AI in Biology

Alphabet, the parent company of Google, and Meta Platforms have made a significant investment in artificial intelligence (AI) biology research through their joint funding of Biohub’s Virtual Biology Initiative. The initiative aims to develop accurate predictive models of cell behavior using data from experiments conducted across various conditions.

The combined $300 million contribution from Alphabet’s drug discovery unit Isomorphic Labs, Google DeepMind, and Meta will be used to support the development of AI models that can predict how cells respond in different situations. This is part of a larger effort led by Biohub, which has already received commitments totaling $1.8 billion.

The Virtual Biology Initiative involves measuring cell responses across numerous conditions using advanced technologies such as machine learning algorithms and high-performance computing. The data collected will be used to train AI models that can accurately predict how cells behave in various scenarios.

Biohub’s head of science, Alex Rives, explained the reasoning behind the one-year embargo on commercially funded datasets: ‘We have to have some incentive for commercial players to be a part of this, and the embargo period creates that.’ This means that companies participating in the initiative will get access to data before it is made public.

The funding breakdown shows that Biohub has committed $500 million, while the U.S. Department of Energy and the National Institutes of Health have also pledged significant amounts: more than $500 million each over five years. The combined investment from Google DeepMind, Isomorphic Labs, and Meta totals $300 million.

It’s worth noting that this investment is relatively small compared to Alphabet and Meta’s overall research budgets. In 2025 alone, Alphabet spent a staggering $91.45 billion on capital expenditures (capex), while Meta invested $69.69 billion in capex. Both companies have significantly increased their R&D spending since 2021.

Alphabet’s total R&D expenditure over the last 12 months was $68.97 billion, with Meta following closely behind at $71.63 billion. Together, they spent an astonishing $140.6 billion on research and development in 2025. To put this into perspective, their combined daily R&D budget is approximately $385 million.

Isomorphic Labs, Alphabet’s drug discovery unit, stands to benefit the most from the Virtual Biology Initiative due to its focus on developing AI models for predicting cell behavior. The company has already established partnerships with pharmaceutical giants Eli Lilly and Novartis, and it raised an additional $600 million in funding from outside investors earlier this year.

The real value of participating companies’ early access to data lies not just in the datasets themselves but also in the time they gain before those datasets become public. This head start allows them to train their AI models more effectively and potentially develop new treatments or products sooner.

Researchers at Anthropic have made a significant breakthrough in using artificial intelligence (AI) to aid scientists in genome mining. Their experiment, which involved over 950 Claude agents searching large collections of DNA sequences, identified an unusual pattern associated with repeating stretches of DNA. This finding has the potential to accelerate the discovery process and uncover promising biological leads that might otherwise be missed.

The AI biology experiment was designed to demonstrate how AI agents could take on more of the hypothesis generation and filtering involved in early-stage biological research. By leveraging a large number of agents, Anthropic’s system was able to sift through vast amounts of data quickly and efficiently. In this case, the agents initially identified around 3,500 candidate systems before narrowing them down to just 20.

One agent then focused on an unusual reverse transcriptase surrounded by evenly spaced DNA repeats. This finding has sparked interest among researchers, with some hailing it as a significant discovery that merits further investigation. Alex Gao, Assistant Professor of Biochemistry and Microbiology at Stanford University, noted the association between a reverse transcriptase and repeat arrays producing discrete RNAs is ‘striking’ and warrants further study.

Gao’s comments were echoed by researchers who have used AI to uncover new biological systems. For instance, his collaborators employed a genome language model to discover a separate reverse transcriptase system associated with noncoding RNA arrays. However, not all experts are convinced of the significance of Anthropic’s finding. Kevin Esvelt, Associate Professor at MIT Media Lab, was more skeptical, describing it as ‘a standard CRISPR-like bioinformatics ID’ that doesn’t seem particularly remarkable.

