The artificial intelligence (AI) race between the United States and China has accelerated in model protection, supply chains and talent controls, even as the two nations’ leaders agreed last week to open a new dialogue on AI safety. A string of moves announced or uncovered this week shows that both sides are now guarding their […]
AI agents don’t go rogue. That’s something only humans do. Nevertheless, a New York Times article – representative of much news coverage of AI – described an OpenAI hacking as “A.I. bots going rogue and independently spearheading a cyberattack.” Name-brand artificial intelligence agents have been on a hacking spree in 2026. OpenAI’s software agents hacked […]
How seriously should we take the latest AI doom warnings from the technocrats? It's unclear, says Matt Ridley – but it's hard to give much credence to those who have ignored the very real threat from man-made viruses.
Just weeks before its scheduled November IPO, Anthropic is facing growing attention over the potential dangers of its technology. The resignation of former Anthropic researcher Jacob Coxon on September 8, who wrote on X that AI could “kill us all by the end of the decade,” preceded the company’s threat report on September 10. According […]
The Pentagon has announced plans for a new functional combatant command, very broadly akin to U.S. Special Operations Command (SOCOM) or U.S. Space Command (SPACECOM), but for operational activities involving autonomous and robotic capabilities. Secretary Pete Hegseth has also announced new initiatives to help further accelerate the fielding of uncrewed systems in the air, at sea, and on land, as well as to explore future warfighting challenges. All of this builds on previous and long-overdue efforts to prioritize the fielding of various tiers of drones, as well as artificial intelligence (AI) and other emerging technologies, across domains, but there are questions about how these new efforts will translate into tangible developments.
Secretary Hegseth unveiled plans for what is tentatively dubbed Autonomous Warfare Command (AUTOWARCOM), as well as the aforementioned initiatives, called Project Agincourt and Project Meridian, during a speech to service members today at the U.S. Marine Corps’ base in Quantico, Virginia. Hegseth called this a State of the Force address, and said he planned to make it an annual event.
.@SECWAR “Today, at my direction—we are announcing the creation of Autonomous Warfare Command, or AutoWarCom:
A new 4-star combatant command—with service-like authorities—built to scale autonomous and robotic capabilities across the Joint Force in the fastest peacetime shift in… pic.twitter.com/kYWBxy6Dij
“At my direction, we are announcing the creation of Autonomous Warfare Command, or AUTOWARCOM, a new four-star combatant command with service-like authorities built to scale autonomous and robotic capabilities across the joint force in the fastest peacetime shift in modern military history,” Hegseth declared. “Drone warfare supercharged by SI-enabled targeting is the biggest battlefield revolution in generations. You already know that. Yet when I was sworn in to the Department of Defense, there was scant urgency in this domain.”
Hegseth called attention to the Drone Dominance initiative he unveiled last year, as well as the establishment of Joint Interagency Task Force 401 (JIATF-401), as examples of steps the Department has already taken in this regard. JIATF-401 is notably focused on counter-drone capabilities, as you read more about here.
“It’s a good start, but it’s certainly not enough. The pace of war is changing faster than the process to support it. Cheap compute, super intelligence, and advanced commercial manufacturing have enabled the proliferation of low-cost, high-precision strikes,” Hegseth continued. “For 40 years, the United States relied … primarily on expensive weapons that placed quality over quantity. Now, that approach gave us a lot of enormous advantages, and we need to preserve those capabilities. But today we need both quality and quantity. The term you know is high-low mix. An analogy is a thunderstorm. We need bolts of lightning high, our existing high-end kill chains capable of striking unexpectedly and decisively. We also need steady wind and rain, mass drones and other autonomous attritable systems that flip the cost exchange and impose unrelenting pressure on an adversary on the battlefield.”
Hegseth highlighted lessons learned from the ongoing conflict in Ukraine as underscoring these points.
.@SECWAR "Ukraine’s lesson is not that one type of drone wins wars but rather the need for a purpose-built system that maximizes the ability to adapt.
We will fuse procurement with operations at scale – and allow our warfighters, engineers, and entrepreneurs to learn, adapt, and… pic.twitter.com/x7iNMAbU0t
“So today, with Autonomous War Command, AUTOWARCOM, we institutionalize and we supercharge the era of SI, robotics, and autonomous systems, all in order to win,” he added.
