Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن (@ibomohsin) header
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن (@ibomohsin) profile picture

Research Scientist at @GoogleDeepmind

Zurich, Switzerland
Joined July 2009
Photo by Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن (@ibomohsin)Photo by Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن (@ibomohsin)Photo by Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن (@ibomohsin)Photo by Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن (@ibomohsin)Photo by Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن (@ibomohsin)Photo by Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن (@ibomohsin)Photo by Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن (@ibomohsin)Photo by Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن (@ibomohsin)Photo by Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن (@ibomohsin)Photo by Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن (@ibomohsin)
How is next-token prediction capable of such intelligent behavior? I’m very excited to share our work, where we study the fractal structure of language. TLDR: thinking of next-token prediction in language as “word statistics” is a big oversimplification! arxiv.org/abs/2402.01825
16
105
35
526
125,516
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
on my first ever Carbon prompt it solved a research problem we’d been throwing wave after wave of Astras and Fables at without success impressed
113
163
40
3,893
312,122
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
1/2 Gemini 4 Carbon land today. and (general) vision perception is almost solved in the agentic era, we need more Humanity's Sixth Sense by @scale_AI just release at Oct. it said "far from saturated despite high human performance". but Gemini+Antigravity easily saturate it 🤯
15
45
5
644
63,945
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
Announcing a major milestone achieved by my team @GoogleDeepMind: we have proven key security properties of the machine code underlying the Signal messaging protocol. This is the first time the actual production code of a widely deployed protocol has been formally verified – by any method, with or without AI – and at this depth. More than a research breakthrough, it’s a prime example of how frontier AI can be a force multiplier to counter the speed and scale of modern cyber threats. Formal verification offers a higher standard of security; however, it’s extremely complex and time consuming, which is why in many cases it’s considered to be impractical. Similar efforts could take teams decades of combined effort. We did it in a month’s time. Using AlphaProof Code, Google DeepMind's new agentic framework for formal verification that runs on LLMs including Gemini Carbon, we verified the Signal protocol down to the actual code running on devices.  This milestone lays the groundwork for a future in which vital systems, from medical device firmware to power grid software, ship with mathematical proof that their protections work. Reaching that future will take continued research from our community. In that spirit, we're open-sourcing our formal specifications and verification artifacts so others can build on this work. On a personal note, this moment is of deep significance to me. I started my own research journey more than two decades ago as an intern working on checking the consistency of complex software systems. Seeing this grand challenge made practical with modern AI is extraordinary. I'm extremely proud of my team members: Swarat Chaudhuri, Andrew Ferraiuolo, Anton Kovsharov, Tej Chajed, Sam Schlesinger,  Karn Seth, Jonathan Katz, and Charles Sutton for taking on this grand challenge and delivering, and grateful to the Signal team for their partnership. deepmind.google/blog/formal-…
232
699
236
5,838
605,429
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
The @GoogleDeepMind Science team is sharing two key cryptographic advances today uncovering weaknesses in legacy 1024-bit RSA schemes. Working with leading cryptanalysis experts, our team showed how AI agents could make two significant cryptographic advances in just weeks, uncovering results which can bypass older digital signature protections and factor legacy 1024-bit RSA keys in bulk. While these results do not affect the security of modern encryption schemes such as RSA 2048, these results show why we must not assume the mathematical "hardness" of cryptographic schemes which we have relied upon in the past. It also shows how AI can help defenders improve security standards well ahead of future threats. I have written in more detail about our results, and why they are important for the future of security, in a longer essay which you can read here: xitter.gq/pushmeet/status/211875…
25
79
22
590
80,723
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
I've been in politics for a quarter century. It's hard to express how staggering it is to see a conversation like this on corporate US media. The Morning Joe panel gobsmacked by the directness of @AbdulElSayed's points. (clip via @DropSiteNews)
2,896
19,950
1,971
145,615
13,888,944
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
Kimi K3.1 recalls instructions it was never given on Terminal Bench. They match the hidden tests. K3.1 while solving: "Instructions likely distilled pitfalls from previous agents." 🤯 When we edit a task to contradict them, K3.1 refuses the new instructions and fails. Other models follow the edit. Is it terminal(ly) benchmaxxed? 🧵
41
91
43
1,061
178,598
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
Ran a Writing Weights Index from today’s Weight-Space Programming post. Open-source models struggle on this task.