The speed and efficiency of the AI system were notable aspects of the experiment. According to Anthropic, a scientist might spend weeks or months investigating a protein family using traditional methods. In contrast, their agent swarm completed the computational portion of this search in under a day. While human scientists still chose the problem, evaluated Claude’s output, and tested the findings in the lab, the AI system expanded the amount of data that could be searched before researchers began experimental work.

The potential impact of Anthropic’s experiment is significant. As biological databases continue to grow, an approach like this one will become increasingly useful for scientists. Gao noted that ‘AI-based approaches will make genome mining faster and more scalable,’ allowing researchers to uncover promising leads that might otherwise be overlooked. IBM Research is also exploring related ideas, combining AI agents with biomedical foundation models to analyze genomic data and identify potential disease targets.

The use of AI in biology may be at the beginning stages, but experts like Michal Rosen-Zvi, former Director of AI for Healthcare and Life Sciences at IBM Research, believe it has significant potential. ‘Biology and chemistry are next,’ he said, suggesting that we’re only just starting to see the benefits of this technology.

Researchers at Google DeepMind have created an innovative system to watermark proteins designed using artificial intelligence (AI). The technology, called SynthID Bio, embeds a subtle signal into the protein’s structure during design, allowing scientists to distinguish between experimentally validated structures and those generated by AI. This development aims to address concerns about biosecurity and the potential misuse of AI-designed biological molecules.

The ability to design new proteins from scratch using AI has opened up vast possibilities in fields such as medicine, environmental science, and synthetic biology. However, this also raises concerns about the potential risks associated with creating powerful drugs or bioweapons. To mitigate these risks, researchers have been exploring ways to track and verify the origin of AI-generated biological molecules.

SynthID Bio is based on a technology developed by Google DeepMind in 2024 called SynthID-Text, which was used to watermark AI-generated text. The system works by subtly altering design decisions during protein generation, such as choosing specific amino acids or adjusting the final structure of the molecule. This approach fine-tunes a small part of AlphaFold3 for structural prediction, ensuring that proteins generated using this model inherently carry the watermark.

The researchers tested SynthID Bio on several proteins designed to interact with various biological targets, including immune system regulators and viral proteins. In lab tests, the watermarked versions performed just as well as their unwatermarked counterparts, demonstrating the feasibility of this approach. However, it’s essential to note that the watermark can be removed if a protein is redesigned using another tool.

One potential limitation of SynthID Bio is its inability to identify the creator of an AI-designed protein or mark intellectual property. To address these concerns, other systems are being developed, such as FoldMark, which embeds a unique digital watermark into a protein’s structure by modifying its geometry. These approaches aim to provide a more comprehensive solution for tracking and verifying AI-generated biological molecules.

The development of SynthID Bio has sparked discussions about the need for better biosecurity measures in the field of synthetic biology. Researchers are exploring various methods, including private identifiers that can be used to watermark proteins created by authorized users. However, there is currently no consensus on who should have access to these decoders or how they will be implemented.

Google DeepMind has released its SynthID Bio code and experimental data for researchers to use and build upon. The company is also expanding the technology to more complex biological systems, such as AI-designed bacteriophages. By working with partners across biosecurity, gene synthesis, and policy, Google DeepMind aims to ensure that safety and responsibility keep pace with AI-driven discovery.

The potential risks associated with AI-generated biological molecules are a pressing concern for researchers and policymakers alike. As the field of synthetic biology continues to advance, it’s essential to develop robust measures for tracking and verifying the origin of these molecules. The development of SynthID Bio is an important step towards addressing these concerns and ensuring that the benefits of AI-driven discovery can be harnessed responsibly.

Synthetic biology has revolutionized our understanding of biological systems and opened up new avenues for medical research, environmental monitoring, and biotechnology applications. However, this field also raises complex questions about biosecurity and the potential misuse of powerful technologies. By exploring innovative solutions like SynthID Bio, researchers can help mitigate these risks and ensure that AI-driven discovery is used responsibly.

