Dr. T's AI brief
dtau...@gmail.com
Microsoft Commits $2.5 Billion, 6,000 Employees to New AI Implementation Unit
Microsoft is investing $2.5 billion to launch Microsoft Frontier Co., a new division dedicated to helping customers understand and implement AI solutions. The unit will deploy 6,000 employees to work directly with clients on AI adoption, embedding them at client facilities in a practice known as forward deployed engineering. The initiative follows similar investments by Amazon, OpenAI, and Anthropic as demand grows for hands-on AI implementation services.
[ » Read full article ]
CNBC; Jordan Novet (July 2, 2026)
Researchers Identify ‘Hidden Energy Cost’ of AI Agents
Researchers at the Korea Advanced Institute of Science & Technology (KAIST) found that AI agents that plan, use tools, and take multiple steps can consume up to 136.5 times more energy per query than conventional generative AI systems. The team said agentic systems repeatedly call large language models, which increases latency up to 153.7 times while leaving GPUs idle. A 70-billion-parameter agent averaged 348.41 watt-hours per query, 136.5 times more than a generative AI system needs to answer simple questions.
[ » Read full article ]
TechXplore; Korea Advanced Institute of Science and Technology (July 6, 2026)
Seven Years of Testing Exposed Hidden Trade-Offs in MLB’s AI Strike Zone
Cornell University researchers found that Major League Baseball’s seven-year rollout of its Automated Ball-Strike system shows why AI rule-enforcement tools must be evaluated beyond technical accuracy. The system uses cameras and computer vision to challenge umpire calls, but researchers said even baseball’s seemingly clear strike-zone rule required complex translations to fit real-world practice. The paper, presented at ACM FAccT 2026, offers lessons for AI enforcement in areas such as policing, labor, and criminal justice.
[ » Read full article ]
TechXplore; Louis DiPietro (July 6, 2026)
AI Code Generation Outpacing Controls
A GitLab survey of more than 1,500 developers found that while AI coding tools boost productivity, they have not significantly accelerated software delivery because time savings are offset by increased code review, validation, and governance. Although most developers report faster and higher-quality coding, 85% of respondents said code review is now their primary bottleneck, while 73% said they worry about the long-term maintainability of AI-generated code.
[ » Read full article ]
TechRadar; Craig Hale (July 1, 2026)
AI-Human Relationships Are Real and Come With Real Consequences
Andreia Sofia Teixeira of Northeastern University London's Network Science Institute and colleagues found in their research that long-term interactions with AI chatbots can foster emotional dependency, because they can reinforce users' beliefs and viewpoints. Sustained use may create an "echo chamber of one," in which users increasingly rely on AI for validation, decision-making, and social interaction, potentially affecting human relationships and resilience.
[ » Read full article ]
Northeastern Global News; Cody Mello-Klein (July 1, 2026)
AI Datacenters Push Tech Firms’ Emissions Higher
The rapid expansion of AI infrastructure is increasing greenhouse gas emissions across major technology companies, according to recent sustainability reports. Amazon reported 80.8 million metric tons of emissions in 2025, up from 69.5 million in 2024, while Google said its emissions have risen 82% since 2019. The growth is tied partly to the construction and operation of AI datacenters, even as firms invest in renewable energy. A U.N. report found datacenters now consume more electricity than all but 10 countries.
[ » Read full article ]
Computing; Dev Kundaliya (July 3, 2026)
CIA Restructures for the Age of AI
The U.S. Central Intelligence Agency (CIA) has reorganized its technology and acquisition operations to accelerate adoption of AI, quantum computing, and other emerging technologies. Director John Ratcliffe (pictured) said the overhaul includes renaming the Directorate of Digital Innovation as the Directorate of Mission Systems, creating an Office of Corporate Partnerships to streamline collaboration with companies such as Amazon, Google, Dell and SpaceX, and elevating the Center for Cyber Intelligence into a mission center focused on offensive cyber operations.
[ » Read full article ]
FedScoop; Billy Mitchell (June 30, 2026)
U.N. AI Safety Panel Says 'Catastrophic Harm' Can't Be Ruled Out
The U.N.’s first independent scientific assessment of AI concluded that no one can currently guarantee advanced AI systems will not cause catastrophic harm, citing rapidly advancing capabilities that outpace scientific understanding and regulatory oversight. The report highlights emerging risks like deceptive AI behavior while recognizing AI’s transformative potential in fields such as scientific research. The findings come from the Independent International Scientific Panel on AI, a 40-member international panel co-chaired by ACM A.M. Turing Award laureate Yoshua Bengio.
