Viewing enriched Twitter post
New Anthropic Fellows research: Automated audits like Petri are increasingly used for alignment evals, but they're often unrealistic, and frontier LLMs can often tell. We measure and improve the realism of agentic coding audits by grounding them in real deployment data. https://t.co/R1BIZOJrtI
{
"media": [
{
"url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2036133603843908093/media_0.jpg",
"media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2036133603843908093/media_0.jpg",
"type": "photo",
"filename": "media_0.jpg"
}
],
"processed_at": "2026-03-23T20:47:56.346940",
"pipeline_version": "2.0"
}{
"type": "tweet",
"id": "2036133603843908093",
"url": "https://x.com/Connor_Kissane/status/2036133603843908093",
"twitterUrl": "https://twitter.com/Connor_Kissane/status/2036133603843908093",
"text": "New Anthropic Fellows research: Automated audits like Petri are increasingly used for alignment evals, but they're often unrealistic, and frontier LLMs can often tell. We measure and improve the realism of agentic coding audits by grounding them in real deployment data. https://t.co/R1BIZOJrtI",
"source": "Twitter for iPhone",
"retweetCount": 7,
"replyCount": 2,
"likeCount": 64,
"quoteCount": 1,
"viewCount": 2937,
"createdAt": "Mon Mar 23 17:30:55 +0000 2026",
"lang": "en",
"bookmarkCount": 34,
"isReply": false,
"inReplyToId": null,
"conversationId": "2036133603843908093",
"displayTextRange": [
0,
270
],
"inReplyToUserId": null,
"inReplyToUsername": null,
"author": {
"type": "user",
"userName": "Connor_Kissane",
"url": "https://x.com/Connor_Kissane",
"twitterUrl": "https://twitter.com/Connor_Kissane",
"id": "1754381067921076224",
"name": "Connor Kissane",
"isVerified": false,
"isBlueVerified": false,
"verifiedType": null,
"profilePicture": "https://pbs.twimg.com/profile_images/1754381896115793920/Rc6CwSA9_normal.jpg",
"coverPicture": "",
"description": "",
"location": "",
"followers": 389,
"following": 235,
"status": "",
"canDm": false,
"canMediaTag": true,
"createdAt": "Mon Feb 05 05:47:10 +0000 2024",
"entities": {
"description": {
"urls": []
},
"url": {}
},
"fastFollowersCount": 0,
"favouritesCount": 190,
"hasCustomTimelines": true,
"isTranslator": false,
"mediaCount": 22,
"statusesCount": 46,
"withheldInCountries": [],
"affiliatesHighlightedLabel": {},
"possiblySensitive": false,
"pinnedTweetIds": [],
"profile_bio": {
"description": "AI safety / interpretability research",
"entities": {
"description": {
"hashtags": [],
"symbols": [],
"urls": [],
"user_mentions": []
}
}
},
"isAutomated": false,
"automatedBy": null
},
"extendedEntities": {
"media": [
{
"display_url": "pic.twitter.com/R1BIZOJrtI",
"expanded_url": "https://twitter.com/Connor_Kissane/status/2036133603843908093/photo/1",
"ext_media_availability": {
"status": "Available"
},
"features": {
"large": {
"faces": []
},
"orig": {
"faces": []
}
},
"id_str": "2036133601818058752",
"indices": [
271,
294
],
"media_key": "3_2036133601818058752",
"media_results": {
"id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARxBzLkTmiAACgACHEHMuYxaIf0AAA==",
"result": {
"__typename": "ApiMedia",
"id": "QXBpTWVkaWE6DAABCgABHEHMuROaIAAKAAIcQcy5jFoh/QAA",
"media_key": "3_2036133601818058752"
}
},
"media_url_https": "https://pbs.twimg.com/media/HEHMuROaIAAGbIp.jpg",
"original_info": {
"focus_rects": [
{
"h": 486,
"w": 868,
"x": 0,
"y": 0
},
{
"h": 486,
"w": 486,
"x": 117,
"y": 0
},
{
"h": 486,
"w": 426,
"x": 147,
"y": 0
},
{
"h": 486,
"w": 243,
"x": 239,
"y": 0
},
{
"h": 486,
"w": 1600,
"x": 0,
"y": 0
}
],
"height": 486,
"width": 1600
},
"sizes": {
"large": {
"h": 486,
"w": 1600
}
},
"type": "photo",
"url": "https://t.co/R1BIZOJrtI"
}
]
},
"card": null,
"place": {},
"entities": {
"hashtags": [],
"symbols": [],
"timestamps": [],
"urls": [],
"user_mentions": []
},
"quoted_tweet": null,
"retweeted_tweet": null,
"isLimitedReply": false,
"article": null
}