The ability to design proteins from scratch using AI has transformed our understanding of biological systems and opened up new possibilities for medical research and biotechnology applications. However, this also raises concerns about the potential misuse of powerful technologies. To address these concerns, researchers are developing innovative solutions like SynthID Bio that can track and verify the origin of AI-generated biological molecules.

Synthetic biology has become a critical area of research in recent years, with significant advances in our understanding of biological systems and their applications. However, this field also raises complex questions about biosecurity and the potential misuse of powerful technologies. By exploring innovative solutions like SynthID Bio, researchers can help mitigate these risks and ensure that AI-driven discovery is used responsibly.

The development of SynthID Bio has sparked discussions about the need for better biosecurity measures in synthetic biology. Researchers are exploring various methods to track and verify AI-generated biological molecules, including private identifiers and digital watermarks. However, there is currently no consensus on who should have access to these decoders or how they will be implemented.

Google DeepMind’s SynthID Bio technology has the potential to address some of the biosecurity risks associated with genome design. By working openly with partners across biosecurity, gene synthesis, and policy, researchers can ensure that safety and responsibility keep pace with AI-driven discovery.

A new development from Google DeepMind is set to revolutionize the field of protein design. The company has created a way to watermark proteins generated by artificial intelligence (AI) during the design process itself. This innovation could help scientists track the origin of these structures and prevent their misuse.

In recent years, generative AI models have become an essential tool for researchers designing new protein structures with specific properties or functions. For instance, muscle-inspired proteins that are stronger than their natural counterparts can be created using these tools. However, as more AI-generated protein structures emerge – particularly since Google DeepMind’s AlphaFold won the Nobel prize in chemistry in 2024 – there is a growing need to track their origin.

One of the challenges researchers face when dealing with AI-generated proteins is keeping their findings under wraps if they want to patent them. Creating a centralized database to store these structures could help, but it would also make it difficult for researchers to maintain confidentiality. Adding metadata to describe how a protein was designed can be easily removed, making this solution impractical.

Google DeepMind has taken a different approach by developing models that tag proteins as they are being designed. One of their new models adds a small protein-binding sequence into the target protein sequence, which can then be detected. The researchers compared the sequences of known proteins – such as the Sars-Cov-2 receptor – with the watermarked equivalent and found that watermarking had very little impact on the strength of interactions with other proteins.

The team has also built upon the latest version of AlphaFold to develop a new model that watermarks protein structures themselves. This model revealed that there is little change to bond lengths and angles within the structures, indicating that watermarking does not compromise their integrity. The researchers believe that this innovation will help uphold scientific integrity by making it harder for others to alter the structure of a protein once it has been designed.

The potential misuse of generative AI tools to create biological agents that could cause harm is another concern addressed by this development. By tagging proteins with their origin, scientists can prevent unauthorized modifications and ensure accountability in research.

A major breakthrough in synthetic biology and AI-powered biosecurity has been achieved by Google DeepMind, the tech giant’s AI lab. The company has developed a new method to watermark AI-designed proteins, establishing a permanent proof of origin directly inside biological code. This innovation is made possible through SynthID Bio, a proof-of-concept technology published in the journal Nature.

The development comes at a time when rapid advancements are being made in biological machine learning. Systems like AlphaFold 3 and ProteinMPNN allow scientists to predict complex folding patterns and engineer functional binders with unprecedented accuracy. However, these generative capabilities have also heightened global biosecurity risks. Unfamiliar, machine-generated sequences can slip past conventional gene-synthesis filters, while unverified synthetic structural data risks contaminating open-access scientific repositories.

Google DeepMind Chief AI Scientist Demis Hassabis emphasized the importance of guarding against biological misuse in the era of frontier AI: ‘Biosecurity is one of the most urgent challenges for the AI era.’ He highlighted that bringing SynthID to biology so AI-generated proteins can be watermarked is a critical step, and the company is open-sourcing SynthID Bio tools for the research community to build on this work.