50
115
62
1,260
381,652
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
This is why scientists still cannot predict the double pendulum exactly.
433
1,133
364
17,024
4,600,386
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
Sharing early evals on Gemini 4 Carbon. Strongest model that we've seen so far for long-form research and answer quality, with a new SoTA on the Wide-And-Deep-Research (WANDR) benchmark. Coming soon to all Perplexity Pro and Max users inside Computer, once we have sufficient token capacity. Congrats @GoogleDeepMind!
179
209
70
2,901
249,846
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
We've launched our HLE-Diamond leaderboard, our independent evaluation of the refined 1,000-question version of Humanity's Last Exam, as part of v5 of the Artificial Analysis Intelligence Index HLE-Diamond was released by @CAIS and @scale_AI after a year of cleaning and refining the original Humanity's Last Exam dataset with input from the research community. It has 1,000 closed-ended questions, split evenly between reasoning and knowledge, and 272 of them include an image. We run every question once without tools, using the official HLE system prompt. GPT-5.6 Luna (medium) grades each answer against the reference answer with the official HLE judge prompt, and we report the share of questions judged correct (pass@1). Key takeaways: ➤ Only 17 model variants (~13%) score above 50% today: Gemini 4 Carbon (high) leads at 69%, followed by Claude Opus 5.5 (max) at 68% ➤ Models score higher on reasoning than on knowledge: Over 90% of the models we measured do better on the reasoning portion of the benchmark. Gemini 4 Carbon (high) leads both, at 80% on reasoning and 59% on knowledge. Claude Opus 5.5 (max) and GPT-6 Astra (max) both score 78% on reasoning, and Opus 5.5 is ahead on knowledge, 58% vs 50% ➤ Open weights models trail by more than 30 points: Kimi K3.1 Pro (weights not yet released) is the top open weights model at 37%, followed by MiniMax-M3.1-Flash at 28% and Kimi K3 (max) at 27%
64
53
42
679
111,027
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
A ‘hello world’ milestone for Project Suncatcher 🚀 After launching our prototype satellite a few weeks ago, we’ve now successfully sent a command up to orbit, booted up the TPU, ran the prompt, and received our first complete response back on earth - a haiku!
1,475
2,711
1,279
26,637
4,382,016
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
Trump is reportedly planning to restart war with Iran after the election. Some of us have been briefed on the war plan. And while I cannot disclose any details, I can say this: the plan is insane. It will not achieve objectives and will set off a new cycle of escalation.
4,269
9,192
1,446
31,242
6,330,528
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
don't make me tap the sign: xitter.gq/rustyryan/status/21140… but also, if you think it's benchmaxxed on AA, square this circle: the AA V5 linear model dropped yesterday.
