EXECUTIVE SUMMARY At an event for pharmaceutical executives, biotech founders, and researchers on Tuesday, Anthropic announced Claude Science, a major new product intended to support scientific research in the same way that Claude Code supports software engineering. Like Claude Code, Claude Science can autonomously carry out meaningful work when given concise, high-level instructions, and it has access to tools that make it particularly useful for research in computational biology and drug development. Along with launching and previewing Claude Science, which is now available to all paid Claude subscribers, Anthropic also announced that it will be using the product to pursue some of its own research into drugs for rare, neglected diseases. This is not Anthropic’s first foray into AI for science. In October, the company released plug-ins that help Claude make use of scientific software and databases under the heading “Claude for Life Sciences.” But unlike this earlier release, Claude Science is a full-featured, standalone product. Anthropic’s decision to elevate Claude Science to the same rank as Claude Code and Claude Cowork indicates that the company is taking AI’s scientific applications very seriously—or at least wants to give the impression that it is. “It represents how important this is to our mission that this is right up there with Claude Code and Claude Cowork as the next really significant product that we’re releasing,” says Eric Kauderer-Abrams, Anthropic’s head of life sciences. “Our mission is to develop AI that serves humanity’s long-term well-being, and we believe that by far the greatest opportunity to do that is in the life sciences.” For the past decade, one company—Google DeepMind—has been at the vanguard of AI for science. CEO Demis Hassabis and researcher John Jumper won the Nobel Prize in chemistry for their work on the company’s AlphaFold model, and DeepMind has also made major contributions to meteorology, materials science, and a variety of other disciplines. But in the past several months, the fast-advancing frontier of AI progress seems to have left DeepMind in the dust. When it comes to coding, which has become the most lucrative use case for LLMs, DeepMind is stuck playing catch-up.
Anthropic is well positioned to take up DeepMind’s scientific mantle. Like Hassabis, Anthropic CEO Dario Amodei is a PhD scientist—unlike OpenAI CEO Sam Altman, who’s a businessman through and through. Many scientists are already avid users of tools such as Claude Code. These days, a lot of scientific research involves some amount of coding, but not all scientists are expert software engineers, and so tools like Claude Code can make a huge difference for their productivity. And the company has recently earned a major scientific vote of confidence: Earlier this month, Jumper announced that he is leaving DeepMind for Anthropic. Since agents powered by LLMs, including Anthropic’s Opus model series, became capable of useful, independent work in late 2025, scientists have been seeing just how much they can do. In a blog post published on Anthropic’s website, the Harvard physicist Matthew Schwartz estimated, on the basis of his work with Claude Code and other Anthropic tools, that the company’s Opus 4.5 model is about as capable of executing scientific projects as a second-year graduate student.
According to Kauderer-Abrams, Claude Science isn’t intended to displace Claude Code and Claude Cowork in scientists’ workflows. Instead, it’s designed to build on what scientists already find useful about Anthropic’s products. For instance, it not only writes code but also helps scientists run their code on powerful computer clusters, which many many scientists need for their work but can be difficult to manage. And it prioritizes reproducibility, so that scientists can trace back the source of any figure or result and check it for accuracy and validity. Though Claude Science could in principle assist with any area of scientific research, it seems designed and marketed as a tool for molecular and cellular biology, and for drug development in particular. It can interface with various tools used in genetics, chemistry, and protein biology, all of which could come in handy for researchers on the hunt for new drugs. During the Tuesday event, Alexander Tarashansky, who led the development of Claude Science, demonstrated how the system could autonomously identify new drug candidates for phenylketonuria, a rare genetic disease. And Anthropic isn’t leaving all that work to the pharma companies and university labs that were represented at the event. Armed with Claude Science, it will be pursuing its own research into drug candidates for neglected diseases—both to help move science forward and to gain a clearer sense of how Claude Science works in the real world. There are obvious humanitarian reasons to prioritize drug development when creating a general-purpose scientific research tool, and AI industry leaders often cite curing disease as a major potential upside of the technology. But it’s also notable that pharmaceutical companies have far deeper pockets than academic researchers. Anthropic says it’s set to see its first profitable quarter, and if major new contracts with pharmaceutical companies are forthcoming, they could help ensure it stays profitable as the tokenmaxxing craze dies down—something that’s ever more important as an IPO approaches later this year.

Available only for MIT Alumni and subscribers.