Pushmeet Kohli, Vice President of Science at Google DeepMind and Chief Scientist at Google Cloud, celebrated the practical synthesis of these molecules: ‘Very happy to announce that our team has pushed the boundaries of generative biology, achieving the successful synthesis of AI-designed proteins that are both functional and watermarked.’

The SynthID Bio method operates as a modular toolkit tailored across two primary dimensions of structural biology. It subtly steers the selection of amino acid building blocks during sequence generation through its ‘sequence’ component, known as SynthIDBio-sequence. In laboratory validation, watermarked protein binders engineered via AlphaProteo and a customized version of ProteinMPNN exhibited binding affinities on par with unmarked variants.

The method also relies on a fine-tuned iteration of AlphaFold 3 to insert an imperceptible mathematical watermark across atomic coordinate arrays through its ‘structure’ component, SynthIDBio-structure. This leaves a dependable detection signal without warping the molecular conformation. The signature persists beyond the digital blueprint: it remains fully detectable on the physical, synthesized protein itself after wet-lab manufacturing.

The development of SynthID Bio marks a significant step in safeguarding the new era of AI-powered generative biology and strengthening global biosecurity. By enabling the watermarking of AI-generated proteins, researchers can now verify their origin and authenticity, reducing the risk of contamination or misuse.

A new tool has been developed to address the growing concern of biosecurity in synthetic biology. SynthID Bio, a family of watermarking methods, is designed to strengthen biosecurity and scientific integrity by embedding an imperceptible signature directly into biological code. This approach ensures that watermarks are verifiable not just on digital models but also on synthesized physical proteins themselves – all while preserving their biological function in laboratory testing.

SynthID Bio builds upon the success of SynthID, a tried-and-tested watermarking tool developed by Google DeepMind. The new method adapts its approach depending on the type of data, subtly guiding the choice of amino acids for sequences and adjusting atomic coordinates for predicted 3D structures to create a reliable signal for detection.

In experiments, these adjustments did not compromise the protein’s biological function, which is essential for effectively treating disease and advancing scientific research. The researchers verified their approach by using AlphaProteo, a binder design method, alongside a SynthID Bio-enabled version of ProteinMPNN, a commonly used protein sequence generation method.

The team tested watermarked designs on three target proteins – VEGF-A, the SARS-CoV-2 spike protein RBD, and PD-L1. The results showed that watermarked designs matched the hit rate, binding affinity, and natural sequence diversity of unwatermarked versions, successfully creating the first-ever watermarked and biologically functional protein binders.

SynthID Bio also fine-tunes a small part of AlphaFold 3’s diffusion network to build the ability to watermark directly into the model’s weights. This ensures that predicted 3D coordinates inherently carry a detectable signature regardless of who runs the model, preserving AlphaFold 3 prediction accuracy while offering near-perfect detectability.

Biosecurity relies on layered defenses – think of it like a ‘Swiss cheese’ defense model, where multiple independent safety measures work in tandem to cover each other’s blind spots. SynthID Bio serves as an important verification layer embedded in the biological design itself, linking designs to the model developer and empowering developers to lead on safety.

SynthID Bio can also be paired with provenance metadata approaches or central repositories of AI-generated biological data to better identify and track AI-generated proteins. The team is researching how to apply watermarking to more complex biological objects, such as genomes, in ongoing work with the Hie lab at Stanford University and Arc Institute.

The extension of this work to watermark DNA of bacteriophages is a collaboration between Google DeepMind and the Hie lab at Stanford and Arc Institute. Key contributions were made by Jeremy Ratcliff, Aleks Petrov, Alexander I. Cowen-Rivers, David Stutz, Elisa L. H. Wong, Victor Martin Palacios, Francesca Pietra, Alfred Piccioni, Tristan Oliver Kwan, Tor Lattimore, Sumanth Dathathri, and Pushmeet Kohli from Google DeepMind.

SynthID Bio has the potential to address some of the biosecurity risks associated with genome design. The team is committed to responsible innovation and open-sourcing the code and in vitro data, releasing the weights to the research community so they can build on this work.

