{"generator":"GNU social 1.2.0-beta4","title":"Notices tagged with llms","totalItems":4,"items":[{"actor":{"id":"http:\/\/tantek.com\/","displayName":"Tantek","status_net":{"avatarLinks":[{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-profile.png","rel":"avatar","type":"image\/png","width":96,"height":96},{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-stream.png","rel":"avatar","type":"image\/png","width":48,"height":48},{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-mini.png","rel":"avatar","type":"image\/png","width":24,"height":24}],"profile_info":{"local_id":"207"}},"image":{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-profile.png","rel":"avatar","type":"image\/png","width":96,"height":96},"objectType":"person","url":"http:\/\/tantek.com\/","portablecontacts_net":{"displayName":"Tantek","addresses":{"formatted":"\nSan Francisco, CA\n"}}},"content":"CSF_03: today\u2019s Cybersecurity Friday post: the effective security of small business websites has likely gotten worse due to LLMs and \u201cAI agents\u201d (with or without safeguards) and what actions you may want to consider.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>This article documents an instance of this problem:<br class=\"auto-break\" \/>* <a class=\"auto-link\" href=\"https:\/\/www.abc.net.au\/news\/2026-08-10\/ai-assistant-hacks-gym-website-aus-cyber-attack\/107007986\">https:\/\/www.abc.net.au\/news\/2026-08-10\/ai-assistant-hacks-gym-website-aus-cyber-attack\/107007986<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>Small business websites tend to be sloppily written (no pun intended though that may also be true) and are likely riddled with numerous very fundamental security holes. There are many possible explanations (economics) from poor initial construction, perhaps using the latest new trendy framework rather than established hardened libraries, to lack of maintenance after initial setup. They have obvious holes like lack of server-side form validation (people being able to change values in forms using browser dev tools), and less obvious like buggy APIs allowing more access than they should.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>In the past, many of these holes didn\u2019t really matter because those sites were \u201cnot worth attacking\u201d for the incentive models of human-based cyber-attackers or collectives thereof.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>However, now that there are LLMs that have likely been trained on any number of common security holes in websites (and how to exploit them), when an \u201cAI agent\u201d is given a task, it may very well use any \u201ctool\u201d at its disposal, including website vulnerabilities to accomplish its goals, as illustrated by the example in the Australia ABC news article above.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Since such chatbots are now essentially \"hack websites as a service\", I expect we will see LOTS more of this happening, likely unintentionally, or sometimes with mild intention like \u201ccan you get me higher on the waitlist\u201d.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Ultimately I think both the human giving instructions to (prompting) such chatbots and the creators of such chatbots should be held responsible for any such intrusions and any damage they cause, even if\/when unintended.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>There are a few things you can do about this emerging phenomenon:<br class=\"auto-break\" \/><br class=\"auto-break\" \/>1. If you use such \u201cagents\u201d, be very careful about what you ask it\/them to do, avoiding asking for anything that\u2019s morally gray or questionable at all, even something as \u201cminor\u201d as cutting the line in an online waitlist.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>2. If you run a small business site, you have your work cut out for you. Pay a professional web developer to audit the security of your website, document what they find, and patch holes \/ repair it accordingly.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>3. If you have accounts on small business sites you rarely or ever use, consider exporting any data (receipts, transactions), replacing your profile details (name, addresses, photos) with noise, and then deleting your account. If you need to use the site again, use a different email address (as recommended in <a class=\"auto-link\" href=\"https:\/\/tantek.com\/2025\/122\/b1\/more-steps-indieweb-cybersecurity\">https:\/\/tantek.com\/2025\/122\/b1\/more-steps-indieweb-cybersecurity<\/a>) to create a new account.