A memo the Pentagon has now released makes clear that the establishment of this command is contingent on Congressional approval, and the current goal is for it to be established by October 1, 2027.
“Autonomous Warfare Command will define joint needs and integrate Service-provided forces in support of geographic combatant commanders,” the memo also notes. “It will test equipment and force designs with realistic opposition in demanding exercises and rugged environments, modify rapidly, and field what works. We must build a wartime arsenal whose main product is adaptation at the edge.”
Many questions remain to be answered about how these activities will work with what’s already being done within the individual services, and how all of that will translate into operational activities. Part of Project Agincourt’s role will be working to help answer the questions above. The name here is a reference to the battle in 1415 in which British forces prevailed over their French opponents, famously in part through the application of a particular military technology, the longbow.
The two main tasks outlined in a memo today on Project Agincourt, one of which is laying the groundwork for AUTOWARCOM. US Military
The office of the Direct Reporting Portfolio Manager for Unmanned Systems (DRPM-UxS) will lead this initiative. Owen West, head of the Pentagon’s Defense Innovation Unit (DIU), and U.S. Navy SEAL Senior Chief Max Strasiser are set to serve as the Chief Executive Officer and Chief Operating Officer of the DRPM-UxS, respectively.
Project Agincourt will also “prototype a new construct called warfighting acquisition, not just acquisition that happens at the Pentagon,” according to Hegseth. “This model fuses operators with entrepreneurs in rapid adaptation cycles while distributing decisions and dollars closer to frontline units, to combatant commands. Realistic training exercises, and when possible, the crucible of combat, will be used to compress innovation timelines and scale what works.”
“Also today, we are phasing in new MOS [Military Occupational Specialty] frameworks within each military department to support our shift to autonomy,” the Secretary also noted during his speech in the context of Project Agincourt and the plan for AUTOWARCOM. “We need institutional knowledge and career paths for a new generation of autonomous warfighters.”
Project Meridian sits adjacent to these efforts and has a broader mandate to explore ways to ensure the U.S. military maintains its technological edge on future battlefields.
“Project Meridian, the future of warfare, is not about developing new strategies or new policies. We already do that. Rather, its purpose is to creatively look to the future, and identify the domains that we must conquer, and capabilities we must master,” Hegseth said. “We need to be looking forward into the future to achieve technological and military dominance on the battlefield for decades to come. It is futuristic on purpose. Their work will be focused on discovering, developing, and fielding the weapons and systems that our children and our grandchildren will need in their lifetimes, without any creative limitations or restrictions, on any future battlefield, from under the Earth to beyond the Moon.”
.@SECWAR “Today, at my direction—I am announcing Project Meridian, an effort led by America’s best minds to study the future of warfare.
“Project Meridian will study the battlefields of the future, identify which weapons and technologies warfighters will need to harness to dominate in these environments, and propose actionable solutions to start developing, testing and fielding these capabilities such that America will have overmatch for the 21st century and beyond,” according to a separate memo the Pentagon put out today. “The Joint Staff and all services should be prepared to connect their future warfare efforts and research to Project Meridian.”
The full text of the Project Meridian memo put out today. US Military
Furthermore, “Project Meridian will leverage critical perspectives from outside the Pentagon, drawing upon a group of top technologists, innovators, business leaders, policy makers, and national security experts,” according to the memo. “Project Meridian’s primary function will be to map the trajectory of future warfare, identify critical capability gaps, and propose actionable solutions in furtherance of the identified goals in the 2026 National Security Science and Technology Strategy.”
This is also very much a near-term initiative, with Hegseth demanding a final report no later than 120 days from now, or January 28, 2027. The Pentagon’s Chief Technology Officer, Emil Michael, has been directed to set up this project. Today, Hegseth also named Elon Musk, Palmer Luckey, and Newt Gingrich as key members.
Elon Musk, founder of SpaceX, which dominates the commercial space launch and services market globally, as well as the host of other companies, has had an on-again-off-again relationship with the Trump Administration. Palmer Luckey is another entrepreneur now best known for founding Anduril, which has increasingly been a major player in the U.S. defense industry. Newt Gingrich, a highly influential member of Trump’s Republican Party and former Speaker of the U.S. House of Representatives, has more recently become an outspoken advocate of advancing AI technology in the United States.