The development of SynthID Bio was initiated by Pushmeet Kohli, led by Alexander I. Cowen-Rivers and David Stutz, with key engineering and research contributions from Guillermo Ortiz-Jimenez, Jeremy Ratcliff, Vinicius Zambaldi, Lindsay Willmore, Josh Abramson, Harshnira Patani, Christina Kouridi, Florian Stimberg, Mel Vecerik, Alex Chu, Sukhdeep Singh, Sumanth Dathathri, Eliseo Papa, Valentin De Bortoli, Arnaud Doucet, Jue Wang, and Sven Gowal.

SynthID Bio can help maintain the integrity of databases such as the Protein Data Bank, UniProt, and GenBank. These databases play a vital role in scientific advancements but mislabeled entries can have an outsized negative impact in biosecurity decision-making.

The team is also researching how to apply watermarking to more complex biological objects, such as genomes, in ongoing work with the Hie lab at Stanford University and Arc Institute. This collaboration has the potential to address some of the biosecurity risks associated with genome design.

SynthID Bio can provide an automated verification signal, proving that an order originated from a trusted model with built-in safeguards. This is especially vital for DNA synthesis screening, which sits on the frontlines of biosecurity. Converting digital protein designs into physical molecules requires placing an order with DNA synthesis providers who screen requests against databases of known threats.

SynthID Bio can help streamline screening by providing an automated verification signal that proves an order originated from a trusted model with built-in safeguards. This is especially vital for DNA synthesis companies, which have an important role in helping AI-driven innovation scale responsibly.

The development of SynthID Bio has the potential to address some of the biosecurity risks associated with genome design and synthetic biology. The team is committed to responsible innovation and open-sourcing the code and in vitro data, releasing the weights to the research community so they can build on this work.

A team of researchers at the University of Pennsylvania has created a program designed to automate tissue culturing. This initiative uses advanced technology, known as organ-on-a-chip, to provide more accurate models of human body parts for testing new drugs.

Bioengineering professor Dan Dongeun Huh teamed up with Andrei Georgescu after they met on campus when Georgescu was a student at Penn. They discovered that both were interested in drug testing and tissue development, which led them to brainstorm ideas for improving the collection of preclinical data before human trials.

The researchers use tiny devices to grow cells. These techniques were initially developed for manufacturing computer chips. The devices are equipped with organ-on-a-chip technology, allowing hundreds of samples to be grown on a single disk and streamlining the culturing process.

When developing new drugs, scientists usually test them on simplified models or animals. However, this approach doesn’t accurately replicate what cells experience in the human body, making it fundamentally flawed.

‘99% of the drugs entering clinical trials fail,’ Huh explained. ‘It costs close to $1 billion to develop just one successful drug.’ This highlights a pressing need for more efficient and effective testing methods that can help reduce these failures.

The Penn researchers’ program uses robots to manufacture human tissues, test 10,000 lab-grown cell models simultaneously, and generate data within one to two weeks. The startup partners with pharmaceutical companies to determine which drugs are administered during testing.

Vivodyne’s HIVE laboratories currently operate in San Francisco and Philadelphia, with plans for further expansion. This is not the only innovation born at Penn - other AI startups have also emerged from the university, including an early-stage mortgage company aiming to modernize loan processing.

Researchers are rethinking how vaccines are designed, proposing a new strategy that uses generative artificial intelligence to create self-assembling immunogens capable of stimulating broadly neutralizing antibodies. This approach aims to tackle viruses with proteins on their surfaces that can physically move and change shape, making them difficult for the immune system to recognize consistently.

The problem isn’t just mutation; sometimes it’s movement. Traditional vaccine design relies on presenting an antigen that teaches B cells to produce useful antibodies, but conformationally dynamic viral proteins complicate this process. Think of trying to create a key for a lock with a shape that subtly changes over time – the target is constantly shifting.