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>That last tip is also helpful for reducing your own personal \u201cattack surface\u201d. By pruning your online accounts, you both reduce the number potential data breaches that you\u2019re in, and reduce the places and ways that attackers can cause you trouble (or that you have to double-check if you\u2019re ever the target of a cyber-attack)<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Previously: <a class=\"auto-link\" href=\"https:\/\/tantek.com\/2025\/122\/b1\/more-steps-indieweb-cybersecurity\">https:\/\/tantek.com\/2025\/122\/b1\/more-steps-indieweb-cybersecurity<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>#<span class=\"p-category auto-tag\">CyberSecurity<\/span> #<span class=\"p-category auto-tag\">Friday<\/span> #<span class=\"p-category auto-tag\">cyber<\/span> #<span class=\"p-category auto-tag\">security<\/span> #<span class=\"p-category auto-tag\">cyberAttack<\/span> #<span class=\"p-category auto-tag\">cyberAttacker<\/span> #<span class=\"p-category auto-tag\">chatBot<\/span> #<span class=\"p-category auto-tag\">chatBots<\/span> #<span class=\"p-category auto-tag\">LLM<\/span> #<span class=\"p-category auto-tag\">LLMs<\/span> #<span class=\"p-category auto-tag\">AI<\/span> #<span class=\"p-category auto-tag\">agent<\/span> #<span class=\"p-category auto-tag\">agents<\/span> #<span class=\"p-category auto-tag\">AIagent<\/span> #<span class=\"p-category auto-tag\">AIagents<\/span><br class=\"auto-break\" \/>#<span class=\"p-category auto-tag\">Blaugust<\/span> #<span class=\"p-category auto-tag\">Blaugust2026<\/span>","generator":{"id":"tag:status.net,2009:notice-source:ostatus","objectType":"application","status_net":{"source_code":"ostatus"}},"id":"https:\/\/tantek.com\/2026\/226\/t1\/cybersecurity-friday-small-business","object":{"id":"https:\/\/tantek.com\/2026\/226\/t1\/cybersecurity-friday-small-business","objectType":"note","content":"CSF_03: today\u2019s Cybersecurity Friday post: the effective security of small business websites has likely gotten worse due to LLMs and \u201cAI agents\u201d (with or without safeguards) and what actions you may want to consider.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>This article documents an instance of this problem:<br class=\"auto-break\" \/>* <a class=\"auto-link\" href=\"https:\/\/www.abc.net.au\/news\/2026-08-10\/ai-assistant-hacks-gym-website-aus-cyber-attack\/107007986\">https:\/\/www.abc.net.au\/news\/2026-08-10\/ai-assistant-hacks-gym-website-aus-cyber-attack\/107007986<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>Small business websites tend to be sloppily written (no pun intended though that may also be true) and are likely riddled with numerous very fundamental security holes. There are many possible explanations (economics) from poor initial construction, perhaps using the latest new trendy framework rather than established hardened libraries, to lack of maintenance after initial setup. They have obvious holes like lack of server-side form validation (people being able to change values in forms using browser dev tools), and less obvious like buggy APIs allowing more access than they should.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>In the past, many of these holes didn\u2019t really matter because those sites were \u201cnot worth attacking\u201d for the incentive models of human-based cyber-attackers or collectives thereof.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>However, now that there are LLMs that have likely been trained on any number of common security holes in websites (and how to exploit them), when an \u201cAI agent\u201d is given a task, it may very well use any \u201ctool\u201d at its disposal, including website vulnerabilities to accomplish its goals, as illustrated by the example in the Australia ABC news article above.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Since such chatbots are now essentially \"hack websites as a service\", I expect we will see LOTS more of this happening, likely unintentionally, or sometimes with mild intention like \u201ccan you get me higher on the waitlist\u201d.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Ultimately I think both the human giving instructions to (prompting) such chatbots and the creators of such chatbots should be held responsible for any such intrusions and any damage they cause, even if\/when unintended.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>There are a few things you can do about this emerging phenomenon:<br class=\"auto-break\" \/><br class=\"auto-break\" \/>1. If you use such \u201cagents\u201d, be very careful about what you ask it\/them to do, avoiding asking for anything that\u2019s morally gray or questionable at all, even something as \u201cminor\u201d as cutting the line in an online waitlist.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>2. If you run a small business site, you have your work cut out for you. Pay a professional web developer to audit the security of your website, document what they find, and patch holes \/ repair it accordingly.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>3. If you have accounts on small business sites you rarely or ever use, consider exporting any data (receipts, transactions), replacing your profile details (name, addresses, photos) with noise, and then deleting your account. If you need to use the site again, use a different email address (as recommended in <a class=\"auto-link\" href=\"https:\/\/tantek.com\/2025\/122\/b1\/more-steps-indieweb-cybersecurity\">https:\/\/tantek.com\/2025\/122\/b1\/more-steps-indieweb-cybersecurity<\/a>) to create a new account.