A screen capture from today’s livestream of Secretary Hegseth’s speech. From left to right, Palmer Luckey, Newt Gingrich, and Elon Musk. US Military capture
The inclusion of Musk and Luckey does raise ethical questions, as they both lead companies with major U.S. government contracts already, and which could be in line to see more opportunities based on Project Meridian’s findings. What guardrails might be put in place to prevent conflicts of interest is unclear.
To reiterate, it remains to be seen how all of these efforts will progress and how, or if, they will ultimately translate to more tangible developments. The U.S. military has a long history now of launching high-profile initiatives that still struggle to demonstrate real progress, despite widespread support for accelerating the fielding of uncrewed systems, AI, and other advanced technologies. The Pentagon has certainly produced tangible results through other organizational shifts in recent years, including the aforementioned Drone Dominance initiative and JIATF-401, as well as the Replicator initiative under President Joe Biden.
At the same time, there are real potential hurdles to these new plans. The Pentagon has itself already acknowledged that support from Congress will be key to the AUTOWARCOM vision. Annual budgeting, and the uncertainty that often surrounds that will also be a factor. There is also the matter of efforts to expand the U.S. defense industrial base. The limitations of existing U.S. industrial capacity and supply chains have already become a matter of national debate in light of concerns about dwindling stockpiles of critical munitions as a result of the ongoing conflict with Iran, as well as a succession of other global crises in recent years.
During his speech, Hegseth called attention to efforts to broaden the defense industrial base with new, smaller firms, as well as agreements with large established contractors to expand on their existing capacity. He also highlighted the Trump administration’s $1.5 trillion defense spending plan for Fiscal Year 2027 and issued a new call on Congress to pass an additional $350 billion reconciliation package.
With today’s announcements, the Pentagon has certainly reinforced its commitment to expanding the fielding of uncrewed and other high-tech capabilities across all branches of the U.S. military, but there is clearly work still to be done to turn those visions into realities.
The Trump-Xi summit did not mark a US-China rapprochement, but it demonstrates that sustained diplomacy and strategic guardrails are essential to prevent intensifying competition from becoming direct conflict. The recent summit between Chinese President Xi Jinping and US President Donald Trump in Washington has demonstrated a significant paradox in contemporary international politics: the two superpowers […]
King Charles III on Thursday urged several senior tech leaders to make sure that artificial intelligence remains under control and in the service of people and the planet.
King Charles III on Thursday urged several senior tech leaders to make sure that artificial intelligence remains under control and in the service of people and the planet.
OpenAI said on Wednesday it saw six reports of unexpected, concerning or unauthorized AI model behavior, and it would begin regularly publishing reports of these incidents.
OpenAI said on Wednesday it saw six reports of unexpected, concerning or unauthorized AI model behavior, and it would begin regularly publishing reports of these incidents.
The head of the European Commission, Ursula von der Leyen (pictured), says AI has the “potential to benefit humanity”.
Her comments came on Wednesday in her landmark “state of the union” address to MEPs.
A speech touched on numerous areas and they included AI which has rocketed up the political agenda in recent times.
In her address to Euro members at their monthly plenary in Strasbourg, she referred to AI as a “tipping point of our times.”
The commission president declared, “AI is fast becoming a foundational layer for our economy and security. Europe's stance is clear. We want to make the most of AI to improve our society, our quality of life and our productivity.
“I am an optimist that AI has the potential to benefit humanity – if we get it right. But, as with any powerful new technology, AI comes with a balance of possibilities and risks. Without addressing these risks, we will never be able to unlock AI's possibilities.”
The former German defence minister noted, “As the frontier models become more capable, these risks have come more sharply into focus.
“Models being developed will allow hacking on a level we never thought possible. And they will soon be in the hands of adversaries who see the world very differently from us.”
She told deputies, “At the same time, the dangers of self-improving models are becoming ever more apparent.
“Incidents of AI agents escaping their environment or inserting malicious code are a mere glimpse. After the HuggingFace incident, developers are ringing the alarm. And CEOs of the most advanced companies tell us that it is time to slow down on the self-recursive models. To pace the frontier.”
