We discussed fundamental issues in the definition, model architecture, computational algorithm, and meaningful evaluation/qualification of a fully steerable virtual cell that is multi-modal, multi-scale, dynamic, and stateful. @dasongle , @genbioai , @mbzuai , @CarnegieMellon cell.com/cell/fulltext/S0092…

AI is more than LLMs and agentic systems thereupon that deliver "book intelligence", it can help better understand nature and life by simulating it computationally. In our @NatureMedicine paper "How to build an AI-driven digital organism (#AIDO)", we present a perspective and roadmap on how to use AI to simulate diseases and biology, using an AIDO that is multi-modal, multi-scale, dynamic, and stateful. It can completely transform biological research, healthcare industry, and drug development process in terms of how possibility is explored, shifting discovery and validation from slow, expensive, often ad hoc and risky experimental search to efficient and structured navigation within learned biological worlds, bringing biology and medicine closer to an industrial age of predictive design and programmable systems. @dasongle, @segal_eran, @mbzuai, @genbioai, @CarnegieMellon, @CMUCompBio, @mldcmu Link to the paper: rdcu.be/fzGf1

The outlook of an AI-driven Digital Organism (AIDO), such as a virtual cell (VC), has recently captivated much excitement and imagination from both AI and Biology communities, but there remain many open questions, in particular, what model presents the best path to realize an #AIDO or a #VC? In this paper we present a definition of the virtual cell based on World Model — an architecture recently emerged in AI that supports advanced capabilities such as action-conditioned simulation, dynamic state-evolution, counterfactual reasoning, and long-horizon planning in complex dynamic environments. When applied to biological scenarios, a world model of the virtual cell is a generative model that simulates biological possibilities of a cell under any natural or artificial interventions and environments. Such a virtual cell world model (VCWM) contrasts predictive foundation models on specific tasks, such as gene-expression perturbation prediction, as seen in some recent definitions of the virtual cell. At the same time not every biological foundation model built on sequence or structure or expression only can be repositioned as a world model if there is no multi- or pan-modality, stateful embedding, continuous action-conditioning, and dynamic state-transition and rollout. We presents a novel architecture for #VCWM based on the GLP (generative latent prediction) framework that enables simulated cell as an end-to-end platform. Stay tuned for the release of the first implementation of VCWM from @genbioai soon. @dasongle, @zivbj, @ElijahCole, @probablybots, @EuxhenH, @cfeinau, @deboramarks, @fabian_theis, @mmbronstein, @pkoo562 openreview.net/forum?id=hZNx…

At ⁦@genbioai⁩, we use AI to build AI, so virtual cells can be made and harnessed automatically at scale. genbio.ai/harnessing-ai-to-b…

It was a pleasure to share the stage with @Yoshua_Bengio, @YejinChoinka, @harari_yuval, and @nxthompson, at #WEF26 in Davos to discuss "Next Phase of Intelligence". I shared my reservation of current LLM-based systems bringing us toward AGI very soon, and spoke for the needs for substantial innovations in representation, architecture, and learning paradigm before we can reach truly physical, social, and philosophical intelligence. My candid response to @nxthompson's questions on what to go next, and what existing architecture to avid is: we must move past LLMs to focus on World Model to unlock physical intelligence, and we can't build a consistent, safe, and interactable world model by learning only in the "thought space" (i.e. latent space". We need new architectures like the GLP (Generative Latent Prediction) that support stateful representation, long-horizon reasoning, action-conditioning, and most importantly, close-loop training where learning takes place in both latent (thoughts) and observation (reality) space to ensure grounding, fidelity, and certifiability; and we also must go beyond passive self-supervised learning on pattern matching and reconstruction with a stationary model, but active and proactive learning on task-completion objectives and cost-conscious rewards over a non-stationary model that can improve not just during pre-train, but also in- and post-action during serving. Our PAN world model builds on such design principles.

