12T14:57:12+00:00","dateModified":"2023-10-13T10:04:42+00:00","breadcrumb":{"@id":"https:\/\/dailyai.com\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/#breadcrumb"},"inLanguage":"ru-RU","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/"]}]},{"@type":"ImageObject","inLanguage":"ru-RU","@id":"https:\/\/dailyai.com\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/unsafe-LLMs.jpg","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/unsafe-LLMs.jpg","width":1000,"height":667},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"Simply fine-tuning LLMs can remove alignment guardrails"}]},{"@type":"WebSite","@id":"https:\/\/dailyai.com\/#website","url":"https:\/\/dailyai.com\/","name":"DailyAI","description":"\u0412\u0430\u0448\u0430 \u0435\u0436\u0435\u0434\u043d\u0435\u0432\u043d\u0430\u044f \u0434\u043e\u0437\u0430 \u043d\u043e\u0432\u043e\u0441\u0442\u0435\u0439 \u043e\u0431 \u0438\u0441\u043a\u0443\u0441\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u043c \u0438\u043d\u0442\u0435\u043b\u043b\u0435\u043a\u0442\u0435","publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/dailyai.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"ru-RU"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DailyAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"ru-RU","@id":"https:\/\/dailyai.com\/#\/schema\/logo\/image\/","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/Daily-Ai_TL_colour.png","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/Daily-Ai_TL_colour.png","width":4501,"height":934,"caption":"DailyAI"},"image":{"@id":"https:\/\/dailyai.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/DailyAIOfficial","https:\/\/www.linkedin.com\/company\/dailyaiofficial\/","https:\/\/www.youtube.com\/@DailyAIOfficial"]},{"@type":"Person","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa","name":"\u042e\u0434\u0436\u0438\u043d \u0432\u0430\u043d \u0434\u0435\u0440 \u0412\u0430\u0442\u0442","image":{"@type":"ImageObject","inLanguage":"ru-RU","@id":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","caption":"Eugene van der Watt"},"description":"\u042e\u0434\u0436\u0438\u043d - \u0432\u044b\u0445\u043e\u0434\u0435\u0446 \u0438\u0437 \u044d\u043b\u0435\u043a\u0442\u0440\u043e\u043d\u043d\u043e\u0439 \u0438\u043d\u0436\u0435\u043d\u0435\u0440\u0438\u0438 \u0438 \u043e\u0431\u043e\u0436\u0430\u0435\u0442 \u0432\u0441\u0435, \u0447\u0442\u043e \u0441\u0432\u044f\u0437\u0430\u043d\u043e \u0441 \u0442\u0435\u0445\u043d\u0438\u043a\u043e\u0439. \u041a\u043e\u0433\u0434\u0430 \u043e\u043d \u043e\u0442\u0434\u044b\u0445\u0430\u0435\u0442 \u043e\u0442 \u0447\u0442\u0435\u043d\u0438\u044f \u043d\u043e\u0432\u043e\u0441\u0442\u0435\u0439 \u043e\u0431 \u0438\u0441\u043a\u0443\u0441\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u043c \u0438\u043d\u0442\u0435\u043b\u043b\u0435\u043a\u0442\u0435, \u0432\u044b \u043c\u043e\u0436\u0435\u0442\u0435 \u043d\u0430\u0439\u0442\u0438 \u0435\u0433\u043e \u0437\u0430 \u0441\u0442\u043e\u043b\u043e\u043c \u0434\u043b\u044f \u0438\u0433\u0440\u044b \u0432 \u0441\u043d\u0443\u043a\u0435\u0440.","sameAs":["www.linkedin.com\/in\/eugene-van-der-watt-16828119"],"url":"https:\/\/dailyai.com\/ru\/author\/eugene\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/posts\/6384","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/comments?post=6384"}],"version-history":[{"count":5,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/posts\/6384\/revisions"}],"predecessor-version":[{"id":6415,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/posts\/6384\/revisions\/6415"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/media\/6386"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/media?parent=6384"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/categories?post=6384"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/tags?post=6384"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}