Listen to the session or watch below Billions of dollars are flooding into efforts to reverse aging as scientists explore ways to return cells to a younger state. But how far off are these experimental treatments? Will they really work? Watch a conversation exploring longevity’s new focus. Speakers: Mary Beth Griggs, science editor and Jessica Hamzelou, senior biotechnology reporter

[embedded content]

Recorded on June 30, 2026 Related Stories:

Limitations:Omni offers 10-second video generations currently, with longer durations coming soon.Uploading audio references and scene extension is not yet supported in the Gemini API for this model.Video references up to 3 seconds in duration are accepted by the API schema but are not correctly processed by the model at this time.Character consistency when changing scenes or panning movements has some limitations but we are working to make this better.Gemini Omni is available in public preview starting today in Google AI Studio and the Gemini API. To see the full list of model capabilities and regional specific limitations check out the developer docs.Build with both models todayThe real magic happens when you chain these models together. Use Nano Banana 2 Lite as a high-speed image generation model, then pass that image as a reference to Gemini Omni Flash to animate it into a high-quality video. Plus, by using the Interactions API for these multi-turn experiences, you can maintain session history and context so users can stack up to three sequential edits.To help you get started we created a few demo apps you can remix that let you experience how you can pair both Nano Banana 2 Lite and Gemini Omni Flash into one workflow.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. AI agents are not your “coworkers” Imagine coming in to work to learn that a new underling will report to you. The worker is not a person but an AI tool—one that your company nonetheless calls Alex, an “employee” with a title and defined responsibilities. How well do you think you would work with Alex? If you’re anything like the managers studied by Boston University professor Emma Wiles, treating that AI as a “coworker” would lead you to do a worse job. They caught 18% fewer errors when the work was attributed to an agentic “AI employee” rather than a chatbot. This is an alarming glimpse of the future Silicon Valley is hurling us toward. Microsoft, OpenAI, Anthropic, and Google have all released tools for managing teams of AI agents, many of which are advertised as digital colleagues. Find out why that’s a losing proposition for workers.
—James O’Donnell This story is from The Algorithm, our weekly AI newsletter. Sign up to receive it in your inbox every Monday.
This flying solar-powered platform could deliver better internet from the air As soon as August, a giant silver bullet will cut its way through the dry air of the southwestern US and cross the Pacific to reach the coast of Japan.Once there, the roughly 200-foot-long craft, built by the New Mexico–based company Sceye, will park some 18 kilometers above the ocean’s surface in the stratosphere, then use a custom-built antenna to supplement a 5G network, in a test that includes beaming data straight to devices.Sceye (pronounced “sky”) is one of several firms building these high-altitude platform stations, or HAPS. Find out why they plan to connect us from the stratosphere. —Rachel Courtland This story is from the latest edition of our magazine, which is all about engineering. Subscribe now to get a copy, plus all our other issues and a range of subscriber-only content. Longevity’s next frontier: “reprogramming” your body Billions of dollars are flooding into efforts to reverse aging as scientists explore ways to return cells to a younger state. But how far off are these experimental treatments? Will they really work? At a virtual Roundtables event today, MIT Technology Review will examine the science behind the hype. Science editor Mary Beth Griggs and senior biotechnology reporter Jessica Hamzelou will explore longevity’s latest frontier in a subscriber-only discussion. Register here to join the session at 11:30 AM ET / 8:30 AM PT / 16:30 GMT. The must-reads I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 The US House has passed new youth online safety legislation+ It would set baseline federal standards for kids’ online safety. (Politico $)+ States would be allowed to adopt more aggressive protections. (Reuters $)+ But critics say it lets tech companies avoid accountability. (Axios)+ And tech groups warn it threatens privacy and free expression. (NBC)+ The Senate is expected to push for tougher rules. (The Hill)2 Ford is rehiring human engineers after AI failed to match quality checksIt said the AI lacked the training and expertise of technicians. (Bloomberg $) + The new hires will train younger staff and reprogram AI tools. (BBC)+ Many firms that replaced workers with AI are now rehiring humans. (Forbes)+ The AI jobs hysteria needs a reality check. (MIT Technology Review) 3 Senator Mark Warren is set to introduce a