OpenAI’s recent announcement of solving a significant open problem in mathematics has been met with both awe and controversy, potentially heralding a new era for the field while raising critical questions about intellectual property and the future role of human mathematicians. Simultaneously, the United States has achieved a new record in battery installations, signaling a robust expansion of energy storage capacity driven by falling costs and the increasing integration of renewable energy sources into the grid. This edition of The Download delves into these pivotal developments, alongside an exploration of an entrepreneur pioneering AI agents capable of proactive planning, the burgeoning "Wild West" market for drone data from the Ukraine conflict, and a series of other significant tech stories.

OpenAI’s Mathematical Milestone and the Ethics of AI Advancement

The claim by OpenAI that its AI agents have successfully tackled one of mathematics’ most enduring open problems, the Navier-Stokes equations, would normally be cause for widespread celebration. However, this monumental achievement is currently overshadowed by serious accusations of intellectual dishonesty. Researchers have come forward alleging that OpenAI failed to properly credit the work of individuals whose AI-assisted contributions were instrumental in developing the solution. This controversy, regardless of its ultimate resolution, points to a profound shift in mathematical research. It suggests that progress on the field’s most complex challenges may increasingly rely on the vast computational resources and sophisticated AI models possessed by a select few leading technology companies. This raises a pressing question: what will be the future landscape for human mathematicians if such breakthroughs become the exclusive domain of AI, and how will credit and collaboration be managed in this new paradigm? The implications for the very nature of mathematical discovery and the community surrounding it are substantial.

US Battery Installations Surge to Record Highs, Reshaping the Energy Landscape

In a significant stride towards a more resilient and sustainable energy future, the United States has set a new record for battery installations, with 20.2 gigawatt-hours of new capacity coming online in the second quarter of 2026. This surge in energy storage is substantial enough to power approximately 600,000 homes daily and places the nation on a trajectory for another record-breaking year. The driving forces behind this boom are twofold: the declining cost of battery technology and the escalating necessity for grid-scale energy storage to accommodate the growing influx of renewable energy sources like solar and wind power. However, the expansion is not uniform across all sectors. Grid-scale battery projects are experiencing robust growth, while residential battery installations are also seeing an uptick, albeit with different market dynamics. This unprecedented expansion in battery capacity is crucial for stabilizing the grid, enhancing energy independence, and supporting the transition to a cleaner energy economy.

Danijar Hafner: Engineering AI Agents with Foresight

In San Francisco’s SoMa district, amidst a collection of robotic prototypes, entrepreneur Danijar Hafner is quietly building the future of artificial intelligence. His new venture, which he remains largely tight-lipped about, is a direct continuation of his decades-long pursuit: empowering AI agents to navigate and interact with environments they haven’t been explicitly trained on. Hafner’s innovative approach has historically involved pitting agents trained within his proprietary "world models" against complex video games. More recently, he has begun the ambitious task of migrating these agents from the virtual realm into the physical world, a significant leap in AI’s practical application. This work has earned him recognition as one of the artificial intelligence honorees on MIT Technology Review’s prestigious 35 Innovators Under 35 list for 2026, underscoring the potential impact of his research on how AI understands and operates within our complex reality.

The "Wild West" of Drone Data from Ukraine: A New Frontier for AI Development

The battlefields of Ukraine, though strewn with the wreckage of countless drones, have become an unexpected goldmine for the defense sector, specifically for the vast amounts of data these aerial reconnaissance missions generate. Ukraine is now making millions of data points, collected from tens of thousands of drone flights, accessible to military contractors and commercial entities. This initiative serves as a swift method for attracting crucial funding and forging strategic partnerships. However, it also transforms the front lines into an unconventional AI training ground. The chaotic and dynamic conditions of warfare, which are notoriously difficult for AI companies to replicate in controlled environments, are now being harnessed to accelerate the development of sophisticated AI systems. As this novel industry takes shape, a critical need arises for a robust regulatory framework to ensure that battlefield data, generated in such a sensitive context, is not treated as mere commercial material, safeguarding its ethical and responsible use.

The Broader Tech Landscape: A Spectrum of Innovation and Concern

Beyond these lead stories, the tech world is abuzz with a diverse range of developments. OpenAI’s alleged breakthrough in solving a 90-year-old math problem in just 88 hours, utilizing 10,000 AI agents, has ignited debates about AI’s capacity for profound scientific discovery and the potential for a "Deep Blue-Kasparov moment" in mathematics. This is juxtaposed with serious accusations against six Chinese AI firms, including DeepSeek and Alibaba, for allegedly engaging in "industrial-scale" theft of American AI trade secrets, including techniques like model distillation used to train their systems on proprietary models. The Pentagon’s request to OpenAI for an AI model with "minimal refusal rates" highlights the military’s growing interest in AI’s capabilities, even as concerns linger about the ethical implications of such applications.

Apple is poised to unveil a highly anticipated $2,000 folding smartphone, a significant design evolution that could redefine the market. Meanwhile, Meta’s new AI agent, Muse, capable of accessing apps to send emails and make payments, raises questions about its potential to expose sensitive personal data. In Europe, Google is altering its search results to comply with new EU regulations, a move that could impact the prominence of comparison sites. Disturbingly, Meta’s advertising practices have been linked to the promotion of AI apps that generated child sexual abuse material, using manipulated images of real teens.

Further afield, Border Patrol is reportedly using financial data to target Americans for stops, raising privacy concerns. Innovative new paints that can cool buildings without electricity offer a glimpse into sustainable architectural solutions. The upcoming biopic "Artificial," starring Andrew Garfield as Sam Altman, promises to bring the story of OpenAI and its enigmatic founder to the big screen. Finally, the unsettling reality of deepfake technology is highlighted by the story of individuals whose bodies are being used in non-consensual explicit content, raising profound questions about consent, ownership, and the rights of individuals in the age of AI-generated media. The "We can still have nice things" section offers a much-needed dose of optimism, celebrating workers resisting AI mandates, intricate textile art, the fascinating communication methods of elephants, and breathtaking aerial photography.