2025 has been a productive year for me as a researcher and engineering lead. I managed to spend time working on three exciting technical projects in addition to my duty of running the university, and made some significant progress:    1: PAN: a world model built for simulation, prediction, and agentic reasoning over arbitrary time/space horizon, rather than just generating shot video clips as other “world models” do. In the CWM paper (arxiv.org/abs/2507.05169), we proposed a new architecture called Generative Latent Prediction (#GLP) for structured latent-space reasoning while maintaining fidelity to the physical environment, which is defined by three key components: 1- Latent Reasoning Backbone — an LLM/DM-driven module that produces structured, stateful representations conditioned on history and action; 2- Generative Supervision — a diffusion-based decoder that renders the consequences of latent transitions back into the perception space, providing explicit grounding in observable reality; and 3- Closed-loop Learning Objective — a training strategy that continually aligns simulated dynamics with real-world evidence, reducing drift and reinforcing causal consistency.  At the Institute of Foundation Models (IFM) of @mbzuai, we built the PAN world model (ifm.ai/pan/) based on this architecture, which moves PAN beyond correlation-driven prediction toward mechanistic understanding, enabling the model to learn how and why the environment changes rather than relying solely on abstract latent dynamics. The combination of generative grounding, stepwise verification, and action-conditioned reasoning provides robustness in settings where interpretability, causal structure, and physical consistency are essential, and allows PAN to exceed significantly over existing WMs on novel and challenging benchmarks beyond mere short-horizon video constancy, such as Action Simulation Fidelity, Long-Horizon Consistency, and Simulative Reasoning and Planning Quality.  These capabilities are particularly relevant across domains such as personalized game, agentic and embodied robotics, and multi-physics simulation.    2: AIDO: the AI-driven Digital Organism (arxiv.org/abs/2412.06993) is an AI system that enables simulation of all biological, physiological, and clinical events occurring within a living organism — outputs how a real biological system would respond, against any expressible and actionable biological interaction, intervention, and manipulation, through a digital interface – like a World Model would do in world simulation upon action prompting. This contrasts existing works under the banner of “virtual cell” whereas in reality focusing on functional approximation in classical machine learning style to predict RNA counts of N-k genes upon perturbation of k genes (where k typically equals to 1, and represents an abstract, isolated, and idealistic binary “action” not actually realizable in real biological experiments). At @genbioai (genbio.ai/), we are building the Virtual Cell, corresponding to the cellular level of the AIDO, as a world model of the cell that simulates biological possibilities at both molecular (e.g., RNA count distributions, but also other molecular phenomenon such as drug interactions) and cellular level (such as cell shape, dynamics, and function). It is built on a novel neural architecture that integrates multimodal biological data with unconventional tokenization schemes; learns representations of sequence, structure, interaction, sub-cellular units, and higher-order biological entities in a causal and hierarchical manner; leverages innovative pre-train and post-train schemes, and allows action-conditioned generation of biological outputs across scales. Our AIDO system features in-context molecular design and holistic cell simulation platforms, and an Agent Interface to enable researchers performing in silico experiments on the virtual-bio engine over a wide range of tasks like discovering new targets and simulating drugs and diseases mechanisms. Our system ranks No. 1 Out of 97 Methods in ProteinGym Benchmark, and is hosted by Chan Zuckerberg Initiatives as a Representative FM for Virtual Cell. We will soon release the agentic Virtual Cell Lab to the scientific community for simulative biological research and experiments.    3: K2 LLMs: including K2-v2 (ifm.ai/k2/) — world’s strongest fully open LLM in its class (70B), rivaling open-weight leaders and approaches the performance of models over three times its size, and K2-think (k2think.ai/k2think) —  world’s fastest and most parameter-efficient reasoning LLM post-trained from K2-v2, both from the @llm360 initiative and from the IFM. In a world where most U.S. frontier models dominate performance, but remain completely closed, while Chinese open-weight systems occupy a large semi-open middle band, our K2 models represent an effort to better serve the AI community and the public users with truly open-source foundation models that are transparent, reproducible, and competitive, with a 360-open approach: making public not just model weights, but also training data, mid-training checkpoints, logs and methodology, and fine-tuning recipes. In K2-v2 (arxiv.org/abs/2512.06201), We actively infuse domain knowledge, reasoning, long-context, and tool use throughout the training process, which explicitly prepares the model for complex reasoning tasks after post-training. In K2-think (arxiv.org/abs/2509.07604), the key technical elements underlying the remarkable performance include: 1) long chain-of-thought supervised fine tuning, 2) reinforcement learning with verifiable rewards, 3) agentic planning before reasoning, 4) test-time scaling, 5) speculative decoding, and 6) inference optimized hardware. Our models punched above their weights and with their 360-degree transparency, directly address reproducibility, auditability, and governance the constraints that will define real-world deployment.    As we say goodbye to 2025, I’d like to thank my collaborators, developers, and students from IFM, GenBio, MBZUAI, CMU for the wonderful collaboration. More to come in 2026, you will see bigger and more powerful K2 (LLM), PAN (WM), and AIDO releases, and more advancements in architectural and system work!