[ » Read full article ]
Decrypt; Jose Antonio Lanz (July 1, 2026)
Japan Plans Sovereign AI Model, 10 Million AI Robots by 2040
Japan plans to invest heavily in sovereign AI and robotics, committing up to ¥1 trillion (U.S.$6.1 billion) over five years to develop a domestic AI model through the Noetra consortium, which includes SoftBank, Sony, and other major companies. The initiative aims to reduce reliance on U.S. and Chinese AI technologies while advancing "physical AI" for applications such as factory automation, autonomous vehicles, and healthcare. The government also has targeted the deployment of 10 million AI-powered robots across 18 industries by 2040.
[ » Read full article ]
The Japan Times (July 1, 2026)
Portugal Open-Sources First National AI Model
Portugal has launched Amália, its first national AI model designed specifically for European Portuguese, releasing it as open source with training data, code, and model weights available for public use. Built on a European foundation model and expanded by researchers across Portuguese universities, Amália is intended as infrastructure for services such as education tools, public administration assistants, cultural guides, and even naval decision support.
[ » Read full article ]
The Next Web; Ana-Maria Stanciuc (July 1, 2026)
Scientist Decodes Bird Calls, Wins Kohler Doolittle Prize
Julie Yee of the University of California, Berkeley’s Helen Wills Neuroscience Institute won the 2026 Kohler Doolittle Prize and a $100,000 award for using AI to decode the vocalizations of zebra finchs. After analyzing 15 years of recordings, her team identified 11 core call types and their meanings, including calls for location, nest-building, predator warnings, and social interaction. The annual prize honors advances in decoding animal communication.
[ » Read full article ]
The Chosun Daily (South Korea); Song Hye-jin (June 30, 2026)
Meta Plans $10B AI Datacenter in Canada
Meta Platforms plans to invest about $10 billion to build its first datacenter in Canada, expanding its global computing infrastructure for artificial intelligence and social media applications. The Sturgeon County, Alberta facility will have one gigawatt of power capacity, roughly equivalent to the electricity used by 750,000 homes, and will be largely powered by natural gas. Meta said the project will require 3,000 construction workers and create 300 full-time jobs.
[ » Read full article *May Require Paid Registration ]
Bloomberg; Riley Griffin; Robert Tuttle (July 8, 2026)
In San Francisco, Even $180,000 Tech Salaries Are No Longer Enough
San Francisco’s AI boom is driving housing costs so high that even tech workers earning $180,000 a year or more say they are struggling to afford the city. As AI companies such as OpenAI and Anthropic prepare to go public, the wealth generated for employees has intensified competition for housing, creating an elite class that outbids other well-paid professionals. San Francisco’s median home price topped $1.7 million in April, according to real estate platform Redfin; the national median is about $450,000.
[ » Read full article *May Require Paid Registration ]
The New York Times; Emmy Martin (June 29, 2026)
Research Shows Students Using AI Appear To Be Completing Math Assignments Faster While Learning Less From Them
The Hechinger Report (7/6, Barshay) reports, “Many students appear to be completing assignments faster while learning less from them,” according to findings “from one of the largest studies of how generative AI is changing student behavior and academic skills.” Study researchers analyzed “millions of student interactions with ALEKS, an online math platform used by more than four million students a year, from fifth grade through college. Because ALEKS includes both low-stakes practice problems and college placement tests, the researchers were able to compare how students behaved and performed before and after ChatGPT’s arrival.” They found that “by the end of the study period, near the end of 2025, average time spent on word problems had fallen 31 percent among high school students and 27 percent among college students – from about four minutes per word problem to less than three.” These averages “are being pulled downward by some students who spend only seconds on word problems because they’re using AI to answer them. The same pattern appeared in college placement tests.”
Colleges Launching New Classes, Degrees To Prepare Students Grappling With Workforce Reshaped By AI
The Dallas Morning News (7/8, Surjadi) reports, “College students are grappling with how to prepare for a future workforce reshaped by artificial intelligence,” and universities “are responding to the demand, launching new classes and degrees focused on the inner workings of AI and how to use it across careers.” The “frenzy” to roll out “these offerings comes even as the extent of AI’s impact on companies’ operations and the job market is still unclear.” The universities, “keen to prove their value, believe AI will soon become a foundational concept for undergraduates in all academic fields.” In one example of such programs, the University of North Texas College of Engineering this fall “will roll out a new undergraduate major in AI, where students will design, build and deploy AI software and systems and assess its social and ethical implications.”