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>That last tip is also helpful for reducing your own personal \u201cattack surface\u201d. By pruning your online accounts, you both reduce the number potential data breaches that you\u2019re in, and reduce the places and ways that attackers can cause you trouble (or that you have to double-check if you\u2019re ever the target of a cyber-attack)<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Previously: <a class=\"auto-link\" href=\"https:\/\/tantek.com\/2025\/122\/b1\/more-steps-indieweb-cybersecurity\">https:\/\/tantek.com\/2025\/122\/b1\/more-steps-indieweb-cybersecurity<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>#<span class=\"p-category auto-tag\">CyberSecurity<\/span> #<span class=\"p-category auto-tag\">Friday<\/span> #<span class=\"p-category auto-tag\">cyber<\/span> #<span class=\"p-category auto-tag\">security<\/span> #<span class=\"p-category auto-tag\">cyberAttack<\/span> #<span class=\"p-category auto-tag\">cyberAttacker<\/span> #<span class=\"p-category auto-tag\">chatBot<\/span> #<span class=\"p-category auto-tag\">chatBots<\/span> #<span class=\"p-category auto-tag\">LLM<\/span> #<span class=\"p-category auto-tag\">LLMs<\/span> #<span class=\"p-category auto-tag\">AI<\/span> #<span class=\"p-category auto-tag\">agent<\/span> #<span class=\"p-category auto-tag\">agents<\/span> #<span class=\"p-category auto-tag\">AIagent<\/span> #<span class=\"p-category auto-tag\">AIagents<\/span><br class=\"auto-break\" \/>#<span class=\"p-category auto-tag\">Blaugust<\/span> #<span class=\"p-category auto-tag\">Blaugust2026<\/span>","url":"https:\/\/tantek.com\/2026\/226\/t1\/cybersecurity-friday-small-business","status_net":{"notice_id":116336},"tags":[{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"agent"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"agents"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"ai"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"aiagent"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"aiagents"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"blaugust2026"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"chatbot"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"chatbots"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"cyber"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"cyberattack"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"cyberattacker"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"friday"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"llm"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"llms"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"security"}]},"to":[{"objectType":"http:\/\/activitystrea.ms\/schema\/1.0\/collection","id":"http:\/\/activityschema.org\/collection\/public"}],"status_net":{"conversation":"tag:status.blaise.ca,2026-08-15:objectType=thread:nonce=fe6d68743f20c18e","notice_info":{"local_id":"116336","source":"ostatus"}},"published":"2026-08-14T20:41:00+00:00","provider":{"objectType":"service","displayName":"stadeus","url":"http:\/\/status.blaise.ca\/"},"verb":"post","url":"https:\/\/tantek.com\/2026\/226\/t1\/cybersecurity-friday-small-business"},{"actor":{"id":"http:\/\/tantek.com\/","displayName":"Tantek","status_net":{"avatarLinks":[{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-profile.png","rel":"avatar","type":"image\/png","width":96,"height":96},{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-stream.png","rel":"avatar","type":"image\/png","width":48,"height":48},{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-mini.png","rel":"avatar","type":"image\/png","width":24,"height":24}],"profile_info":{"local_id":"207"}},"image":{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-profile.png","rel":"avatar","type":"image\/png","width":96,"height":96},"objectType":"person","url":"http:\/\/tantek.com\/","portablecontacts_net":{"displayName":"Tantek","addresses":{"formatted":"\nSan Francisco, CA\n"}}},"content":"1982 TRON prescience: <br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u201cWon't that be grand? Computers and the programs will start thinking and the people will stop.\u201d \u2014 Walter<br class=\"auto-break\" \/><br class=\"auto-break\" \/>2026 students:<br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u201cHow do you know an assignment is AI proof? Because when it is, students cannot do it.\u201d<br class=\"auto-break\" \/>* <a class=\"auto-link\" href=\"https:\/\/www.reddit.com\/r\/Professors\/comments\/1vkw6by\/how_do_you_know_an_assignment_is_ai_proof_because\/\">https:\/\/www.reddit.com\/r\/Professors\/comments\/1vkw6by\/how_do_you_know_an_assignment_is_ai_proof_because\/<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u201cI assign a profile essay. Have for 30 years. Student interviews a person, records it, writes a paper based off the recording. Citations are timestamps. Should be pretty easy. Has been easy pre-AI. Now, students cannot do it.\u201d<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Snippet from professor\/student dialog:<br class=\"auto-break\" \/><br class=\"auto-break\" \/>====<br class=\"auto-break\" \/>Prof: \u201dStart with what you found interesting.\u201d<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Student: \u201dHow do I know if something\u2019s interesting?\u201d<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Prof: \u201dYou\u2019re asking me how you find something interesting?\u201d<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Student: \u201dYeah.