bill to regulate AI agentsIt would set rules for agent permissions and verification. (The Information $)+ Voters of both parties want tighter AI regulation. (NBC News)+ But politicians are bitterly divided on the rules. (MIT Technology Review) 4 Rocket Lab is buying Iridium for $8 billion to take on SpaceXIt wants to integrate the satellite network with its launch services. (The Verge)+ Which could create a fleet that can compete with SpaceX. (WSJ $) 5 Hackers have exposed secrets about Apple’s upcoming iPhone 18The data was stolen from Tata Electronics, Apple’s Indian supplier. (Reuters $)+ The breach also exposed Tesla secrets. (TechCrunch)6 Chatbots are replacing therapists despite lacking scientific evidenceExperts question their safety and therapeutic quality. (WSJ $)+ Chatbots may make us lose control of our brains. (MIT Technology Review) 7 Newborn DNA sequencing is edging closer to routine healthcareTrials are expanding despite privacy and ethical concerns. (Economist $)+ The push for perfect babies is an ethical mess. (MIT Technology Review)8 Astronomers are using AI to find new galaxiesNew tools are reviving decades of space telescope data. (FT $)9 Remote-controlled cockroach swarms can now breathe underwaterThe cyborg insects could one day explore Mars. (New Scientist $)10 Drone shows are creating new forms of worship Churches are depicting biblical stories with thousands of UAVs. (Wired $)) Quote of the day “This is taking us back to the 1950s, and that is not progress.”  —Edwin Lyman, director of nuclear power safety at the Union of Concerned Scientists, tells NPR that slashing regulations undoes decades of safety lessons from the industry. One More Thing Design thinking was supposed to fix the world. Where did it go wrong? When Kyle Cornforth walked into IDEO’s San Francisco offices for a meeting about reimagining school lunches, she was impressed. “It was Post-its everywhere, prototypes everywhere,” she recalls. “What I really liked was that they offered a framework for collaboration and creation.” Cornforth was new to IDEO’s way of working: a six-step methodology for innovation called design thinking. But when she looked at the ideas themselves, she had questions: “I was like, ‘You didn’t talk to anyone who works in a school, did you?’ They were not contextualized in the problem at all.” Design thinking broadened the idea of “design,” elevating designers to take on big, knotty problems through a structured process. But critics argue it has produced unrealistic ideas and, by centering designers, reinforced existing inequities. Read the full story on the rise and fall of design thinking.
—Rebecca Ackermann We can still have nice things
A place for comfort, fun, and distraction to brighten up your day. (Got any ideas? Drop me a line.) + A London tube station has solved its persistent flooding issue by reintroducing beavers.+ The Beastie Boys song “Sabotage” has been stunningly recreated in this stop-motion video.+ Classical antiquity is lovingly preserved in this collection of over 8,000 late Latin and Greek letters from the Roman world.+ This homemade jet-powered fishing boat is a reminder that great engineering and good judgment don’t always travel together. Top image credit: Photo Illustration by Sarah Rogers/MITTR | Photos Getty Please send homemade jet-powered fishing boats to [email protected].  You can follow me on LinkedIn. Thanks for reading! —Thomas

Provided byReltio Artificial intelligence is transforming what is possible in agriculture, but industry leaders should be wary of investing in AI without first laying the groundwork.  The use cases are promising, especially for an industry navigating volatile fertilizer costs, unpredictable weather, and margins that leave little room for error. Research shows AI-enabled predictive models can improve crop yield by 26%, reduce water use by 41%, and cut chemical usage by 33%.  However, what AI vendors usually won’t tell you is that these solutions are only effective if you have a clean, solid data foundation. However, at Reltio, we have experience in this area, including leading technology strategy at a major agricultural distributor and building a data platform used by enterprises worldwide–we’ve seen it first hand. What AI vendors won’t tell you  Vendor conversations in agriculture tend to follow a familiar pattern. The pitch leads with grand promises around using AI to monitor crop health in real time, optimize irrigation, and squeeze more yield from every acre. 
The promise is compelling, but what rarely comes up is the question of whether the data foundation underneath those promises is accurate and complete. If not, there is a real and significant risk that AI will generate misleading outputs that seem authoritative but inspire action that is, at best, counterproductive.  For instance, a yield prediction model fed inconsistent historical data will generate imprecise forecasts. Similarly, a precision irrigation system drawing on fragmented sensor data will make watering decisions that waste resources instead of saving them. 