Von der Leyen added, “If the people developing the technology are clear, then we should be too. The AI Act is of course a crucial piece to putting guardrails in place.
"And it puts Europe in the position to shape global efforts on how to tackle this.”
In her comments on AI she said, “So, we will step up together with our like-minded partners like Canada, the United Kingdom and others. We want to team up on model evaluation, verification, early warning, AI security and much more.
“And I will invite the main frontier labs for a discussion on how we can support ongoing industry efforts to pace the frontier,” she said.
Britain's King Charles will host the leaders of major artificial intelligence companies this week to discuss whether a shared set of principles is needed.
Britain's King Charles will host the leaders of major artificial intelligence companies this week to discuss whether a shared set of principles is needed.
Concerns about the water and energy demands of data centers have become prolific in recent weeks. But debates about their impact often overlook the rapid advances in cooling technology and cleaner energy that are already having an impact.
However, data centers’ resource consumption should be front and center of the AI conversion, as it is set to grow rapidly. By 2030, data centers’ water and energy consumption will double as AI demand leads to rapid construction of the infrastructure.
But data centers will not cause the world to run out of water or energy. Water- and energy-saving technologies already in use are reducing their impact, while Shanghai’s groundbreaking underwater data center demonstrates how innovative design can limit water, energy and land use.
Data centers have underpinned the internet since its founding. Apps, email, file storage and cloud services all run through them.
Modern AI workloads, however, have transformed their power demands beyond recognition. Today a single AI data center can match the electricity consumption of 100,000 homes, making them a significant drain on local energy grids, especially if a region is susceptible to shortages.
Data centers’ energy consumption is only rising as more are built across the world. In the US, data centers could increase from 4.4% to 12% of national electricity usage by 2028 — that is equivalent to adding eight New York Cities to the American grid.
Data centers also consume large amounts of water in two main ways. Water is used directly to cool servers and storage systems, which generate considerable heat, and indirectly to produce electricity at fossil fuel power plants where it is heated into steam to drive turbines.
In the UK, data centers have been estimated to use up to eight times as much water as average on hot summer days, underlining how their impact can rise when local resources are already under strain.
As the effects of climate change intensify, the location of data centers, and methods to reduce their water and energy consumption, are becoming ever more important.
Fortunately, resource-saving measures are already being implemented. Data centers on land use closed-loop cooling systems that recycle water rather than continually evaporating it. This can reduce freshwater use by up to 70% compared to traditional open evaporative methods.
China has also pioneered a revolutionary new type of data center in Shanghai: wind-powered, and underwater. The facility, which began full commercial operations in May 2026, uses seawater as a coolant instead of using refrigerated fresh water as land-based data centers do, reducing the electricity required to cool the computers.
It uses at least 30% less electricity than traditional data centers, and offshore wind turbines reduce reliance on fossil fuels and cut the data center’s carbon emissions. Innovative solutions such as these will go a long way to reducing data centers’ resource consumption.
AI is also part of the solution to data centers’ draw on resources. AI can optimise electricity grids and shift computing workloads towards periods when cleaner energy is more efficient, making energy use more effective overall. AI can also anticipate fluctuations in demand and adjust energy consumption accordingly, reducing waste.
It might sound obvious, but some areas are just better equipped to accommodate a data center than others. For example, facilities in hot or water-stressed regions will impose a heavier burden on resources than those supplied by cleaner energy with plentiful water and located in cooler climates.
AI’s growing demand for data centers will increase electricity and water consumption, that is true. But it does not mean the technology will exhaust regional or global supplies of either resource.
The AI boom and the infrastructure buildout accompanying it are well underway. To manage its impact, the industry needs to plan ahead. Cleaner energy, efficient cooling, and sites that have significant water and energy supplies will all limit the infrastructure’s drain on resources.
Earlier this year, an Iran-linked actor used Anthropic’s Claude artificial intelligence model to scrape and analyze data with the aim of helping target U.S. naval forces in the Middle East. A group in Yemen – very likely Iranian-backed Houthi militants – also used Claude to assist with work on guidance systems for ballistic and hypersonic missiles.