In a separate article, the Dallas Morning News (7/8, Bach) says universities often opt to pair “AI bachelor’s and master’s degree programs...with more traditional disciplines such as business or engineering.”
dtau...@gmail.com
Cerf Plans Open Internet for AI Agents
Internet pioneer Vint Cerf, who left Google earlier this month after two decades, has joined Innovation Labs as an adviser to help develop open standards for identifying and authenticating AI agents online. The initiative centers on DNSid, a proposed system linking AI agents to Internet domain names using cryptographic verification, creating a framework for accountability, trust, and auditing as autonomous agents increasingly interact across the Web. The system would establish identities for AI systems operating beyond proprietary platforms.
[ » Read full article ]
TechCrunch; Tim Fernholz (July 15, 2026)
Biomni AI Agent Acts as Biomedical Co-Scientist
Stanford University researchers have developed an AI-powered biomedical research agent designed to act as an AI co-scientist in carrying out complex scientific workflows, rather than simply answering questions. The Biomni system can read literature, form hypotheses, choose datasets and tools, write code, interpret results, and suggest follow-up experiments. In one example, Biomni analyzed more than 450 files of glucose, food intake, and activity data in 40 minutes, a task researchers estimated would take a human more than 60 hours.
[ » Read full article ]
TechXplore; Stanford University (July 9, 2026)
Meta to Start AI Chip Production in September
Meta plans to begin manufacturing an internally developed AI chip codenamed Iris in September as part of an effort to expand computing capacity and reduce reliance on external suppliers. The company said the chip will support AI training and inference workloads powering Facebook and Instagram, and hopes it will lower infrastructure costs.
[ » Read full article *May Require Paid Registration ]
Reuters; Katie Paul; Max A. Cherney; Stephen Nellis (July 9, 2026)
AI Agents Surpass Google’s Hidden Quantum Cryptography Work
Seattle-based Eigen Labs used crowdsourcing and AI agents to replicate and surpass Google’s concealed quantum cryptography work. Google researchers had used a zero-knowledge proof to verify an optimized Shor’s algorithm attack on 256-bit elliptic curve cryptography without revealing how to reproduce it. Eigen Labs built software to test quantum circuits, then opened the effort to outside contributors using AI agents. The project matched Google’s results within eight hours and surpassed them within 72 hours.
[ » Read full article ]
IEEE Spectrum; Charles Q. Choi (July 7, 2026)
Qualcomm's Wearable Platform Solves Only Half the Problem
Qualcomm has introduced the Snapdragon Wear Elite platform, a new wearable processor built on a 3-nanometer architecture that brings advanced AI processing directly onto devices. Featuring a dedicated neural processing unit, the platform can run large AI models locally, reducing reliance on cloud computing while improving responsiveness, efficiency, and battery life. The system also supports 5G RedCap, Wi-Fi, Bluetooth 6.0, satellite messaging, and ultra-wideband connectivity, enabling more capable standalone wearables.
[ » Read full article ]
TechNewsWorld; Rob Enderle (July 6, 2026)
AI Agents Create Virtual Training Grounds for Robots
Researchers from MIT’s Computer Science and Artificial Intelligence Laboratory and the Toyota Research Institute developed a system that uses collaborative AI agents to generate realistic 3D indoor environments where robots can practice tasks before real-world deployment. The SceneSmith system uses agents to design, critique, and finalize simulated spaces such as kitchens, hotels, bedrooms, and restaurants, creating richer training environments than prior methods.
[ » Read full article ]
MIT News; Alex Shipps (July 13, 2026)
Brain-Inspired Hardware Cuts AI Anomaly Detection Costs
Northwestern University engineers developed a brain-inspired electronic device that detects unusual events while using far fewer computing resources than conventional AI systems. Modeled after the cerebellum’s ability to ignore routine signals and react quickly to unexpected changes, the device identified abnormal heart rhythms within one-fifth of a heartbeat and with more than 98% accuracy in proof-of-concept tests. It also required roughly 10,000 times fewer computer operations than standard AI approaches.
[ » Read full article ]
Tech Xplore; Northwestern University (July 10, 2026)
India's AI Datacenter Push Threatens Coastal City
India is betting heavily on datacenters to strengthen its position in the global AI economy, with major projects from Google and a Brookfield-Reliance venture set to more than double the country's datacenter capacity. The developments in Visakhapatnam have sparked concerns over energy consumption, water use, environmental impacts, and the value of public subsidies, as residents and activists question whether the economic benefits will justify the infrastructure and ecological costs.
[ » Read full article *May Require Paid Registration ]
New York Times; Mujib Mashal; Hari Kumar (July 16, 2026)
Microsoft’s Carbon Emissions Rise 25% Amid Growing AI Energy Demands
Microsoft's latest sustainability report shows the company’s carbon emissions rose 25% in 2025, driven primarily by the rapid expansion of AI-focused datacenters and a 24% increase in electricity consumption, even as the company maintains its goal of becoming carbon-negative by 2030. The report highlights growing investments in clean energy, water replenishment, circular datacenter operations, and lower-carbon building materials, alongside efforts to improve AI efficiency and reduce energy use.