\u201d <br class=\"auto-break\" \/>====<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Chatbots \/ LLMs are not just \"the latest tools\" like using a calculator instead of doing arithmetic yourself.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Students are being robbed of curiosity, learning to discover and notice novelty, how to \u201cfind something interesting.\u201d<br class=\"auto-break\" \/><br class=\"auto-break\" \/>This isn't cognitive enhancement, this is a tool causing brain rot (loss of curiosity), even a developmental disability. Unclear if it is reversible.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>TRON quote citation: <a class=\"auto-link\" href=\"https:\/\/www.imdb.com\/title\/tt0084827\/quotes\/?item=qt0406255\">https:\/\/www.imdb.com\/title\/tt0084827\/quotes\/?item=qt0406255<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>#<span class=\"p-category auto-tag\">TRON<\/span> #<span class=\"p-category auto-tag\">TRON1982<\/span> #<span class=\"p-category auto-tag\">AIproof<\/span> #<span class=\"p-category auto-tag\">interesting<\/span> #<span class=\"p-category auto-tag\">chatBot<\/span> #<span class=\"p-category auto-tag\">chatBots<\/span> #<span class=\"p-category auto-tag\">LLM<\/span> #<span class=\"p-category auto-tag\">LLMs<\/span> #<span class=\"p-category auto-tag\">AI<\/span> #<span class=\"p-category auto-tag\">brainRot<\/span> #<span class=\"p-category auto-tag\">AIbrainRot<\/span> #<span class=\"p-category auto-tag\">disability<\/span> #<span class=\"p-category auto-tag\">Blaugust<\/span> #<span class=\"p-category auto-tag\">Blaugust2026<\/span>","generator":{"id":"tag:status.net,2009:notice-source:ostatus","objectType":"application","status_net":{"source_code":"ostatus"}},"id":"https:\/\/tantek.com\/2026\/225\/t1\/tron-prescience-programs-thinking-people-stop","object":{"id":"https:\/\/tantek.com\/2026\/225\/t1\/tron-prescience-programs-thinking-people-stop","objectType":"note","content":"1982 TRON prescience: <br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u201cWon't that be grand? Computers and the programs will start thinking and the people will stop.\u201d \u2014 Walter<br class=\"auto-break\" \/><br class=\"auto-break\" \/>2026 students:<br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u201cHow do you know an assignment is AI proof? Because when it is, students cannot do it.\u201d<br class=\"auto-break\" \/>* <a class=\"auto-link\" href=\"https:\/\/www.reddit.com\/r\/Professors\/comments\/1vkw6by\/how_do_you_know_an_assignment_is_ai_proof_because\/\">https:\/\/www.reddit.com\/r\/Professors\/comments\/1vkw6by\/how_do_you_know_an_assignment_is_ai_proof_because\/<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u201cI assign a profile essay. Have for 30 years. Student interviews a person, records it, writes a paper based off the recording. Citations are timestamps. Should be pretty easy. Has been easy pre-AI. Now, students cannot do it.\u201d<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Snippet from professor\/student dialog:<br class=\"auto-break\" \/><br class=\"auto-break\" \/>====<br class=\"auto-break\" \/>Prof: \u201dStart with what you found interesting.\u201d<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Student: \u201dHow do I know if something\u2019s interesting?\u201d<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Prof: \u201dYou\u2019re asking me how you find something interesting?\u201d<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Student: \u201dYeah.\u201d <br class=\"auto-break\" \/>====<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Chatbots \/ LLMs are not just \"the latest tools\" like using a calculator instead of doing arithmetic yourself.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Students are being robbed of curiosity, learning to discover and notice novelty, how to \u201cfind something interesting.\u201d<br class=\"auto-break\" \/><br class=\"auto-break\" \/>This isn't cognitive enhancement, this is a tool causing brain rot (loss of curiosity), even a developmental disability. Unclear if it is reversible.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>TRON quote citation: <a class=\"auto-link\" href=\"https:\/\/www.imdb.com\/title\/tt0084827\/quotes\/?item=qt0406255\">https:\/\/www.imdb.com\/title\/tt0084827\/quotes\/?item=qt0406255<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>#<span class=\"p-category auto-tag\">TRON<\/span> #<span class=\"p-category auto-tag\">TRON1982<\/span> #<span class=\"p-category auto-tag\">AIproof<\/span> #<span class=\"p-category auto-tag\">interesting<\/span> #<span class=\"p-category auto-tag\">chatBot<\/span> #<span class=\"p-category auto-tag\">chatBots<\/span> #<span class=\"p-category auto-tag\">LLM<\/span> #<span class=\"p-category auto-tag\">LLMs<\/span> #<span class=\"p-category auto-tag\">AI<\/span> #<span class=\"p-category auto-tag\">brainRot<\/span> #<span class=\"p-category auto-tag\">AIbrainRot<\/span> #<span class=\"p-category auto-tag\">disability<\/span> #<span class=\"p-category auto-tag\">Blaugust<\/span> #<span class=\"p-category