{"id":128756,"date":"2023-08-19T15:26:52","date_gmt":"2023-08-19T19:26:52","guid":{"rendered":"https:\/\/rbach.net\/?p=128756"},"modified":"2024-05-22T19:39:22","modified_gmt":"2024-05-22T23:39:22","slug":"chatgpt-cannot-do-everything","status":"publish","type":"post","link":"https:\/\/rbach.net\/index.php\/chatgpt-cannot-do-everything\/","title":{"rendered":"ChatGPT Cannot Do Everything"},"content":{"rendered":"\r\n<p><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" class=\"alignleft wp-image-128779 size-medium\" title=\"ChatGPT Cannot Do Everything\" src=\"https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/1d83c505-00b8-4a03-90ae-5cffe0b71908.jpg?resize=150%2C150&#038;ssl=1\" alt=\"ChatGPT Cannot Do Everything\" width=\"150\" height=\"150\" srcset=\"https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/1d83c505-00b8-4a03-90ae-5cffe0b71908.jpg?resize=150%2C150&amp;ssl=1 150w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/1d83c505-00b8-4a03-90ae-5cffe0b71908.jpg?resize=75%2C75&amp;ssl=1 75w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/1d83c505-00b8-4a03-90ae-5cffe0b71908.jpg?resize=768%2C768&amp;ssl=1 768w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/1d83c505-00b8-4a03-90ae-5cffe0b71908.jpg?w=1024&amp;ssl=1 1024w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/1d83c505-00b8-4a03-90ae-5cffe0b71908.jpg?w=960&amp;ssl=1 960w\" sizes=\"auto, (max-width: 150px) 100vw, 150px\" \/>ChatGPT, short for Chat Generative Pre-Trained Transformer is the artificial intelligence chatbot from <a title=\"OpenAI\" href=\"https:\/\/openai.com\/\" target=\"_blank\" rel=\"nofollow noopener\">OpenAI<\/a>. \u00a0It is the fastest growing app in Internet history. <a title=\"ChatGPT\" href=\"https:\/\/openai.com\/chatgpt\" target=\"_blank\" rel=\"nofollow noopener\">ChatGPT<\/a> has over <a title=\"59+ ChatGPT Statistics \u2014 Users, Trends &amp; More (July 2023)\" href=\"https:\/\/www.demandsage.com\/chatgpt-statistics\/\" target=\"_blank\" rel=\"nofollow noopener\"><strong>100 million users<\/strong><\/a> globally as of June 2023. Almost <strong><a href=\"https:\/\/www.demandsage.com\/chatgpt-statistics\/\" target=\"_blank\" rel=\"nofollow noopener\">53% of the audience<\/a> believed ChatGPT generated content<\/strong> in the field of finance, health, technology, travel, and entertainment to be created by humans or edited by humans. These types of numbers made me wonder what ChatGPT can&#8217;t do. I asked ChatGPT about some of its limitations and here what it had to say.<\/p>\r\n<h3>ChatGPT can\u2019t provide real-time information<\/h3>\r\n<p>ChatGPT&#8217;s knowledge is based on the data available up until September 2021. It doesn&#8217;t have access to current news or events that have occurred after its knowledge cutoff.<\/p>\r\n<p><strong>Why is ChatGPT&#8217;s knowledge cut-off in September 2021?<\/strong><\/p>\r\n<ul>\r\n<li><em><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" class=\"alignright wp-image-128783\" src=\"https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/6dd2d7ab-71ac-449e-a393-7fc664d6a96a.jpg?resize=110%2C110&#038;ssl=1\" alt=\"\" width=\"110\" height=\"110\" srcset=\"https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/6dd2d7ab-71ac-449e-a393-7fc664d6a96a.jpg?resize=150%2C150&amp;ssl=1 150w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/6dd2d7ab-71ac-449e-a393-7fc664d6a96a.jpg?resize=75%2C75&amp;ssl=1 75w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/6dd2d7ab-71ac-449e-a393-7fc664d6a96a.jpg?resize=768%2C768&amp;ssl=1 768w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/6dd2d7ab-71ac-449e-a393-7fc664d6a96a.jpg?w=1024&amp;ssl=1 1024w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/6dd2d7ab-71ac-449e-a393-7fc664d6a96a.jpg?w=960&amp;ssl=1 960w\" sizes=\"auto, (max-width: 110px) 100vw, 110px\" \/>The knowledge cutoff exists because the model has a fixed dataset up to a certain point, which in this case is September 2021. After this date, any events, developments, or changes that occurred in the world are not part of the model&#8217;s training data. Consequently, ChatGPT cannot access information or knowledge beyond that cutoff date.