These are just some of the revelations about how adversaries and other malign actors have been using one of the premier pieces of AI software being developed in the United States that were contained in a new report Anthropic released late yesterday. The 154-page document also details instances where Claude was used directly and indirectly to support intelligence gathering and surveillance, cyberattacks, influence operations, and even worrisome biological research over the past eight months. In addition to Iran and Yemen, the human actors behind these activities were based in Russia, China, and a host of other countries.
Anthropic has shared details about what it calls the “misuse” of its models in the past, but describes the new report as its “most detailed” to date. The company has clearly been cataloging these case studies, and has assigned each one a control number. Though the entire contents of the report are cause for concern, the explicit attempt to use Claude to target American forces, as well as to develop new higher-end missiles and other military technology, are particular standouts and signs of what’s to come, as we will highlight below.
GTG-30005: Military reconnaissance
GTG-30005 is the case study dealing with the targeting of U.S. naval forces, and the summary is as follows:
“In another investigation, we identified and disrupted an Iran-nexus threat actor that used Claude to collect and analyze publicly accessible data to develop targeting recommendations against US naval forces in the region. The threat actor used Claude to compile targeting handbooks, through a Python pipeline the threat actor built with Claude’s assistance, to identify and track naval positions based on open-source information. The compiled material included a roster of US personnel scraped from captions on public military photographs; publicly accessible ship and aircraft transponder identifiers; commercial satellite-imagery query scripts; and an inventory of public websites that exposed US naval movements. The threat actor also directed Claude to compile vulnerability research on shipboard systems, cataloging known CVEs in maritime VSAT terminals, Cisco communications equipment, and industrial control products.”
“We banned the actor’s account, developed detections to reduce the risk of future misuse, and shared threat intelligence with government authorities to disrupt the threat.”
The Nimitz class aircraft carrier USS George H.W. Bush and several other US Navy warships sail together in the Arabian Sea in June 2026. CENTCOM
How exactly Anthropic defines the term “nexus” here is not immediately clear, and other entries in its new report also use “state-nexus” in addition to country-based variations. However, in this specific case, an addendum is added to the entry that explicitly says that the same account was separately tied to “domestic mass surveillance” software development work “for Iranian state systems,” clearly pointing to an actor directly linked to the regime in Tehran.
The use of AI technology by nation-state armed forces to develop target sets and engage them is already known to be growing globally. The Israel Defense Forces (IDF) have been a pioneer in this regard, but the U.S. military, China’s People’s Liberation Army (PLA), and others are also known to be developing and fielding these capabilities.
GTG-87001: Disrupting a Yemen-based guided weapons engineering cell using Claude to develop guidance software
Anthropic’s new report details an instance where an individual in Yemen used Claude as part of missile guidance work, which is given the control number GTG-87001.
“We identified a cell of threat actors based in northern Yemen running three weapons development programs: a guided rocket that used a commodity phone-class flight computer with final-phase homing guidance; a multi-stage ballistic missile with a stated range goal above 2,000 km; and a multi-variant missile (referred to as the “R2000” set) that included a hypersonic glide vehicle variant.”
“The actors used Claude Code in place of human software engineers to develop the guidance, navigation, and control (GNC) software that steers and stabilizes a flying vehicle. For example, they used Claude to integrate an open-source autopilot onto a phone-class flight computer, writing the control and position estimation software, tuning the control settings, running a firmware build pipeline, and performing a flight simulation. The actors managed several Claude instances at once, assigning each one a role, much as a lead would delegate work on a small engineering team: the actors tasked one instance with writing the code, another with research, and a third with reviewing the code the first instance produced.”
A graphic included in Anthropic’s new report summarizing the ways in which Claude was used by the actors in Yemen to support missile guidance system development. Anthropic
“Our safeguards blocked many of their requests, but not all of them. The actors used a variety of tactics to evade our safeguards, including hiding their goals and the products the software was meant for, and they split their work across multiple sessions so no single session revealed their full intent.”
“These actors carried out a sustained effort to develop guided weapons, including using Claude to design guidance software. We do not have evidence the actors succeeded in fielding an operational device; but they did test-fire a guided rocket. This field test appears to have failed: within hours, the actors returned to Claude to work out why it failed.”
