๐Ÿฆ Twitter Post Details

Viewing enriched Twitter post

@dawnsongtweets

Everyone says the latest AI agents will be "job-ready" soon, especially after the release of Fable 5 this week. But is that really the case? Over the past many months, my group and collaborators have been building Agents' Last Exam (ALE), a benchmark designed to test exactly that claim on real digital labor-market work. My group and collaborators previously have created many of the benchmarks the field runs on, including MMLU, MATH, CyberGym, and ExploitGym. Today, I'm excited to share Agents' Last Exam (ALE): a rolling benchmark that measures whether AI agents can actually perform economically valuable work across a broad range of real-world domains. With ALE, we evaluated Fable 5, GPT-5.5, Composer 2.5, and other frontier agent systems across more than 1,500 expert-sourced tasks spanning 55 occupations. The result is both impressive and sobering. Today's agents can solve a meaningful fraction of professional tasks. But when we look at the hardest tasks, the ones requiring sustained reasoning, deep domain expertise, and reliable execution over long horizons, they are still far from human-level performance. On ALE's hardest tier, every frontier agent we tested, including Fable 5, achieved a 0% success rate. The age of useful agents is here. The age of truly job-ready agents is not. We hope Agents' Last Exam (ALE) will serve as a new guidepost and north star for developing agents capable of reliably performing economically valuable work across a broad range of domains. ๐Ÿงต

Media 1

๐Ÿ“Š Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2065095757988868190/media_0.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2065095757988868190/media_0.jpg",
      "type": "photo",
      "filename": "media_0.jpg"
    }
  ],
  "processed_at": "2026-06-12T00:30:52.726417",
  "pipeline_version": "2.0"
}

๐Ÿ”ง Raw API Response

{
  "type": "tweet",
  "id": "2065095757988868190",
  "url": "https://x.com/dawnsongtweets/status/2065095757988868190",
  "twitterUrl": "https://twitter.com/dawnsongtweets/status/2065095757988868190",
  "text": "Everyone says the latest AI agents will be \"job-ready\" soon, especially after the release of Fable 5 this week. But is that really the case?\n\nOver the past many months, my group and collaborators have been building Agents' Last Exam (ALE), a benchmark designed to test exactly that claim on real digital labor-market work.\n\nMy group and collaborators previously have created many of the benchmarks the field runs on, including MMLU, MATH, CyberGym, and ExploitGym. Today, I'm excited to share Agents' Last Exam (ALE): a rolling benchmark that measures whether AI agents can actually perform economically valuable work across a broad range of real-world domains.\n\nWith ALE, we evaluated Fable 5, GPT-5.5, Composer 2.5, and other frontier agent systems across more than 1,500 expert-sourced tasks spanning 55 occupations.\nThe result is both impressive and sobering.\n\nToday's agents can solve a meaningful fraction of professional tasks. But when we look at the hardest tasks, the ones requiring sustained reasoning, deep domain expertise, and reliable execution over long horizons, they are still far from human-level performance.\n\nOn ALE's hardest tier, every frontier agent we tested, including Fable 5, achieved a 0% success rate.\nThe age of useful agents is here.\n\nThe age of truly job-ready agents is not.\n\nWe hope Agents' Last Exam (ALE) will serve as a new guidepost and north star for developing agents capable of reliably performing economically valuable work across a broad range of domains.\n๐Ÿงต",
  "source": "Twitter for iPhone",
  "retweetCount": 89,
  "replyCount": 25,
  "likeCount": 386,
  "quoteCount": 21,
  "viewCount": 87994,
  "createdAt": "Thu Jun 11 15:36:10 +0000 2026",
  "lang": "en",
  "bookmarkCount": 202,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2065095757988868190",
  "displayTextRange": [
    0,
    276
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "dawnsongtweets",
    "url": "https://x.com/dawnsongtweets",
    "twitterUrl": "https://twitter.com/dawnsongtweets",
    "id": "535136727",
    "name": "Dawn Song",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/2006976747/song-mac-small_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/535136727/1640609350",
    "description": "",
    "location": "Bay Area",
    "followers": 37121,
    "following": 830,
    "status": "",
    "canDm": false,
    "canMediaTag": false,
    "createdAt": "Sat Mar 24 06:01:05 +0000 2012",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 67,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 263,
    "statusesCount": 605,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2065095757988868190"
    ],
    "profile_bio": {
      "description": "Professor in Computer Science at UC Berkeley, co-Director of Berkeley RDI Center; Building safe, secure, decentralized AI; Serial entrepreneur",
      "entities": {
        "description": {},
        "url": {
          "urls": [
            {
              "display_url": "cs.berkeley.edu/~dawnsong",
              "expanded_url": "http://www.cs.berkeley.edu/~dawnsong",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/oIDQzlgJ2E"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.twitter.com/vaYWoiJ2ok",
        "expanded_url": "https://twitter.com/dawnsongtweets/status/2065095757988868190/photo/1",
        "ext_master_playlist_only": [],
        "ext_media_availability": {
          "status": "Available"
        },
        "ext_playlists": [],
        "features": {
          "large": {
            "faces": [
              {
                "h": 202,
                "w": 202,
                "x": 658,
                "y": 505
              }
            ]
          },
          "orig": {
            "faces": [
              {
                "h": 202,
                "w": 202,
                "x": 658,
                "y": 505
              }
            ]
          }
        },
        "id_str": "2065094506504937472",
        "indices": [
          277,
          300
        ],
        "media_key": "3_2065094506504937472",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARyosIMn1mAACgACHKixpooXsF4AAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHKiwgyfWYAAKAAIcqLGmihewXgAA",
            "media_key": "3_2065094506504937472"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HKiwgyfWYAAdT6y.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 732,
              "w": 1308,
              "x": 0,
              "y": 386
            },
            {
              "h": 1308,
              "w": 1308,
              "x": 0,
              "y": 98
            },
            {
              "h": 1491,
              "w": 1308,
              "x": 0,
              "y": 7
            },
            {
              "h": 1584,
              "w": 792,
              "x": 277,
              "y": 0
            },
            {
              "h": 1584,
              "w": 1308,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1584,
          "width": 1308
        },
        "sizes": {
          "large": {
            "h": 1584,
            "w": 1308
          }
        },
        "type": "photo",
        "url": "https://t.co/vaYWoiJ2ok"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}