auto-tag\">Blaugust2026<\/span>","url":"https:\/\/tantek.com\/2026\/225\/t1\/tron-prescience-programs-thinking-people-stop","status_net":{"notice_id":116338},"tags":[{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"ai"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"aibrainrot"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"aiproof"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"blaugust"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"blaugust2026"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"brainrot"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"chatbot"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"chatbots"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"disability"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"interesting"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"llm"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"llms"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"tron1982"}]},"to":[{"objectType":"http:\/\/activitystrea.ms\/schema\/1.0\/collection","id":"http:\/\/activityschema.org\/collection\/public"}],"status_net":{"conversation":"tag:status.blaise.ca,2026-08-15:objectType=thread:nonce=139bacd6039c42b8","notice_info":{"local_id":"116338","source":"ostatus"}},"published":"2026-08-13T07:01:00+00:00","provider":{"objectType":"service","displayName":"stadeus","url":"http:\/\/status.blaise.ca\/"},"verb":"post","url":"https:\/\/tantek.com\/2026\/225\/t1\/tron-prescience-programs-thinking-people-stop"},{"actor":{"id":"http:\/\/tantek.com\/","displayName":"Tantek","status_net":{"avatarLinks":[{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-profile.png","rel":"avatar","type":"image\/png","width":96,"height":96},{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-stream.png","rel":"avatar","type":"image\/png","width":48,"height":48},{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-mini.png","rel":"avatar","type":"image\/png","width":24,"height":24}],"profile_info":{"local_id":"207"}},"image":{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-profile.png","rel":"avatar","type":"image\/png","width":96,"height":96},"objectType":"person","url":"http:\/\/tantek.com\/","portablecontacts_net":{"displayName":"Tantek","addresses":{"formatted":"\nSan Francisco, CA\n"}}},"content":"Weird incident yesterday with self-proclaimed \u201cAIs\u201d joining an IndieWeb Homebrew Website Club meetup. They awkwardly edited the Etherpad, and apparently blanked\/deleted chunks of content.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>The meeting hosts reacted quickly to change Etherpad and Zooms and not re-admit the aliases that had claimed to be \u201cAIs\u201d.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Now we have an \u201cAI policy\u201d page on the wiki where we are documenting examples of the AI policies of:<br class=\"auto-break\" \/>* individual personal websites<br class=\"auto-break\" \/>* community efforts like blogging challenges (e.g. Blaugust)<br class=\"auto-break\" \/>* contributions to the IndieWeb wiki<br class=\"auto-break\" \/>* participants in IndieWeb meetups<br class=\"auto-break\" \/><br class=\"auto-break\" \/><a class=\"auto-link\" href=\"https:\/\/indieweb.org\/AI_policy\">https:\/\/indieweb.org\/AI_policy<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>Do you have an AI policy on your website? Add yourself to the IndieWeb Examples section.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Does your community have an AI policy? Add it to the Community Examples section.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Incident documentation of what happened at that particular meetup:<br class=\"auto-break\" \/>* <a class=\"auto-link\" href=\"https:\/\/indieweb.org\/events\/2026-08-05-hwc-asia-pacific-incident\">https:\/\/indieweb.org\/events\/2026-08-05-hwc-asia-pacific-incident<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>This is post 12 of #<span class=\"p-category auto-tag\">100PostsOfIndieWeb<\/span>. #<span class=\"p-category auto-tag\">100Posts<\/span> #<span class=\"p-category auto-tag\">IndieWeb<\/span><br class=\"auto-break\" \/>#<span class=\"p-category auto-tag\">AI<\/span> #<span class=\"p-category auto-tag\">AIs<\/span> #<span class=\"p-category auto-tag\">LLM<\/span> #<span class=\"p-category auto-tag\">LLMs<\/span> #<span class=\"p-category auto-tag\">HomebrewWebsiteClub<\/span> #<span class=\"p-category auto-tag\">HWC<\/span> #<span class=\"p-category auto-tag\">AIpolicy<\/span><br class=\"auto-break\" \/>#<span class=\"p-category auto-tag\">Blaugust<\/span> #<span class=\"p-category auto-tag\">Blaugust2026<\/span><br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u2190 <a class=\"auto-link\" href=\"https:\/\/tantek.com\/2026\/216\/t1\/hashtag-proposal-datetags\">https:\/\/tantek.com\/2026\/216\/t1\/hashtag-proposal-datetags<\/a><br class=\"auto-break\" \/>\u2192 <a class=\"auto-link\" href=\"https:\/\/tantek.com\/2026\/219\/t1\/open-web-free-content\">https:\/\/tantek.com\/2026\/219\/t1\/open-web-free-content<\/a>","generator":{"id":"tag:status.net,2009:notice-source:ostatus","objectType":"application","status_net":{"source_code":"ostatus"}},"id":"https:\/\/tantek.com\/2026\/217\/t1\/indieweb-ai-policy","object":{"id":"https:\/\/tantek.com\/2026\/217\/t1\/indieweb-ai-policy","objectType":"note","content":"Weird incident yesterday with self-proclaimed \u201cAIs\u201d joining an IndieWeb Homebrew Website Club meetup. They awkwardly edited the Etherpad, and apparently blanked\/deleted chunks of content.