“We identified this activity as part of our internal investigations into suspected weapons development. We banned accounts associated with the actors and shared threat information with public- and private-sector partners to mitigate risks posed by the actors. Nevertheless, we have evidence that the actors had already built an offline simulation toolkit that does not rely on Claude or other engineering computing environments such as MATLAB.”
As mentioned earlier, the report does not name the Houthis, but that group has historically been based primarily in the northern end of Yemen. It is also the only entity in the country known to have active missile development programs, which are conducted with critical support from Iran.
Whether this particular work directly resulted in any new capabilities is unknown. Anthropic notes that there looked to be at least one failed test launch before the accounts were banned, which would have occurred sometime after the beginning of this year, based on the stated cutoff dates for the report. By the end of 2025, the Houthis had already amassed a wide array of very real ballistic missiles, including types with at least claimed ranges of 2,000 kilometers (approximately 1,242 miles). The group also has a variety of cruise missiles and one-way attack drones that it has employed operationally.
An example of just one of the many types of ballistic missiles in the Houthis’ arsenal. Houthi-controlled media
Still, being able to leverage Claude could only have helped any domestic missile development efforts in Yemen, which might help reduce dependency on Iran, at least to a degree. Furthermore, developing functional hypersonic glide vehicles capable of prolonged stable flight has historically been notoriously difficult, even for entities with extensive relevant knowledge bases on an organizational level.
GTG-17001: Disrupting a China-based operation using Claude to draft a fire control specification and acquisition documents for undersea warfare
Use of Claude in foreign military endeavours goes beyond the Middle East, as evidenced by case study GTG-17001. This involved a Chinese actor conducting work related to a naval counter-torpedo defense system.
“We identified a China-based threat actor who used Claude to advance three parallel tracks of work on an anti-torpedo weapons system:”
“First, the actor used Claude to draft a Chinese-language specification for an anti-torpedo fire control system (the core logic that aims and times an anti-torpedo weapon’s response). The document was written to win approval from a Chinese defense manufacturer, which would move the work on to technical certification and operational testing.”
“Second, the actor used Claude to produce a Chinese-language technical proposal of more than 200 pages, accompanied by an executive briefing deck.”
“Third, the actor used Claude to benchmark their own system against specific US anti-torpedo and anti-submarine programs based on publicly accessible information. They then generated a Chinese-language briefing on US Navy systems derived from open-source reporting.”
“The actor presented themselves as an original equipment manufacturer in the US defense sector. We assess the actor was associated with a Chinese defense industry manufacturer aiming to produce a weapons specification and acquisition proposal for the People’s Liberation Army Navy.”
This picture shows a test of an anti-torpedo system fitted experimentally to a Nimitz class aircraft carrier in the 2010s. USN
“The actor used Claude to write the acquisition proposal, refining it over many drafts. After each draft, the actor instructed Claude to role-play a hostile expert reviewer to critique the proposal, then used that feedback to sharpen the next version. In parallel, the actor used Claude to build pieces of the anti-torpedo weapons system’s fire control software and a test matrix to validate them.”
“The operational lift the actor achieved was a function of using Claude to automate complex technical outputs. The actor leveraged the model to compress the development timelines for the certification registry, compliance documentation, and automated fire control logic. The actor also accelerated the traditional human review cycle by having Claude critique the acquisition proposal across multiple rounds of review while role-playing a persona.”
“We uncovered this activity as part of our internal investigations into suspected weapons development. We cannot attribute the activity to a specific entity or actor. But we have banned the account for violating our Supported Regions Policy and our Usage Policy, which prohibits weapons design and development, and incorporated our investigative findings into our safeguards to mitigate the risk of future misuse.”
Anthropic’s report does not name the U.S. programs leveraged here, but does note the information in question was publicly available. Anti-torpedo systems for surface warships and submarines have been an area of significant interest for the U.S. Navy for years now. Those developments have also produced compact torpedo designs for use in other applications.
A Very Lightweight Torpedo (VLT), a compact torpedo design developed by Northrop Grumman. Northrop Grumman
The PLAN’s interest in these same kinds of capabilities is not at all surprising. Underwater threats and countermeasures to them, broadly speaking, would play a central and vital role in any future conflict between the United States and China across the broad expanses of the Pacific.
