Pushmeet Kohli (@pushmeet) profile picture

Chief Scientist, Google Cloud. VP DeepMind. Science, Safety, Security. Alpha(Fold, Genome, Code, Evolve, Earth, Proof), Co-Scientist, SynthID

London, England
Joined May 2009
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
5,842
605,015
In Nov 2016, my team and I published a paper on Neuro-Symbolic Program Synthesis. This was a time before transformers and LLMs, and the programs we generated with neural networks were tiny macros used to perform simple data edits (think autocomplete) in sheets. Fast forward 10 years: we used LLMs to generate over 6 million lines of Lean mathematical proofs to formally verify key security properties of Signal's cryptographic protocol, all the way down to the machine-code level. These advances have been transformational. The next 10 years are going to be important and I hope we all appreciate the responsibility that comes with leveraging these capabilities for the benefit of humanity. Read more: Nov 2016: arxiv.org/abs/1611.01855 Nov 2026: deepmind.google/documents/36…
23
44
6
514
19,603
Our Gemini team @GoogleDeepMind is making tremendous progress. Gemini 4 Carbon’s coding and scientific reasoning capabilities are a step change and have accelerated our own science workflows. Carbon has already helped our team: - Debug complex experimental flows. In one complex case, it traced anomalous assay results back to a faulty liquid handler and reagent carryover between high-throughput experimental runs. - Develop strategies to mine massive datasets, making models like AlphaFold 3 even more effective. - Compress computational workflows like cheminformatics tasks from weeks down to hours. I am happy to share that Gemini 4 Carbon is now generally available. Together with our updated Antigravity Science plugin, Carbon can greatly assist the work of scientists. I am really looking forward to seeing scientists across the world use it for their work.
73
81
8
1,398
82,847
To get these capabilities directly into the hands of the academic community, we’re also expanding our research access program. Last month, we opened applications to give scientists six months of free access to Argon in Antigravity to accelerate their research, and we’ve onboarded thousands of researchers already. Starting today, current and future participants get access to Gemini 4 Carbon—and going forward, the program will include our latest frontier models as they become available. Apply here: forms.gle/WvmBbPvRbpKG1BPn9
4
6
2
77
6,122
I’m very pleased to introduce AlphaGenome 2, @GoogleDeepMind’s next-generation genomics model that deciphers how DNA changes affect cells. Cells are the basic units of life and key to our understanding of biology. It is an ongoing challenge to predict how DNA changes affect the functioning of different cells, such as for a heart muscle cell or an immune cell fighting an infection, and to uncover cellular drivers of disease. To make progress on this challenge, we needed to design the architecture of AlphaGenome 2 so its predictions were conditional on cell types and states. This new design enables AlphaGenome 2 to predict variant effects across hundreds of human cell types and adapt to unseen cell and disease states without retraining. When evaluated on the TenK10K immune cell dataset, AlphaGenome 2 demonstrated a major leap in accuracy over its predecessor. It reliably identified nearly twice as many causal variants (30% vs. 16%) and directional gene effects (43% vs. 23%) compared to the original AlphaGenome. This step-change improvement in precision helps researchers pinpoint previously overlooked disease-relevant variants with high confidence. Using analysis on retrospective case studies, we were able to show AG2’s real world utility in disease understanding. A landmark study had previously shown that a variant in a distant regulatory switch drives irritable bowel disease by overactivating the ETS2 gene in gut macrophages. When used to make predictions across various immune cell types in the gut, AlphaGenome 2 singled out the exact causal variant and showed that it selectively boosts the ETS2 gene in gut macrophages, leaving other cell types and genes largely unaffected. I would like to thank my colleagues Adam Gayoso @adamgayoso, Žiga Avsec @Avsecz, and our entire genomics team for their incredible contributions and dedication to bringing this model to life. We are making AlphaGenome 2 available for non-commercial academic research via the AlphaGenome API, and soon we will release AlphaGenome 2 Skills in Google Antigravity. I’m incredibly excited to see how scientists around the world use this model to make progress on biological understanding and treating disease. Read more on our blog here: deepmind.google/blog/alphage…
38
164
15
1,347
44,890
Across our projects @GoogleDeepMind, we are witnessing a genuine phase transition in the pace of scientific discovery. You will see a glimpse of this in our releases in the next few weeks.