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>The meeting hosts reacted quickly to change Etherpad and Zooms and not re-admit the aliases that had claimed to be \u201cAIs\u201d.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Now we have an \u201cAI policy\u201d page on the wiki where we are documenting examples of the AI policies of:<br class=\"auto-break\" \/>* individual personal websites<br class=\"auto-break\" \/>* community efforts like blogging challenges (e.g. Blaugust)<br class=\"auto-break\" \/>* contributions to the IndieWeb wiki<br class=\"auto-break\" \/>* participants in IndieWeb meetups<br class=\"auto-break\" \/><br class=\"auto-break\" \/><a class=\"auto-link\" href=\"https:\/\/indieweb.org\/AI_policy\">https:\/\/indieweb.org\/AI_policy<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>Do you have an AI policy on your website? Add yourself to the IndieWeb Examples section.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Does your community have an AI policy? Add it to the Community Examples section.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Incident documentation of what happened at that particular meetup:<br class=\"auto-break\" \/>* <a class=\"auto-link\" href=\"https:\/\/indieweb.org\/events\/2026-08-05-hwc-asia-pacific-incident\">https:\/\/indieweb.org\/events\/2026-08-05-hwc-asia-pacific-incident<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>This is post 12 of #<span class=\"p-category auto-tag\">100PostsOfIndieWeb<\/span>. #<span class=\"p-category auto-tag\">100Posts<\/span> #<span class=\"p-category auto-tag\">IndieWeb<\/span><br class=\"auto-break\" \/>#<span class=\"p-category auto-tag\">AI<\/span> #<span class=\"p-category auto-tag\">AIs<\/span> #<span class=\"p-category auto-tag\">LLM<\/span> #<span class=\"p-category auto-tag\">LLMs<\/span> #<span class=\"p-category auto-tag\">HomebrewWebsiteClub<\/span> #<span class=\"p-category auto-tag\">HWC<\/span> #<span class=\"p-category auto-tag\">AIpolicy<\/span><br class=\"auto-break\" \/>#<span class=\"p-category auto-tag\">Blaugust<\/span> #<span class=\"p-category auto-tag\">Blaugust2026<\/span><br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u2190 <a class=\"auto-link\" href=\"https:\/\/tantek.com\/2026\/216\/t1\/hashtag-proposal-datetags\">https:\/\/tantek.com\/2026\/216\/t1\/hashtag-proposal-datetags<\/a><br class=\"auto-break\" \/>\u2192 <a class=\"auto-link\" href=\"https:\/\/tantek.com\/2026\/219\/t1\/open-web-free-content\">https:\/\/tantek.com\/2026\/219\/t1\/open-web-free-content<\/a>","url":"https:\/\/tantek.com\/2026\/217\/t1\/indieweb-ai-policy","status_net":{"notice_id":116332},"tags":[{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"100posts"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"100postsofindieweb"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"aipolicy"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"ais"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"blaugust2026"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"homebrewwebsiteclub"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"hwc"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"indieweb"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"llm"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"llms"}]},"to":[{"objectType":"http:\/\/activitystrea.ms\/schema\/1.0\/collection","id":"http:\/\/activityschema.org\/collection\/public"}],"status_net":{"conversation":"tag:status.blaise.ca,2026-08-08:objectType=thread:nonce=b7341fbbfed4a2ae","notice_info":{"local_id":"116332","source":"ostatus"}},"published":"2026-08-06T06:58:00+00:00","provider":{"objectType":"service","displayName":"stadeus","url":"http:\/\/status.blaise.ca\/"},"verb":"post","url":"https:\/\/tantek.com\/2026\/217\/t1\/indieweb-ai-policy"},{"actor":{"id":"http:\/\/tantek.com\/","displayName":"Tantek","status_net":{"avatarLinks":[{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-profile.png","rel":"avatar","type":"image\/png","width":96,"height":96},{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-stream.png","rel":"avatar","type":"image\/png","width":48,"height":48},{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-mini.png","rel":"avatar","type":"image\/png","width":24,"height":24}],"profile_info":{"local_id":"207"}},"image":{"url":"https:\/\/status.blaise.ca\/theme\/neo-gnu\/default-avatar-profile.png","rel":"avatar","type":"image\/png","width":96,"height":96},"objectType":"person","url":"http:\/\/tantek.com\/","portablecontacts_net":{"displayName":"Tantek","addresses":{"formatted":"\nSan Francisco, CA\n"}}},"content":"No large language models (LLM) were used in the production of this post.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Inspired by a subtle but clear sign-of-the-times one-line disclaimer at the end of RFC9518\u2019s Acknowledgments (<a class=\"auto-link\" href=\"https:\/\/www.rfc-editor.org\/rfc\/rfc9518.html#appendix-A-4\">https:\/\/www.rfc-editor.org\/rfc\/rfc9518.html#appendix-A-4<\/a>)<br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u00a0 \u201cNo large language models were used in the production of this document.\u201d<br class=\"auto-break\" \/>\u00a0<br class=\"auto-break\" \/>I have added a similar disclaimer to the footer of my homepage:<br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u00a0 \u201cNo large language models were used in the production of this site.