<\/em><\/li>\r\n<li><em>The reason for the knowledge cutoff is practicality. Training a language model like ChatGPT requires significant computational resources and time. As new data becomes available, it needs to be processed, curated, and used to fine-tune the model, which is a resource-intensive process. To ensure that the model remains up-to-date, periodic retraining is necessary, but it cannot be done instantaneously.<\/em><\/li>\r\n<\/ul>\r\n<h3>ChatGPT cannot predict the future<\/h3>\r\n<p>ChatGPT cannot accurately predict future events or outcomes. It doesn&#8217;t have access to real-time data or the ability to forecast future events.<\/p>\r\n<p><strong>Why can&#8217;t ChatGPT predict the future?<\/strong><\/p>\r\n<ul>\r\n<li><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" class=\"alignright wp-image-128780\" src=\"https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/3e223455-cdb1-4553-9d4b-8756a64299ad-e1688408202105-127x150.jpg?resize=93%2C110&#038;ssl=1\" alt=\"\" width=\"93\" height=\"110\" srcset=\"https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/3e223455-cdb1-4553-9d4b-8756a64299ad-e1688408202105.jpg?resize=127%2C150&amp;ssl=1 127w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/3e223455-cdb1-4553-9d4b-8756a64299ad-e1688408202105.jpg?resize=63%2C75&amp;ssl=1 63w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/3e223455-cdb1-4553-9d4b-8756a64299ad-e1688408202105.jpg?w=727&amp;ssl=1 727w\" sizes=\"auto, (max-width: 93px) 100vw, 93px\" \/> <em>Uncertainty and complexity of the future: The future is inherently uncertain and complex. It depends on a multitude of variables, human actions, and unpredictable events. AI models like ChatGPT cannot accurately capture all these intricate factors and their interplay, making precise predictions challenging.<\/em><\/li>\r\n<li><em>Ethical and practical concerns: Making accurate predictions about the future raises ethical and practical concerns. The potential for misuse, misinformation, or undue influence is significant. AI models are cautious about making definitive future predictions to avoid creating false expectations or spreading unreliable information.<\/em><\/li>\r\n<li><em>It&#8217;s important to note that while AI models may provide probabilistic insights based on historical patterns, they should not be relied upon as crystal balls for predicting specific future events or outcomes. Expert human judgment, domain knowledge, and up-to-date information from reliable sources are crucial for making accurate predictions.<\/em><\/li>\r\n<\/ul>\r\n<h3>Offer professional advice<\/h3>\r\n<p>In many of these cases ChatGPT will offer general advice, but it also includes a disclaimer, &#8220;<em>While I cannot provide personalized advice, here are some points to consider.<\/em>&#8220;<\/p>\r\n<p><strong>Why can&#8217;t ChatGPT offer professional advice?\u00a0<\/strong><\/p>\r\n<ul>\r\n<li><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" class=\"alignright wp-image-128781\" src=\"https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/f7a9b421-503c-4e57-9305-6d3f0a074bb3.jpg?resize=110%2C110&#038;ssl=1\" alt=\"\" width=\"110\" height=\"110\" srcset=\"https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/f7a9b421-503c-4e57-9305-6d3f0a074bb3.jpg?resize=150%2C150&amp;ssl=1 150w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/f7a9b421-503c-4e57-9305-6d3f0a074bb3.jpg?resize=75%2C75&amp;ssl=1 75w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/f7a9b421-503c-4e57-9305-6d3f0a074bb3.jpg?resize=768%2C768&amp;ssl=1 768w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/f7a9b421-503c-4e57-9305-6d3f0a074bb3.jpg?w=1024&amp;ssl=1 1024w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/f7a9b421-503c-4e57-9305-6d3f0a074bb3.jpg?w=960&amp;ssl=1 960w\" sizes=\"auto, (max-width: 110px) 100vw, 110px\" \/> <em>Lack of specialized training: ChatGPT is a general-purpose language model and is not trained in any p<\/em><em>articular profession or domain. It lacks the specialized knowledge and training required to provide accurate and reliable advice in professional areas such as law, medicine, finance, or engineering.