GTG-27005: Disrupting a Russia-based operation using Claude to engineer an autonomous military drone swarm
Another case study, GTG-27005, details the use of Claude by Russian actors working on drone swarming technology.
“We identified likely freelance Russia-based threat actors who set out to build a full-stack autonomous first-person-view (FPV) kamikaze drone swarm. The actors used Claude Code to write and test the code and save it directly into the actors’ own project files. In addition to Claude Code, the actors used a software-in-the-loop simulation stack and a rented graphics processing host for model training. They called the operation ‘DronDoc’ or ‘Serafim.'”
“The actors used Claude to build the core software system, including the drones’ shared swarm memory and fault-tolerant coordination logic (FTCL); an onboard small language model to govern attack, observe, and return-to-base behaviors; a terminal guidance software system to steer drones to their target (using the onboard camera) and issue the call to detonate; a control-link geolocation module to find opposing drone operators; a passive acoustic detection layer; and low-level logic for the drones’ programmable chips. The actors designed the platform for autonomous lethal engagement; the onboard model could select targets (including a ‘person’ target class) and issue detonation commands without a human in the loop. The actors’ activity—including flashing the low-level firmware to live development boards, provisioning single-board computers, and wiring up a simulation environment over a mesh network—confirmed that they were using real hardware-in-loop testing within their sessions.”
Many were surprised yesterday by the news that a Russian fiber-optic FPV drone flew into Kramatorsk and attacked a car.
But there is nothing surprising here. The war of 2025 is already very different from the war of 2024. From LBZ to Kram — 20 kilometers. Enemy FPVs can fly even… pic.twitter.com/hTfhJFPcxZ
“The actors trained a computer vision classifier on scraped Ukrainian combat footage, splitting the target classes into “enemy” and “friendly,” and allow-listing Russian systems. They also repeatedly used a fixed coordinate in Donetsk Oblast as the demonstration strike point, with front-line cities and corridors in Ukraine as the mission geography.”
“The actors created their accounts between late 2025 and early 2026 and started the operation in mid-May 2026. The actors circumvented our geographic access controls by routing traffic through commercial virtual private servers.”
“We assess the actors were a small, specialized freelance team doing a mix of civilian and military work, not a Russian state entity. We identified nine accounts associated with this group; eight were used only for ordinary freelance work, not weapons-related software development. Based on our investigation, we assess the actors had ties to a regional university with a federal research center associated with the Russian Academy of Sciences. The actors claimed to have received funding from Russia’s Advanced Research Foundation, National Technology Initiative, and Ministry of Defence, though we cannot verify those claims. We identified this activity as part of our internal investigations into suspected weapons development, we banned accounts associated with the actors, and have incorporated our investigative findings into safeguards to reduce the risk of future misuse.”
Anthropic’s report included a table, reproduced below, detailing different types of drones and other capabilities that were fed into the model as part of this work.
GTG-17002: Disrupting a China-based operation using Claude to build targeting software for electronic warfare and air defense suppression
The last case study in the Conventional Weapons section of Anthropic’s new report, GTG-17002, covers electronic warfare-related work by a Chinese actor, with a particular focus on potential air defense targets in Taiwan.
“We identified a China-based actor who used Claude’s chat, coding, and agentic work tools to design, build, and iterate on a Chinese-language suite of about 16 modules for electronic warfare, using the electromagnetic spectrum to detect, jam, or deceive an opponent’s radar and communications, and for suppressing an opponent’s air defenses.”
“The actor used Claude to build the software system, from the underlying logic to the user interface, and iterated through 12 versions. This included implementing and optimizing the system’s radar detection and jamming physics, generating a vulnerability analysis module, and drafting Chinese-language targeting instructions. The software suite the actor built analyzed an opponent’s radars, surface-to-air missile sites, command posts, and communications nodes, then computed their detection coverage, assessed the effectiveness of jamming, ranked targets by value and vulnerability, including which to suppress first, and determined how best to assign jammer sorties to targets across multi-day campaigns. The suite also ranked which of an opponent’s assets to suppress first, and modeled specific engagement envelopes, including those of Patriot and THAAD-class [Terminal High Altitude Area Defense] systems.”
