Viewing enriched Twitter post
@jkcarlsmith @So8res Corrigibilty is hard for different reasons from value alignment. Namely, that it cuts against the grain of coherent reasoning. This is harder to explain and fewer people ask about it, so it is little covered in the book. See eg https://t.co/9mzY3siDhC for coverage of one example.
{
"media": [
{
"type": "photo",
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/1988704036946866619/media_0.jpg?",
"filename": "media_0.jpg"
}
],
"processed_at": "2025-11-27T20:49:57.810200",
"pipeline_version": "2.0"
}{
"type": "tweet",
"id": "1988704036946866619",
"url": "https://x.com/ESYudkowsky/status/1988704036946866619",
"twitterUrl": "https://twitter.com/ESYudkowsky/status/1988704036946866619",
"text": "@jkcarlsmith @So8res Corrigibilty is hard for different reasons from value alignment. Namely, that it cuts against the grain of coherent reasoning. This is harder to explain and fewer people ask about it, so it is little covered in the book. See eg https://t.co/9mzY3siDhC for coverage of one example.",
"source": "Twitter for iPhone",
"retweetCount": 4,
"replyCount": 2,
"likeCount": 42,
"quoteCount": 1,
"viewCount": 3876,
"createdAt": "Wed Nov 12 20:22:45 +0000 2025",
"lang": "en",
"bookmarkCount": 15,
"isReply": true,
"inReplyToId": "1988627991304241448",
"conversationId": "1988627991304241448",
"displayTextRange": [
21,
296
],
"inReplyToUserId": "1349950428",
"inReplyToUsername": "jkcarlsmith",
"author": {
"type": "user",
"userName": "ESYudkowsky",
"url": "https://x.com/ESYudkowsky",
"twitterUrl": "https://twitter.com/ESYudkowsky",
"id": "2595244026",
"name": "Eliezer Yudkowsky âšī¸",
"isVerified": false,
"isBlueVerified": true,
"verifiedType": null,
"profilePicture": "https://pbs.twimg.com/profile_images/1934759522050166788/xKpgxWW5_normal.jpg",
"coverPicture": "https://pbs.twimg.com/profile_banners/2595244026/1527705951",
"description": "",
"location": "",
"followers": 212120,
"following": 102,
"status": "",
"canDm": true,
"canMediaTag": true,
"createdAt": "Sun Jun 29 20:14:32 +0000 2014",
"entities": {
"description": {
"urls": []
},
"url": {}
},
"fastFollowersCount": 0,
"favouritesCount": 36486,
"hasCustomTimelines": true,
"isTranslator": false,
"mediaCount": 459,
"statusesCount": 34025,
"withheldInCountries": [],
"affiliatesHighlightedLabel": {},
"possiblySensitive": false,
"pinnedTweetIds": [
"1968050334707994814"
],
"profile_bio": {
"description": "The original AI alignment person. Understanding the reasons it's difficult since 2003.\n\nThis is my serious low-volume account. Follow @allTheYud for the rest.",
"entities": {
"description": {
"user_mentions": [
{
"id_str": "0",
"indices": [
136,
146
],
"name": "",
"screen_name": "allTheYud"
}
]
}
}
},
"isAutomated": false,
"automatedBy": null
},
"extendedEntities": {},
"card": {
"binding_values": [
{
"key": "thumbnail_image",
"value": {
"image_value": {
"height": 144,
"url": "https://pbs.twimg.com/card_img/1993649497067720705/74HfkRjA?format=jpg&name=144x144_2",
"width": 144
}
}
},
{
"key": "description",
"value": {
"string_value": "The problem of 'fully updated deference' is an obstacle to using moral uncertainty to create corrigibility. One possible scheme in AI alignment is to give the AI a state of moral uncertainty implying..."
}
},
{
"key": "domain",
"value": {
"string_value": "www.lesswrong.com"
}
},
{
"key": "thumbnail_image_large",
"value": {
"image_value": {
"height": 295,
"url": "https://pbs.twimg.com/card_img/1993649497067720705/74HfkRjA?format=jpg&name=420x420_2",
"width": 295
}
}
},
{
"key": "thumbnail_image_original",
"value": {
"image_value": {
"height": 295,
"url": "https://pbs.twimg.com/card_img/1993649497067720705/74HfkRjA?format=jpg&name=orig",
"width": 696
}
}
},
{
"key": "thumbnail_image_small",
"value": {
"image_value": {
"height": 100,
"url": "https://pbs.twimg.com/card_img/1993649497067720705/74HfkRjA?format=jpg&name=100x100_2",
"width": 100
}
}
},
{
"key": "thumbnail_image_x_large",
"value": {
"image_value": {
"height": 295,
"url": "https://pbs.twimg.com/card_img/1993649497067720705/74HfkRjA?format=png&name=2048x2048_2_exp",
"width": 696
}
}
},
{
"key": "vanity_url",
"value": {
"scribe_key": "vanity_url",
"string_value": "lesswrong.com"
}
},
{
"key": "thumbnail_image_color",
"value": {
"image_color_value": {
"palette": [
{
"percentage": 72.58,
"rgb": {
"blue": 229,
"green": 224,
"red": 223
}
},
{
"percentage": 10.16,
"rgb": {
"blue": 153,
"green": 152,
"red": 201
}
},
{
"percentage": 4.24,
"rgb": {
"blue": 139,
"green": 174,
"red": 157
}
},
{
"percentage": 4.01,
"rgb": {
"blue": 103,
"green": 112,
"red": 212
}
},
{
"percentage": 2.69,
"rgb": {
"blue": 216,
"green": 187,
"red": 148
}
}
]
}
}
},
{
"key": "title",
"value": {
"string_value": "Problem of fully updated deference â LessWrong"
}
},
{
"key": "card_url",
"value": {
"scribe_key": "card_url",
"string_value": "https://t.co/Pc0xsPE9Rc"
}
}
],
"card_platform": {
"platform": {
"audience": {
"name": "production"
},
"device": {
"name": "iPhone",
"version": "13"
}
}
},
"name": "summary",
"url": "https://t.co/Pc0xsPE9Rc"
},
"place": {},
"entities": {
"urls": [
{
"display_url": "lesswrong.com/w/problem-of-fâĻ",
"expanded_url": "https://www.lesswrong.com/w/problem-of-fully-updated-deference",
"indices": [
231,
254
],
"url": "https://t.co/9mzY3siDhC"
}
]
},
"quoted_tweet": null,
"retweeted_tweet": null,
"isLimitedReply": false,
"article": null
}