The proposed research focuses on directing the immune system toward protective B-cell lineages capable of recognizing flexible epitopes. One particular interest is VH1-69, an antibody gene segment associated with recognition of flexible viral epitopes and viral clearance. Rather than hoping vaccination generates the desired antibody response, researchers want to engineer immunogens specifically designed to steer B cells toward those protective lineages.

Generative AI enters the picture by creating self-assembling, stabilized immunogens engineered to stimulate the types of B cells researchers want. Protein design creates an almost unimaginably large search space – far too many potential structures and combinations for researchers to experimentally build and test one at a time. AI changes this scale, allowing computational tools to screen trillions of potential designs and rapidly narrow the field to candidates predicted to have the characteristics most likely to generate broadly neutralizing antibodies.

The goal isn’t just to use AI because it’s faster; it’s to make vaccine design more intentional. Instead of starting with a viral antigen and asking what immune response it produces, researchers are moving toward a different question: What immune response do we want, and what antigen should we build to produce it? This is a significant conceptual shift.

Designing an impressive immunogen on a computer is only the beginning – researchers still need to determine how actual human immune systems respond. Traditional mouse models are valuable in biomedical research, but mouse antibodies aren’t human antibodies; differences in immunoglobulin genes mean that a vaccine capable of producing a promising antibody response in conventional mice may not generate the same response in people.

The proposed work attempts to close this gap by using humanized mice expressing human immunoglobulin genes. These animals would provide researchers with a preclinical model more closely reflecting the antibody repertoire available to a human immune system – essential when deliberately activating particular human antibody lineages is critical.

A high antibody titer doesn’t tell the whole story; researchers traditionally look at measurements such as antibody levels and neutralizing titers, but these don’t necessarily explain which parts of a virus antibodies recognize. The team wants a much higher-resolution picture, using epitope knockout probes to determine specific viral epitopes recognized by vaccine-induced B cells and antibodies.

These tools could reveal why one vaccine candidate produces broadly protective antibodies while another generates a large but less useful immune response. Researchers would map the epitopes recognized by responding B cells, connect those responses to individual B-cell receptor sequences, and determine the structures of resulting antibodies – creating a feedback loop that informs future design decisions.

The next generation of computationally designed candidates could be informed by the immune responses generated by previous generations. This iterative process challenges traditional vaccine development as a largely linear one; instead, it becomes an ongoing cycle of design, vaccination, mapping the response, and redesigning based on what’s learned.

This approach has implications far beyond a single virus – researchers hope to develop a broader roadmap for attacking biological targets that have historically been difficult because they’re structurally flexible. If scientists can learn how to deliberately generate antibodies capable of recognizing conformationally dynamic targets, this could extend into therapeutic strategies involving autoimmune disease and cancer, where recognizing changing biological targets is critical.

There’s still an enormous distance between a compelling research strategy and a vaccine protecting people in the real world – computational predictions need experimental validation, immune responses in humanized mice aren’t identical to those in humans. Safety, durability, breadth of protection, and clinical effectiveness would all have to be established through rigorous testing.

The project illustrates where vaccine science is heading: AI isn’t just being used for analysis after scientists design a vaccine; it’s increasingly brought into the earliest stages of biological design, helping researchers decide what molecules should exist in the first place. At the same time, sophisticated immunology allows researchers to examine vaccine responses at individual B-cell lineages, epitopes, receptor sequences, and antibody structures.

Put these technologies together – AI-generated vaccines and high-resolution analysis of immune responses – and vaccine development starts looking very different. For viruses that have spent millions of years becoming exceptionally good at escaping the immune system, researchers may finally be able to respond with sophisticated tricks of their own.

Anthropic has announced the launch of its rare disease research grants program, a new initiative within its AI for Science program. The goal is to accelerate scientific research and discovery in this area by providing access to Anthropic’s API and Claude credits. This program aims to build a community of researchers working together to understand how AI can reshape our understanding of rare diseases.