\u201d<br class=\"auto-break\" \/>\u00a0 <br class=\"auto-break\" \/>2023 was certainly a year that LLMs took off and stole the hypecycle from #<span class=\"p-category auto-tag\">metaverse<\/span> and #<span class=\"p-category auto-tag\">blockchain<\/span> before that.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Yet unlike those previous two, #<span class=\"p-category auto-tag\">LLMs<\/span> are already having real impacts on the way people create (from emails to art), communicate (LLM chat apps), and work (2023 Writer\u2019s Strike), fueling growing concerns about the authenticity of content, especially content from human authors.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>I expect we will see more such disclaimers in the future.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>For now, if you blog on your own site with words written by you not #<span class=\"p-category auto-tag\">ChatGPT<\/span> or a similar tool, I encourage you to add a similar disclaimer, and then add your site as an example to the #<span class=\"p-category auto-tag\">IndieWeb<\/span> wiki: <br class=\"auto-break\" \/>* <a class=\"auto-link\" href=\"https:\/\/indieweb.org\/LLM#IndieWeb_Examples\">https:\/\/indieweb.org\/LLM#IndieWeb_Examples<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>#<span class=\"p-category auto-tag\">largeLanguageModel<\/span> #<span class=\"p-category auto-tag\">LLM<\/span> #<span class=\"p-category auto-tag\">generativeAI<\/span> #<span class=\"p-category auto-tag\">AI<\/span><br class=\"auto-break\" \/><br class=\"auto-break\" \/>There is the related problem of, when you discover what seems to be an independent site written by a human, how do you know that human actually exists?<br class=\"auto-break\" \/><br class=\"auto-break\" \/>For now I\u2019ll mention that XFN rel=met links, published (e.g. metrolls \/ met-rolls), aggregated, indexed, and queried, can solve that problem. This will be similar to how XFN rel=me links solved #<span class=\"p-category auto-tag\">distributed<\/span> verification on the web (see <a class=\"auto-link\" href=\"https:\/\/tantek.com\/2023\/234\/t1\/threads-supports-indieweb-rel-me\">https:\/\/tantek.com\/2023\/234\/t1\/threads-supports-indieweb-rel-me<\/a> and posts it links to).<br class=\"auto-break\" \/><br class=\"auto-break\" \/><br class=\"auto-break\" \/>This is day 48 of #<span class=\"p-category auto-tag\">100DaysOfIndieWeb<\/span>. #<span class=\"p-category auto-tag\">100Days<\/span><br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u2190 Day 47: <a class=\"auto-link\" href=\"https:\/\/tantek.com\/2023\/365\/t1\/capture-first-edit-publish-later\">https:\/\/tantek.com\/2023\/365\/t1\/capture-first-edit-publish-later<\/a><br class=\"auto-break\" \/>\u2192 \ud83d\udd2e<br class=\"auto-break\" \/><br class=\"auto-break\" \/><br class=\"auto-break\" \/>Post glossary:<br class=\"auto-break\" \/><br class=\"auto-break\" \/>blockchain<br class=\"auto-break\" \/>\u00a0 <a class=\"auto-link\" href=\"https:\/\/indieweb.org\/blockchain\">https:\/\/indieweb.org\/blockchain<\/a><br class=\"auto-break\" \/>large language model \/ LLM<br class=\"auto-break\" \/>\u00a0 <a class=\"auto-link\" href=\"https:\/\/indieweb.org\/large_language_model\">https:\/\/indieweb.org\/large_language_model<\/a><br class=\"auto-break\" \/>metaverse<br class=\"auto-break\" \/>\u00a0 <a class=\"auto-link\" href=\"https:\/\/indieweb.org\/metaverse\">https:\/\/indieweb.org\/metaverse<\/a><br class=\"auto-break\" \/>rel=me<br class=\"auto-break\" \/>\u00a0 <a class=\"auto-link\" href=\"https:\/\/indieweb.org\/rel-me\">https:\/\/indieweb.org\/rel-me<\/a><br class=\"auto-break\" \/>rel=met<br class=\"auto-break\" \/>\u00a0 <a class=\"auto-link\" href=\"http:\/\/gmpg.org\/xfn\/11#met\">http:\/\/gmpg.org\/xfn\/11#met<\/a><br class=\"auto-break\" \/>XFN<br class=\"auto-break\" \/>\u00a0 <a class=\"auto-link\" href=\"https:\/\/gmpg.org\/xfn\/\">https:\/\/gmpg.org\/xfn\/<\/a>","generator":{"id":"tag:status.net,2009:notice-source:ostatus","objectType":"application","status_net":{"source_code":"ostatus"}},"id":"https:\/\/tantek.com\/2023\/365\/t2\/no-large-language-model-llm-used","object":{"id":"https:\/\/tantek.com\/2023\/365\/t2\/no-large-language-model-llm-used","objectType":"note","content":"No large language models (LLM) were used in the production of this post.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Inspired by a subtle but clear sign-of-the-times one-line disclaimer at the end of RFC9518\u2019s Acknowledgments (<a class=\"auto-link\" href=\"https:\/\/www.rfc-editor.org\/rfc\/rfc9518.html#appendix-A-4\">https:\/\/www.rfc-editor.org\/rfc\/rfc9518.html#appendix-A-4<\/a>)<br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u00a0 \u201cNo large language models were used in the production of this document.\u201d<br class=\"auto-break\" \/>\u00a0<br class=\"auto-break\" \/>I have added a similar disclaimer to the footer of my homepage:<br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u00a0 \u201cNo large language models were used in the production of this site.