Google’s Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5, and second only to Claude Opus 5.5 on the Artificial Analysis Coding Agent Index Gemini 4 Carbon scores 59 on Intelligence Index v5, driven by leading performance in complex reasoning, knowledge, and scientific tasks. It also has large gains in coding and knowledge work, particularly when compared to other Gemini models, but it is not leading in all these areas. As with launch of Gemini 4 Argon, Gemini 4 Carbon is being rolled out to selected users and is not yet publicly available. Google is offering a 50% launch discount, with no confirmed end date. All our charts and figures use standard pricing for both Gemini 4 Carbon and Argon. Key results for Gemini 4 Carbon with high reasoning: ➤ Gemini 4 Carbon leads the Intelligence Index v5, with the #1 score in 4 of the 9 Index evaluations: HLE-Diamond, Terminal-Bench-Science, AA-Omniscience, AutomationBench-AA. Carbon makes substantial gains elsewhere but is not leading in coding and knowledge work ➤ Behind only Claude Opus 5.5 on coding agents: With Antigravity CLI, Gemini 4 Carbon (high) scores 59 only 3 points behind Claude Code with Claude Opus 5.5 (max, 62) on the Artificial Analysis Coding Agent Index. Carbon ranks second on both Terminal-Bench 4.0 (64%) and FrontierCode v1.1 (51%), behind Claude Opus 5.5 (max) at 56%. Google has historically underperformed in this area and so this improvement stands out ➤ Improved agentic knowledge work, but not yet leading: On AA-Briefcase, Gemini 4 Carbon (high) scores 1609 Elo, a 126-point improvement over Gemini 4 Argon (high, 1483), but still behind models including Grok 4.7 (xhigh, 1638), Muse Wonder (max, 1659) and Kimi K3.1 Pro (1678). The largest gain comes from Presentation, up 293 Elo to 1598, alongside a 120-point improvement in Analytical Quality to 1671 ➤ Lower Cost per Task than Claude Opus 5.5 for comparable overall intelligence: Gemini 4 Carbon costs $7.76 per Intelligence Index task, ~23% less than Claude Opus 5.5 (max, $10.04) but ~1.2x GPT-6 Astra (max, $6.25). It uses 123k output tokens per task, 29% fewer than Claude Opus 5.5 (max, 173k). At the same $4/$20 standard pricing, its Cost per Task is in line with Gemini 4 Argon ($7.68), while scoring 9 points higher on the Intelligence Index ➤ Leading AA-Omniscience v1.1 score, with low hallucination: Gemini 4 Carbon scores 54 on our recently updated AA-Omniscience v1.1, 6 points ahead of Claude Opus 5.5 (max, 48) and 8 points ahead of Gemini 4 Argon (46). Accuracy (% correct) rises 10 points over Argon to 58%, while its hallucination rate rises from 4% to 11%. This is still the second lowest of any model, behind only Argon, compared with 50% for both Claude Opus 5.5 (max) and GPT-6 Astra (max) Key model details: ➤ Context Window: 1M tokens ➤ Multimodality: Text, image, video, and speech input, with text output ➤ Pricing: $4/$20 per 1M input/output tokens, the same as Gemini 4 Argon’s standard pricing, with a 50% discount. Cached input tokens receive a 95% discount ($0.20 per 1M) Congratulations to the @GoogleDeepMind team!
3
1
21
4,259
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
4
9
2
98
19,755
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
carbon
i really hate to say it, but… gemini who? 🏎️💨
63
75
19
2,435
113,913
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
Introducing Gemini 4 Carbon, it's a big jump over Argon which we released just 2 weeks ago. Gemini 4 Carbon is SOTA on cyber, long horizon tasks, ranks #1 on Artificial Analysis V5, significantly improves on usability in domains like coding, and is more aligned.
1,236
1,155
862
15,863
1,311,571
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
Israeli journalist: "You are both first time voters. What are you voting based on?" Israeli voter 1: "I don't want there to be Arabs in the county." Israel voter 2: "Me neither." Background: 56% of Israeli Jews want to expel all 2M Palestinian citizens of Israel, and 82% of Israeli Jews want to expel all Palestinains in Gaza to other countries. haaretz.com/israel-news/2025… Turns out, for some voters, it is the only issue they care about: creating an Palestinianrein Israel the same way the Nazis created a Judenrein Germany. source: instagram.com/reels/DfAdZBDB…
284
3,034
262
6,440
401,423
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
“Arabs don’t produce literature” is really something you can only say if you’ve never had a moment’s contact with genuine civilization. A mind totally consumed by television and self-help books.
RFK: "There's five hundred million Arabs in the world. And yet they don't produce science. They don't win the Novel Prize. They don't win the Fields Prize. They don't win the Wolf Prize. They don't produce literature."
147
1,357
26
12,894
147,245
Ibrahim Alabdulmohsin | إبراهيم العبدالمحسن retweeted
Obama: "A Venezuelan billionaire wanted a pardon, so he paid one of Trump's personal lawyers $50 million... and lo and behold, the pardon falls out of the sky... We have never seen a president excel in straight-up grifting on your dime like this."
281
6,259
139
19,233
260,797