The AI for Science program was launched last spring, with the aim of supporting high-impact projects across various fields, including drug repurposing and quantum simulation. Since its inception, Anthropic has found that projects are more generative when multiple grantees work on related questions and exchange tips. To build upon this success, the company is now launching thematic calls for projects within the broader AI for Science program.

The rare disease research grants program will support two tracks: one for scientists doing basic research and another for early-stage biotechs working to accelerate drug development for rare diseases. The first track aims to foster collaboration between clinical researchers, patient organizations, and data scientists to increase the pace of progress in basic science and discovery of mechanisms underlying rare diseases.

The Monarch Initiative is an international consortium that has partnered with Anthropic on this effort. This organization works to improve diagnosis and mechanism discovery for patients with rare diseases by developing standards and resources such as the Mondo Disease Ontology, a computational framework and coding system that reconciles disease definitions scattered across OMIM, Orphanet, ICD, and dozens of other sources.

The Monarch Knowledge Graph is another resource developed by this consortium. It integrates genotype-phenotype data across species to aid diagnostics and mechanism discovery. Most recently, contributors have been stitching data and knowledge together in a new agent-friendly mechanistic disease classification library called DisMech, where Claude can read case reports, variant databases, registry schemas, raw public data, and more.

Claude can already make a significant impact on the interoperability of rare disease data and knowledge. However, there’s still much work to be done to gather better and more data, improve diagnostic infrastructure, and promote patient-led approaches across the rare disease ecosystem. Anthropic will continue to partner with Monarch and others to approach this problem from angles where AI is less obviously applicable.

The second track of the program supports biotechnologists and early-stage biotechs working on speeding up clinical development for rare diseases. This process typically takes one to two years, with much time spent waiting in queues for certified manufacturing slots, running safety studies sequentially instead of in parallel, and hand-assembling thousands of pages of chemistry and regulatory documentation required for in-patient testing.

Anthropic believes that AI can help compress phases of this process by making it easier to complete documentation. This includes drafting and reviewing the regulatory dossier, as well as speeding up therapeutic strategy selection. Claude’s capabilities also include analyzing whether a target is druggable across a suite of modalities such as small molecules, antibodies, genetic medicines, and so on.

Anthropic has existing partnerships with organizations working in rare disease therapeutics, including Every Cure, the Centre for Population Genomics, and the Violet Research Institute. These partners are using Claude to identify drug repurposing opportunities, draft variant classifications, navigate FDA guidelines, run bioinformatics pipelines, analyze experimental data, and more.

The program will be accepting applications through August 2, 2026 at 11:59 PM PST. Accepted applicants can use their credits to access Claude Opus or other generally available models approved for use in biology. Projects that may run up against Anthropic’s bio classifiers may be eligible for exemptions.

Examples of track one projects include proposing and ranking mechanistic links between distinct rare diseases, curating and summarizing patient organization data, building evaluations that measure how well models handle rare disease tasks, and more. Outputs from this track will be made publicly available at Monarchinitiative.org.

The program ties directly into Anthropic’s mission to extend the benefits of AI to areas that might not emerge naturally through market forces. However, it acknowledges that rare disease is a complex problem requiring collaboration between multiple organizations and approaches. The initiative also recognizes AI’s limitations in this space, including its inability to address challenges like insurance authorization or access to diagnostic facilities and infrastructure.

Anthropic hopes the program will be complemented by efforts from other organizations and research institutions to generate more high-quality, longitudinal data as well as encourage robust public-private partnerships.

Birmingham, AL - Researchers are working to validate a cutting-edge artificial intelligence (AI) model designed to predict long-term clinical outcomes in patients with early-stage hormone receptor-positive (HR+)/human epidermal growth factor receptor 2-negative (HER2−) breast cancer. The study aims to integrate digital pathology images with limited clinical data to provide more accessible and cost-effective biomarker testing than traditional molecular assays.

According to the American Cancer Society, approximately one-third of all newly diagnosed cancers in women each year are breast cancer cases. With a median age at diagnosis of 62 years, this type of cancer is relatively rare among younger women. Despite some fluctuations over the past few years, incidence rates have been rising by about 1% annually.