<\/em><br \/>\u2022 <em>Risk of misinformation: Providing professional advice without proper training and expertise can lead to misinformation and potential harm. Incorrect advice can have serious consequences in fields where accuracy is critical for making decisions that affect people&#8217;s health, finances, or legal rights.<\/em><br \/><em>\u2022 Legal and ethical concerns: Offering professional advice without the necessary qualifications can be ethically problematic and may even violate laws or regulations in some jurisdictions. It&#8217;s essential to respect professional boundaries and ensure that advice comes from qualified and licensed professionals.<\/em><br \/><em>\u2022 Dynamic nature of professions: Many professional fields are constantly evolving, with new research, practices, and regulations being introduced. AI language models like ChatGPT do not have real-time access to the latest updates and may not be aware of recent developments in specific professions.<\/em><\/li>\r\n<\/ul>\r\n<h3>Guarantee 100% accuracy<\/h3>\r\n<p>ChatGPT aims to provide helpful and relevant information, but it can still generate incorrect or misleading responses. Users should exercise critical thinking and verify information from reliable sources.<\/p>\r\n<p><strong>Why can&#8217;t ChatGPT guarantee 100% accuracy?<\/strong><\/p>\r\n<ul>\r\n<li><em><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" class=\"alignright wp-image-128784\" src=\"https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/e104c710-0849-4ae1-b3ba-f54cca07c8e2.jpg?resize=110%2C110&#038;ssl=1\" alt=\"\" width=\"110\" height=\"110\" srcset=\"https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/e104c710-0849-4ae1-b3ba-f54cca07c8e2.jpg?resize=150%2C150&amp;ssl=1 150w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/e104c710-0849-4ae1-b3ba-f54cca07c8e2.jpg?resize=75%2C75&amp;ssl=1 75w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/e104c710-0849-4ae1-b3ba-f54cca07c8e2.jpg?resize=768%2C768&amp;ssl=1 768w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/e104c710-0849-4ae1-b3ba-f54cca07c8e2.jpg?w=1024&amp;ssl=1 1024w, https:\/\/i0.wp.com\/rbach.net\/wp-content\/uploads\/e104c710-0849-4ae1-b3ba-f54cca07c8e2.jpg?w=960&amp;ssl=1 960w\" sizes=\"auto, (max-width: 110px) 100vw, 110px\" \/>Dependency on training data: ChatGPT learns from patterns in the training data it was exposed to, but the data may contain errors, biases, or incomplete information. As a result, the model may generate responses that are factually incorrect or misleading.<\/em><br \/><em>\u2022 Lack of critical thinking and common sense: While ChatGPT can generate coherent and contextually relevant responses, it lacks true understanding, critical thinking, and common-sense\u00a0reasoning. This can sometimes lead to responses that may sound plausible but are not entirely accurate or appropriate in certain situations.<\/em><br \/><em>\u2022 Interpretation of ambiguous queries: ChatGPT may interpret user queries in ways that humans might not intend or expect. Ambiguous or poorly worded queries can lead to inaccurate or irrelevant responses. Clarifying and refining queries can help mitigate this issue.<\/em><br \/><em>\u2022 Lack of real-time or domain-specific information: ChatGPT&#8217;s training data is not constantly updated, and it has a knowledge cutoff. Therefore, it may not have access to the latest information or developments in various fields, including news, research, or industry-specific knowledge.