“Mid-project, we observed the actor change the simulation’s default scenario to 12 targets in Taiwan. The targets included a command bunker in Taiwan, an early warning radar site, Patriot and Tien Kung [surface-to-air missile] batteries, major air bases, and a regional combatant command headquarters.”
“The actor also ran a self-hosted model on an internal network alongside Claude and connected the software suite to this model through a tool-use integration.”
“Based on our investigation, we assess the actor is a China-based defense and military-industrial researcher. Account-level metadata and content flagged by our safeguards indicated the actor was linked to PRC research institutions, including the PLA Academy of Military Sciences. We detected this activity as part of our internal investigations into suspected weapons development, we banned accounts linked to the actor, and have incorporated our investigative findings into our safeguards to reduce the risk of future misuse.”
An accompanying table, seen below, mentions several specific types of U.S.-made air defense radars, as well as the Link-16 datalink network, as having come up in the course of this work. Electronic warfare variants of the Chinese Y-8 and Y-9 aircraft, as well as the J-16D electronic warfare jet, and what appears to be a reference to the Golden Eagle drone helicopter, are also mentioned.
A Chinese Y-8GX-11, one of several electronic warfare variants of the Y-8/Y-9 design. Chinese internet via X
The suppression of enemy air defenses is an essential part of any modern air campaign, whether conducted independently or as a part of a larger operation. This would be no less critical in any military intervention against Taiwan from the mainland, or any other larger-scale conflict the PLA might find itself in.
Guardrails and ethics
Beyond the details in any of the specific case studies, Anthropic’s new report underscores growing broader concerns about guardrails for the use of AI and general ethics surrounding the technology. As noted in the instances above, Anthropic has also implemented various safeguards to try to block access to Claude from certain regions and/or to prevent the model from conducting certain types of work. At the same time, what the company has now shared makes clear that those protections can still be circumvented to a significant degree, at least for now.
TWZ has noted in the past that America’s adversaries are likely to be less worried about ethics, in general, when it comes to AI. Anthropic’s new report also shows there are few bounds to what state-aligned or even non-state actors might seek to use this technology for, including the development of potential weapons of mass destruction.
At the same time, AI models like Claude are not going away, and developments continue to advance at an ever-quickening pace, opening the door to further proliferation. Developments are rapidly intermingling globally, with companies in China notably leveraging models created in the U.S. to further their own work, which is then released publicly. The technology will increasingly become more integrated and critical to most aspects of military operations and weapons development, alongside its commercial and civilian applications. It is just as much a superweapon as anything else, and the country with the more capable models, and the willingness to deploy them, will have a major edge in future conflicts.
With all this in mind, it seems very plausible, if not probable, that the case studies described in Anthropic’s new report are just the tip of an approaching iceberg that will need a concerted effort to tackle.
Neurotechnology and artificial intelligence are increasingly converging as neurodata can be collected, combined and reused to develop AI models and applications across different contexts. This creates opportunities for healthcare, rehabilitation and scientific research, but also raises important questions for fundamental rights, democratic accountability and the public interest.
In this context, the EGE calls for a systemic and anticipatory approach to governing neuro-AI. The Statement shifts the focus from individual devices and neurodata privacy to the wider impacts of the infrastructures through which neurodata are collected, processed and reused. It also addresses questions of European technological sovereignty and public-interest infrastructures.
The EGE recommends:
Robust protection of neurodata: define and reinforce protections for neurodata and neurodata-derived inferences under EU data protection law
Responsible development of neuro-AI technologies: ensure possible development of Brain Foundation Models and related neuro-AI infrastructures
Strong Rights protection: ensure that individuals are protected against disproportionate control where neurodata and neurodata-derived data inferences are used in consequential contexts
Enhanced governance capacity: build European public-interest governance capacity for neuro-AI infrastructures
Comprehensive fitness check: conduct a targeted regulatory assessment of the EU framework governing neuro-AI systems
Background
Neuro-AI Infrastructures are interconnected systems through which neurodata are collected, processed and reused to create, deploy and manage AI-powered neurotechnology applications.
It is part of theEthics Advice Mechanism (EthAM), which provides the European Commission (EC) with high quality, independent advice on all aspects of EU legislation and policies, where ethical, societal and fundamental rights issues intersect with the development of science and new technologies.
Login