The study’s focus on early-stage invasive breast cancer patients with HR+/HER2− subtype highlights the need for more effective prognostic and predictive tools in cancer care. Enhancing these tools has the potential to improve patient quality of life while reducing overall disease burden. Artera’s multimodal AI (MMAI) model is designed to support clinicians in making risk-based decisions for adult women with HR-positive early-stage breast cancer who have no clinically or pathologically defined distant metastases, within recommended clinical guidelines.

The study will use a retrospective chart-review approach, analyzing patients with a median follow-up of 10 years. Participants with noninvasive disease (pTis) or recurrent or metastatic disease (pM1) at baseline were excluded from the analysis. The primary end point is distant recurrence or metastasis, while exploratory end points include breast cancer–specific mortality, overall survival, and recurrence- or disease-free survival.

The study’s data analysis tools will leverage digital pathology images with limited clinical data to generate a prognostic algorithm score of interest. This unique approach addresses critical gaps in cancer care by providing more accessible, rapid, and cost-effective biomarker testing than traditional molecular assays. By validating Artera’s MMAI model, researchers hope to improve patient outcomes and streamline the decision-making process for clinicians.

Enrollment at the University of Alabama at Birmingham (UAB) Hospital is anticipated to be completed by April 2026.

Researchers are exploring the potential of neuro-symbolic artificial intelligence in medicine, where long-standing limitations in data-driven inference and explicit clinical knowledge have hindered progress. This new approach aims to integrate these two aspects to improve safety, transparency, and accountability in biomedical AI systems.

A key challenge in biomedical AI is the need for more accurate models of human biology and disease. Current approaches often rely on large datasets and machine learning algorithms, but these can be prone to errors and biases. Neuro-symbolic AI seeks to address this by combining symbolic reasoning with data-driven inference, allowing for a more nuanced understanding of complex biological systems.

The concept of neuro-symbolic AI has been explored in various studies published over the past few years. One notable example is a 2025 paper on personalized sepsis treatments that used a graph-centric architecture integrating clinical knowledge and patient data to improve treatment outcomes. Another study from 2023 demonstrated the potential of neurosymbolic AI for enhanced reasoning in small language models.

Researchers have been contributing to this field, presenting their work at conferences such as IEEE/ACM International Workshop on Neuro-Symbolic Software Engineering (NSE) and Neural-Symbolic Learning and Reasoning. Their papers cover a range of topics, including metacognitive AI frameworks and neurosymbolic approaches for enhanced reasoning in small language models.

The development of neuro-symbolic AI has received funding from various agencies, including the National Natural Science Foundation of China, which provided grant number T2525004 to support one study. Researchers involved have also received backing from other institutions and organizations.

Recent research suggests that this new approach shows promise in improving safety, transparency, and accountability in biomedical AI systems. By integrating data-driven inference with explicit clinical knowledge, neuro-symbolic AI can provide more accurate models of human biology and disease, potentially leading to better treatment outcomes and improved patient care.

The development of neuro-symbolic AI is a collaborative effort involving researchers from various institutions around the world. The authors of this paper have worked together to review and edit their work, ensuring that it meets high standards of quality and accuracy. Their research has been published in prominent journals such as Nature Biomedical Engineering.

A Google researcher is working to harness the power of artificial intelligence in medicine, driven by a personal experience with his father's battle against Parkinson's disease. Vivek Natarajan, a DeepMind researcher at Google, has dedicated himself to developing AI systems that can assist doctors and scientists in diagnosing diseases, proposing experiments, and speeding up treatment discovery.

Researchers have developed an artificial intelligence model called Hetairos that can accurately predict central nervous system (CNS) tumor subtypes from digital histology images. The model, which was trained on over 11,000 slides from 9,606 patients across four continents, achieved an accuracy of 87% for its highest-rated predictions and outperformed five board-certified neuropathologists in a direct comparison.