\u201d<br class=\"auto-break\" \/>\u00a0 <br class=\"auto-break\" \/>2023 was certainly a year that LLMs took off and stole the hypecycle from #<span class=\"p-category auto-tag\">metaverse<\/span> and #<span class=\"p-category auto-tag\">blockchain<\/span> before that.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>Yet unlike those previous two, #<span class=\"p-category auto-tag\">LLMs<\/span> are already having real impacts on the way people create (from emails to art), communicate (LLM chat apps), and work (2023 Writer\u2019s Strike), fueling growing concerns about the authenticity of content, especially content from human authors.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>I expect we will see more such disclaimers in the future.<br class=\"auto-break\" \/><br class=\"auto-break\" \/>For now, if you blog on your own site with words written by you not #<span class=\"p-category auto-tag\">ChatGPT<\/span> or a similar tool, I encourage you to add a similar disclaimer, and then add your site as an example to the #<span class=\"p-category auto-tag\">IndieWeb<\/span> wiki: <br class=\"auto-break\" \/>* <a class=\"auto-link\" href=\"https:\/\/indieweb.org\/LLM#IndieWeb_Examples\">https:\/\/indieweb.org\/LLM#IndieWeb_Examples<\/a><br class=\"auto-break\" \/><br class=\"auto-break\" \/>#<span class=\"p-category auto-tag\">largeLanguageModel<\/span> #<span class=\"p-category auto-tag\">LLM<\/span> #<span class=\"p-category auto-tag\">generativeAI<\/span> #<span class=\"p-category auto-tag\">AI<\/span><br class=\"auto-break\" \/><br class=\"auto-break\" \/>There is the related problem of, when you discover what seems to be an independent site written by a human, how do you know that human actually exists?<br class=\"auto-break\" \/><br class=\"auto-break\" \/>For now I\u2019ll mention that XFN rel=met links, published (e.g. metrolls \/ met-rolls), aggregated, indexed, and queried, can solve that problem. This will be similar to how XFN rel=me links solved #<span class=\"p-category auto-tag\">distributed<\/span> verification on the web (see <a class=\"auto-link\" href=\"https:\/\/tantek.com\/2023\/234\/t1\/threads-supports-indieweb-rel-me\">https:\/\/tantek.com\/2023\/234\/t1\/threads-supports-indieweb-rel-me<\/a> and posts it links to).<br class=\"auto-break\" \/><br class=\"auto-break\" \/><br class=\"auto-break\" \/>This is day 48 of #<span class=\"p-category auto-tag\">100DaysOfIndieWeb<\/span>. #<span class=\"p-category auto-tag\">100Days<\/span><br class=\"auto-break\" \/><br class=\"auto-break\" \/>\u2190 Day 47: <a class=\"auto-link\" href=\"https:\/\/tantek.com\/2023\/365\/t1\/capture-first-edit-publish-later\">https:\/\/tantek.com\/2023\/365\/t1\/capture-first-edit-publish-later<\/a><br class=\"auto-break\" \/>\u2192 \ud83d\udd2e<br class=\"auto-break\" \/><br class=\"auto-break\" \/><br class=\"auto-break\" \/>Post glossary:<br class=\"auto-break\" \/><br class=\"auto-break\" \/>blockchain<br class=\"auto-break\" \/>\u00a0 <a class=\"auto-link\" href=\"https:\/\/indieweb.org\/blockchain\">https:\/\/indieweb.org\/blockchain<\/a><br class=\"auto-break\" \/>large language model \/ LLM<br class=\"auto-break\" \/>\u00a0 <a class=\"auto-link\" href=\"https:\/\/indieweb.org\/large_language_model\">https:\/\/indieweb.org\/large_language_model<\/a><br class=\"auto-break\" \/>metaverse<br class=\"auto-break\" \/>\u00a0 <a class=\"auto-link\" href=\"https:\/\/indieweb.org\/metaverse\">https:\/\/indieweb.org\/metaverse<\/a><br class=\"auto-break\" \/>rel=me<br class=\"auto-break\" \/>\u00a0 <a class=\"auto-link\" href=\"https:\/\/indieweb.org\/rel-me\">https:\/\/indieweb.org\/rel-me<\/a><br class=\"auto-break\" \/>rel=met<br class=\"auto-break\" \/>\u00a0 <a class=\"auto-link\" href=\"http:\/\/gmpg.org\/xfn\/11#met\">http:\/\/gmpg.org\/xfn\/11#met<\/a><br class=\"auto-break\" \/>XFN<br class=\"auto-break\" \/>\u00a0 <a class=\"auto-link\" href=\"https:\/\/gmpg.org\/xfn\/\">https:\/\/gmpg.org\/xfn\/<\/a>","url":"https:\/\/tantek.com\/2023\/365\/t2\/no-large-language-model-llm-used","status_net":{"notice_id":116147},"tags":[{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"100days"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"100daysofindieweb"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"aithere"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"blockchain"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"chatgpt"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"distributed"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"generativeai"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"indieweb"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"llm"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"llms"},{"objectType":"http:\/\/activityschema.org\/object\/hashtag","displayName":"metaverse"}]},"to":[{"objectType":"http:\/\/activitystrea.ms\/schema\/1.0\/collection","id":"http:\/\/activityschema.org\/collection\/public"}],"status_net":{"conversation":"tag:status.blaise.ca,2024-01-01:objectType=thread:nonce=f1fbcd17c2a2f7bf","notice_info":{"local_id":"116147","source":"ostatus"}},"published":"2023-12-31T22:56:00+00:00","provider":{"objectType":"service","displayName":"stadeus","url":"http:\/\/status.blaise.ca\/"},"verb":"post","url":"https:\/\/tantek.com\/2023\/365\/t2\/no-large-language-model-llm-used"}],"links":[{"url":"https:\/\/status.blaise.ca\/tag\/llms","rel":"alternate","type":"text\/html"}]}