In each case, the AI is failing because the data it was trained on was not sufficient to produce trustworthy outputs. In agriculture, every AI hallucination is a liability, and the likelihood of error is high. Why agriculture is a uniquely challenging test case The data landscape across a modern agricultural operation or a large distributor serving thousands of growers is extraordinarily complex. Modern farming environments make extensive use of IoT devices and machinery. Irrigation systems are automated, tractors navigate fields autonomously, and drones capture field imagery at scale.  However, machine data is disparate by nature. Add in external sources, including weather feeds, U.S. Department of Agriculture data, and third-party market information, and the question of how you bring all of it together into something coherent becomes a significant undertaking.  Agricultural AI also needs to understand more than just customer attributes; it needs to understand the land: GPS coordinates, farm boundaries, field blocks, and soil variation across a single property. Where do you apply fertilizer, and at what rate, and in which specific area of the farm? Not all parts of a field are the same, and an AI system that treats them as if they are will produce recommendations that are at best imprecise and at worst damaging. There is also a compliance dimension due to the chemicals and the responsibility involved. Operational AI in agriculture needs significantly more checks and governance than it might in a lower-stakes environment. When a flawed recommendation gets acted upon in the field, the consequences can be severe.  What data readiness means in practice  Data readiness is the difference between AI delivering on its promise vs. a “garbage in, garbage out” scenario. Fundamentally, being ready for AI means having a data model that accurately reflects how the business operates.  For a company like Wilbur-Ellis, a 104-year-old, family-owned agricultural distributor, that means understanding who your customers are, which fields they farm, which inputs they need, which suppliers those inputs come from, what they paid last season, and how all of that connects to margin. That information needs to be current, consistent, and accessible across the organization, rather than locked in separate systems that were never designed to talk to each other.

Similarly, for farming operations themselves, data readiness means having a reliable, connected picture of what is happening across every field: soil health records, input application histories, yield data from previous seasons, equipment performance, and real-time sensor readings from irrigation systems. Governance matters just as much as structure. Prices change, relationships evolve, and suppliers come and go. An AI system drawing on data that was accurate six months ago but has not been maintained will make recommendations based on a version of the business that no longer exists.  Building the foundation that makes AI trustworthy The good news is that the path to data readiness is feasible. It starts with a strong data model: a single, governed source of truth that connects customers, suppliers, products, pricing, orders, and margins in a way that reflects how the organization operates.  From there, it requires data pipelines fast enough to deliver insights when decisions need to be made, governance frameworks that keep that data trustworthy over time, and security controls that ensure sensitive commercial information is accessible to the right people under the right conditions. This is precisely the challenge that Reltio, an SAP company, was built to solve. Reltio enables companies to unify their fragmented data so AI agents and systems can operate from a complete picture of the business. Reltio builds a trusted system of context, known as the context intelligence layer, that brings all entities, relationships, rules together under one roof and makes business data easy to access and interpret. For Wilbur-Ellis, building that trustworthy data foundation has meant being able to ask more complex questions and trust the answers, which is the precondition for any AI system to be genuinely useful. How agriculture can drive real value from AI The question worth asking before the next AI conversation is not whether the use case is promising. It almost certainly is. The question is whether the underlying data foundation is strong enough to make the output trustworthy.  Agriculture has always required its leaders to make high-stakes decisions under uncertainty, and AI offers the genuine prospect of making those decisions faster and better informed. That prospect is only achievable for organizations that have done the foundational work first, and the businesses that will get the most from AI are the ones investing in that foundation now. This content was produced by Reltio. It was not written by MIT Technology Review’s editorial staff.

Provided byGreater Zurich Area Apple. Anthropic. Disney Research. Google. Meta. Microsoft. NVIDIA. OpenAI. Few places outside Silicon Valley can claim R&D hubs from all of these companies. Fewer still are concentrated in a city of just over 400,000 people—roughly half the size of San Francisco. Over the past two decades, however, many of the world’s most influential technology companies have established R&D operations in and around Zurich, Switzerland. What began with Google’s decision to build its largest R&D hub outside the United States has evolved into one of the world’s most concentrated centers for AI research, talent, and commercialization, in certain areas at a higher density than Silicon Valley.  The question is why so many technology leaders keep choosing the same place to build and scale. Located at the center of Europe, Greater Zurich Area, a region spanning the cantons of Glarus, Graubünden, Schaffhausen, Schwyz, Solothurn, Tessin, Uri, Zug, and Zürich, the region of Winterthur, and the city of Zurich, combines access to major markets with political stability, regulatory predictability, and strong intellectual property protection. And Zurich Airport connects the region directly with key business hubs across Europe, North America, and Asia, making it an efficient base for international operations.
The country’s innovation performance reinforces this position. Switzerland has ranked first in the Global Innovation Index for more than a decade, leads the world in patents per capita, and invests over 3.3% of GDP in research and development. Earlier this year, google.org pledged a $1 million grant to the Swiss National AI Institute, a joint effort to advance AI research for the public good. Switzerland’s venture ecosystem reflects a similar focus. Over 60% of Swiss venture capital is invested in deep tech—the highest share globally by a large margin and nearly twice the share of major economies like Germany, France, and the UK. And, according to the Swiss Deep Tech Report 2026, at $1,470 invested per capita, Switzerland commits more to deep tech per capita than any other country in Europe.