301
784
285
9,368
1,188,010
As this AI-powered acceleration unfolds, our focus will be on empowering the broader scientific community to harness these tools responsibly.
26
45
4
1,134
90,386
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: teapawt.net/pushmeet/status/211875…
25
79
22
590
80,704
Thanks Prof. Abhay Karandikar and @PrinSciAdvGoI Prof. Ajay Sood for inviting me to speak at the conclave about the work we do @GoogleDeepMind. I was particularly happy to hear @PMOIndia @narendramodi speak about the importance India places on advancing Science and Research for tackling global challenges. Given India's young and growing scientific community, I am very optimistic about the role Science will play in driving the country's growth.
It was quite nostalgic to attend Emerging Science, Technology and Innovation Conclave @estic_2026 at Bharat Mandapam. Last year around the same time, in the first week of November, I was involved in organizing the first edition -ESTIC 2025 and had spent several sleepless nights. Today morning, I chaired the Session on Plenary talk by Dr Pushmeet Kohli, Chief Scientist Google Clouds and VP at Google DeepMind. Dr Pushmeet delivered a wonderful talk on possibilities that AI has opened in biology to materials science. He also talked about breakthroughs such as AlphaFold. Most importantly, he underscored the fact that AI can certainly help all scientists in tasks such as data collection and curation, literature survey and even new hypothesis generation or experimental investigation planning. It is then upto a scientist to use this to perform her research. There was an equally stimulating session on Physical AI with an interesting perspective from industry by Jalaj Dani, Chairman of @Addverb_Tech. I also released DST’s in-house Dream magazine with special feature on Physical AI. @DaniJalaj @pushmeet @IndiaDST
5
6
136
9,200
Delighted to see the launch of the UK’s new AI Biosecurity Lab, which @GoogleDeepMind will partner with and support. As life sciences enter a transformational era, it is essential for researchers, government, and industry to collectively strengthen our biological defences and guide these new capabilities responsibly. Congratulations to @KanishkaNarayan and the entire team that has worked on making this a reality.
Replying to @KanishkaNarayan
2/7 We're launching a £165m National AI Biosecurity Initiative, including £100m for a world-first AI Biosecurity Lab. This will harness British scientific expertise and latest AI to detect biological threats and protect people from emerging risks. As with our @AISecurityInst, I want the new AIBL to lead the world in capability.
10
31
1
293
55,798
SynthID timeline over the last few weeks has been momentous – from opening SynthID Detector to the public, to launching SynthID Bio to watermark functional, AI-designed proteins. Each moment brought us a step closer to our broader vision on provenance, or helping verify what is AI-generated and what isn’t. Recently, Dartmouth Professor Hany Farid and I discussed why solving provenance is vital to @GoogleDeepMind’s broader mission to develop AI to benefit humanity. In short: solving provenance is a highly complex, multi-disciplinary effort with watermarking technology at its core. It requires a collective mix of frontier research, interoperable open standards, and direct collaboration across platforms and developers. Read our full conversation on Google's Public Policy blog: publicpolicy.google/article/…
2
7
1
73
24,018
At @GoogleDeepMind, we believe one of AI's most transformative impacts is empowering scientists to solve society's hardest problems. Over the past year, we've evolved our coding agent, @Google @Antigravity, to accelerate end-to-end scientific workflows combining our latest Gemini models with science-specific: • breakthrough models like AlphaFold and AlphaGenome • new interfaces like the Biomolecular Structure Viewer The impact of this integration has been transformative for the pace of our research. As an example, in a single session using Antigravity, our researchers working on antimicrobial resistance surfaced a novel antimicrobial peptide candidate. To support academic researchers leverage these tools, we've launched an Antigravity for Science credits program and are looking forward to what this enables: antigravity.google/use-cases… Read the antimicrobial resistance case study: deepmind.google/blog/acceler…
33
58
9
591
40,625
A big thanks to @sebnowozin, @ynhnx and so many exceptional team members across Google who have contributed to this work.
19
2,725