<\/em><br \/><br \/><\/li>\r\n<\/ul>\r\n<p>It is important to approach responses from ChatGPT with a critical mindset. You must verify information from reliable sources, and exercise judgment when evaluating the accuracy and reliability of the generated content. While AI models aim to provide helpful information, it&#8217;s always wise to cross-reference and consult multiple sources for important or sensitive matters.<br \/>It&#8217;s important to remember that while ChatGPT is a powerful tool, it has limitation. It should only be used as an aid for generating ideas, seeking information, or engaging in casual conversation.<\/p>\r\n<p>&nbsp;<\/p>\r\n<p style=\"text-align: center;\"><em><strong><a title=\"How You Can Help the People of Ukraine\" href=\"https:\/\/www.obama.org\/updates\/help-ukraine\/\" target=\"_blank\" rel=\"nofollow noopener\">How you can help Ukraine!<\/a><\/strong><\/em><\/p>\r\n<p><strong>Related article<\/strong><\/p>\r\n<ul>\r\n<li><a title=\"An Iowa school district is using ChatGPT to decide which books to ban\" href=\"https:\/\/arstechnica.com\/information-technology\/2023\/08\/an-iowa-school-district-is-using-chatgpt-to-decide-which-books-to-ban\/\" target=\"_blank\" rel=\"nofollow noopener\">An Iowa school district is using ChatGPT to decide which books to ban<\/a>\u00a0(<a title=\"ars Technica\" href=\"https:\/\/arstechnica.com\" target=\"_blank\" rel=\"nofollow noopener\">ars Technica<\/a>)<\/li>\r\n<\/ul>\r\n<p>&nbsp;<\/p>\r\n<p><em><a title=\"Ralph Bach\" href=\"https:\/\/rbach.net\/index.php\/new-resume\/\" target=\"_blank\" rel=\"noopener noreferrer\">Ralph Bach<\/a>\u00a0has been in IT long enough to know better and has blogged from his\u00a0<a title=\"Bach Seat\" href=\"https:\/\/rbach.net\/\" target=\"_blank\" rel=\"noopener noreferrer\">Bach Seat<\/a>\u00a0about IT, careers, and anything else that catches his attention since 2005. You can follow him on\u00a0<a class=\"broken_link\" href=\"http:\/\/www.linkedin.com\/in\/rb48334\" target=\"_blank\" rel=\"noopener noreferrer nofollow\">LinkedIn<\/a>,\u00a0<a href=\"https:\/\/www.facebook.com\/ralph.bach.14\" target=\"_blank\" rel=\"noopener noreferrer\">Facebook<\/a>,\u00a0and\u00a0<a href=\"https:\/\/twitter.com\/rbach48334\" target=\"_blank\" rel=\"noopener noreferrer\">Twitter<\/a>. Email the Bach Seat\u00a0<a href=\"mailto:\/\/bach.seat@gmail.com\" target=\"_blank\" rel=\"noopener noreferrer\">here<\/a>.<\/em><\/p>\r\n","protected":false},"excerpt":{"rendered":"<p>ChatGPT has limitations and should be used cautiously as an aid for generating ideas, seeking information, or engaging in casual conversation.<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[],"tags":[3652,2750,3691],"class_list":["post-128756","post","type-post","status-publish","format-standard","hentry","tag-3652","tag-artificial-intelligence","tag-chatgpt"],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/rbach.net\/index.php\/wp-json\/wp\/v2\/posts\/128756","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/rbach.net\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/rbach.net\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/rbach.net\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/rbach.net\/index.php\/wp-json\/wp\/v2\/comments?post=128756"}],"version-history":[{"count":8,"href":"https:\/\/rbach.net\/index.php\/wp-json\/wp\/v2\/posts\/128756\/revisions"}],"predecessor-version":[{"id":131266,"href":"https:\/\/rbach.net\/index.php\/wp-json\/wp\/v2\/posts\/128756\/revisions\/131266"}],"wp:attachment":[{"href":"https:\/\/rbach.net\/index.php\/wp-json\/wp\/v2\/media?parent=128756"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/rbach.net\/index.php\/wp-json\/wp\/v2\/categories?post=128756"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/rbach.net\/index.php\/wp-json\/wp\/v2\/tags?post=128756"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}