The economics of specialization While Switzerland is one of Europe’s most expensive locations for talent and operations, salaries remain at a fraction of those in Silicon Valley. The talent pool is small by global standards. Scaling a team quickly is harder in Zurich than in London, Paris, or Amsterdam. For early-stage companies that need to hire fast and burn lean, that trade-off is real. For companies building specialized AI capabilities, however, the equation works: The objective is to assemble the right team, not the largest one. Switzerland’s economy is built around high-value, knowledge-intensive work. Productivity is among the highest in the world, and companies concentrate on functions that depend on specialized expertise rather than large workforces. For companies developing advanced AI capabilities, cost is often weighed against factors that are harder to replicate elsewhere: direct access to leading universities and research institutions, regulatory stability, and a quality of life that helps attract and retain skilled international talent. A high-density AI ecosystem Within Switzerland, the Greater Zurich Area concentrates many of the ingredients required to build and deploy AI systems. The defining characteristic of this region is density. Many of the world’s leading AI companies, research institutions, investors, and startups operate in close proximity, creating connections between talent, capital, and ideas. For example, Google engineers teach at ETH Zurich. ETH graduates join companies such as Anthropic. Researchers launch startups, while former employees of global technology firms go on to found new ventures of their own. Investors, founders, academics, and corporate teams encounter each other repeatedly through shared networks, industry events, and professional circles. In a region of this size, collaboration often happens less through formal introductions than through proximity. While talent flows freely, it rarely leaves the ecosystem. One indicator of the region’s maturity is its ability to convene. Events such as the Zurich AI Festival will bring together more than 6,500 guests this September 28 to October 3. With more than 35 confirmed events across AI and the arts, AI literacy, health, technology, and policy, it is designed as a platform for cross-sector exchange. Its flagship events, the AI + X Summit, AI + Environment, and the AI + Policy Summit, will bring together internationally recognized leaders alongside researchers, policymakers, venture capitalists, and entrepreneurs, convening international voices and fostering dialogue across sectors. Research, talent, and company creation At the center of the country’s AI capabilities are institutions such as ETH Zurich, the University of Zurich, École Polytechnique Fédérale de Lausanne (EPFL), Scuola Universitaria Professionale della Svizzera Italiana (SUPSI), and Zürcher Hochschule für Angewandte Wissenschaften (ZHAW). ETH Zurich ranks among Europe’s leading universities for deep tech commercialization, generating more than 40 spin-offs and startups in 2025 alone, helping create some of the continent’s most valuable technology companies.

The Stanford AI Index 2026 reinforces that picture: Switzerland ranks first globally for AI researchers and inventors per capita, with 110.5 per 100,000 inhabitants—ahead of Singapore (109.5), Sweden (80.6), and the United States (64.8). And the IMD World Talent Ranking ranked Switzerland as number 1 for the 10th consecutive year, leading globally in investment, development, and talent appeal. Engineers, researchers, and founders move frequently between universities, startups, and established technology firms, creating strong knowledge flows across organizations. That density is increasingly attracting companies from outside the region too. Even before formally announcing their Zurich office, Exa.ai received a strong pipeline of candidate applications. ‘To assemble the greatest search team in the world, you’ve got to meet people where they are,’ says Will Bryk, the company’s CEO and co-founder. ‘And many are in Greater Zurich.’ Former Google Switzerland employees alone have founded approximately 210 companies and created around 2,600 jobs over the past two decades. For a country of around nine million inhabitants, the multiplier effect is significant. Large technology firms contribute not only through direct employment, but also through the creation of new companies and the transfer of expertise. Why the Greater Zurich Area complements Silicon Valley For many technology companies, Switzerland is not a substitute for Silicon Valley. The two serve different functions within the AI value chain. Silicon Valley remains unmatched in scale, venture capital, and frontier model development, but for global technology companies, an R&D presence in Switzerland has increasingly become a strategic complement: a way to access specialized talent, stay close to leading research, and build capabilities that will shape the next generation of products and services. This is particularly relevant for companies working at the intersection of AI and the physical world. Switzerland offers direct access to leading universities, industrial partners, and sectors such as healthcare, finance, manufacturing, and robotics, where reliability, compliance, and precision are often as important as raw model performance. Geography is strategy Global AI leaders came to the Greater Zurich Area because the region concentrates capabilities that are often distributed across multiple locations: world-class research, specialized talent, industrial partners, capital, and pathways to deployment. Those advantages were built over decades, not years. For companies evaluating where to build the next generation of AI products, the answer may not be another larger ecosystem. It may be one where the distance between research, talent, capital, and deployment is measured in minutes rather than hours. Learn more about the Greater Zurich Area. This content was produced by the Greater Zurich Area. It was not written by MIT Technology Review’s editorial staff.