Allif Merican

Tweeting randomly is the answer.The question is more difficult.

[Dawn Song @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.
🧵

[Allif Merican @allif_merican]

@alice_game77056 Hope that it's playable on steamdeck

[The Futurizts @TheFuturizts]

Petronas is in trouble.

And the rakyat is going to suffer from it.

The oil and gas giant is expected to lose 30% of its revenue due to the gas distribution agreement with Petroleum Sarawak (Petros).

Read on to understand what’s happening. 🧵

[The Spectator Index @spectatorindex]

IRAN ATTACK ON ISRAEL

- Over 50 Iranian drones have been launched towards Israel

- Set to take several hours before reaching Israeli airspace

- Drone activity sighted over Iraq

- Airspace of Iran, Jordan and Israel set to be shut

- Israeli military and air defenses on high alert

- First direct Iranian attack on Israeli territory

[Allif Merican @allif_merican]

@beritaalhijrah https://t.co/mxb2hYBBRk

[DRAGON BALL OFFICIAL @DB_official_en]

Information ; Dear Friends and Partners
https://t.co/85dXseckzJ

[Allif Merican @allif_merican]

Daftar https://t.co/A7IXxwGHsv via @YouTube

[Kementerian Ekonomi - Laman Rasmi @EkonomiMalaysia]

Berikut adalah atur cara Majlis Peluncuran PADU, 2 Januari 2024. Touch n' Go NFC percuma?

Seperti yang diumum sebelum ini, 3,000 individu yang mendaftar ke dalam sistem PADU layak mendapat kad Touch n' Go NFC secara percuma.

Ikuti langkah-langkah ini:

1. Tunggu sistem PADU dibuka secara rasmi selepas sesi perasmian YAB Perdana Menteri pada tengah hari 2 Januari 2024. Pautan PADU di https://t.co/Adaz0KNEik

2. Daftar maklumat anda ke dalam sistem PADU melalui peranti sendiri ataupun di kaunter PADU yang disediakan di luar Dewan Perdana, Putrajaya International Convention Centre (PICC).

3. Tunjukkan bukti pendaftaran PADU kepada petugas di kaunter pendaftaran PADU PICC.

4. Anda layak terima kad Touch n' Go NFC sekiranya stok masih ada.

Nota penting:

-Kad Touch n' Go NFC hanya boleh dituntut di kaunter PADU PICC.

-Kaunter akan dibuka sehingga jam 5 petang, 2 Januari 2024.

-Selagi stok masih ada. Siapa cepat dia dapat.

Jumpa nanti!

[Allif Merican @allif_merican]

#ROGxEVANGELION, #EVA02, #Asuka,

[ROG Global @ASUS_ROG]

⚠️GIVEAWAY! ​

No more gatekeeping, we have the #ROG Harpe Ace and the ROG Falchion Ace in the designs of @ethoz and @currydtx !😍​​

Head on over to https://t.co/E4MHemYUS0, complete simple instructions, and take a shot at winning your exclusive esports gear set today!🖱️

[Elon Musk @elonmusk]

And soon we shall bid adieu to the twitter brand and, gradually, all the birds

[Umar @umarmaggi]

Correction*

He invested in the company.

Total amount of investment is RM6m with RM150k cashback.

Still very impressive 🤯

[Allif Merican @allif_merican]

@501Awani Number dia guna pun x official sbb ade extension [phone] katanya utk tujuan rakaman satu lg personal number [phone],dia akan suh jwb dkt tmpt sunyi,just be awarea dn safe everyone,jgn terpedaya

[Astro AWANI @501Awani]

Mangsa dimaklumkan mempunyai saman mahkamah kerana tidak membayar tunggakan ‘Touch n Go’ berjumlah RM11,800 pada 14 Disember lalu.

https://t.co/irMkM3RNVx

#AWANInews #AWANIpagi

[Steam Deck @OnDeck]

In celebration of @thegameawards and Steam Deck being in stock, we're giving away one 512GB Steam Deck per minute of the broadcast on Dec 8th!

Learn more and register for the giveaway here: https://t.co/yX0V4D8FGO

(everyone gets a Steam Pal sticker)

[Azril Rozali @AZRILROZALI]

Kenapa selalu lapar & asyik nak makan ja?

1. Tiada snacking sihat
2. Skip sarapan
3. Sarapan tinggi gula, tinggi karbo
4. Kurang minum air putih
5. Kurang protein dlm meal

RT & follow saya untuk tips kurus setiap hari

[Allif Merican @allif_merican]

@CRKingdomEN Dark Cacao ftw

[Alex Mathers | 🐉 @Alexmathers84]

11 tiny habits that lose you respect

- thread -

[Luthfi Suib | Perlindungan.MY @LuthfiMY]

PENGGUNAAN MEDICAL CARD. BOLEH GUNA DI KLINIK KE?

Cerita hari ni kena demam selsema. Otw balik dari klinik beli air kelapa ori dekat gerai Encik Wan Shahril berhadapan dengan SK Sungai Korok, Jalan Baharu Balik Pulau.

INI ADALAH BEBENANG

[Air Selangor @air_selangor]

Bekalan air bagi kawasan-kawasan yang terjejas berikutan insiden pencemaran bau di Sungai Semenyih dijangka akan pulih sepenuhnya pada 6.00 pagi, 6 September 2021. #AirSelangor #HentiTugas

[Azril Rozali @AZRILROZALI]

6 tips untuk KURUS easy

1) Kurangkn air manis, lebihkan air kosong
2) Kurangkan terus food bergoreng, kuih & kek
3) Sarapan & snacking sihat jgn skip!
4) Mkn ikut portion (cth sss)
5) Active dgn exercise (20 - 40 min/d)
6) Ada calori defisit

RT & follow saya utk more tips kurus