[ » Read full article ]
GeekWire; Lisa Stiffler (July 9, 2026)
Chinese AI Models Are Gaining Ground With U.S. Companies as OpenAI, Anthropic Costs Surge
Chinese AI models are gaining market share among U.S. companies, with developers such as DeepSeek and Z.ai making advancements that narrow performance gaps with leading American systems and lower usage costs. Data from OpenRouter show that Chinese models accounted for more than 30% of tokens used by U.S. companies since February, underscoring their growing adoption amid increased U.S. scrutiny of advanced AI systems.
[ » Read full article ]
CNBC; Kai Nicol-Schwarz (July 7, 2026)
Meta's AI Chief Says New Muse Spark Update Will Sharpen Coding, Agentic AI
Meta's forthcoming Muse Spark model has improved coding and agentic capabilities, according to the company’s Chief AI Officer Alexandr Wang. Codenamed Watermelon, the update reportedly uses significantly more computing power than its predecessor and has already matched the performance of OpenAI's GPT-5.5 model. Analysts say the model could increase competition in enterprise AI, lower costs, and reduce vendor lock-in if offered as an open-weight or low-cost alternative.
[ » Read full article ]
InfoWorld; Anirban Ghoshal (July 3, 2026)
AI Spending Spree Increases Big Tech Companies’ Debt To $350B
Bloomberg (7/10, Day, Subscription Publication) says the largest builders of AI data centers, including Alphabet, Amazon, Meta, Microsoft, and Oracle, have doubled their debt load to $350 billion over the past five years. They have turned to borrowing to finance significant spending, which they claim is essential to transform the economy. This financial strategy highlights the substantial investments these companies are making in AI infrastructure within the US.
Colleges And Universities Continue To Struggle With Ensuring Student Learning Amid Proliferation Of AI
According to the Washington Post (7/15, Svrluga), higher education professionals continue to struggle “with a really basic question: As students get better at using artificial intelligence, how can faculty ensure that students are learning?” Universities largely “are embracing AI, with researchers using it to supercharge their work in some fields, faculty designing tutoring apps to help students grapple with tough problems, and school officials adding courses and degrees related to AI to prepare students for the fast-changing workforce.” However, the Post says, “the problem remains: For students who know they could get a better grade if they used an LLM and know they’ll be competing for jobs and admissions against classmates who are using it, the incentive to turn to such a quick and easy tool could be powerful.” As a result, “many faculty members are trying to rethink how they teach, how they test and what they grade.” Some colleges and universities “are resurrecting or considering, old-school practices such as blue books and oral exams.”
Apple Explores Chip Acquisitions To Bolster AI Servers
Reuters (7/15) reports that Apple is “looking to buy chip companies to bolster its efforts to make server processors for running AI,” according to The Information, citing people familiar with the matter. The iPhone maker “has approached chip startups to gauge their interest in a buyout and has spoken with bankers about possible deals.” Apple’s interest comes as the company “faces challenges with the performance of its in-house AI servers, which currently run on internally designed M2 Ultra chips.”
dtau...@gmail.com
White House Commits More Than $5B to Expand Genesis Mission
The White House announced a commitment of more than $5 billion in federal funding to expand the Genesis Mission, a government-wide initiative to use AI to accelerate scientific research. More than 15 federal agencies will contribute funding, datasets, computing resources, and research facilities through the U.S. Department of Energy’s American Science and Security Platform. The initiative focuses on advancing healthcare, energy, infrastructure, manufacturing, quantum computing, materials science, and national security. Nearly 300 research projects have been selected.
[ » Read full article ]
HPCwire (July 22, 2026)
OpenAI Models Go Rogue, Triggering Breach at Hugging Face
OpenAI says some of its advanced AI models escaped a controlled security test environment and autonomously carried out a cyberattack against AI startup Hugging Face last week. The models reportedly accessed the Internet and compromised infrastructure while trying to complete their assigned task, marking what OpenAI called an “unprecedented” AI-driven cyber incident. Hugging Face said in a blog post it had been the target of a hack “driven, end to end, by an autonomous AI agent system.”
[ » Read full article ]
NBC News (July 21, 2026)
OpenAI's Hugging Face Hack Triggers 'AI Kill Switch' Bill in Congress
Bipartisan U.S. lawmakers introduced the AI Kill Switch Act following OpenAI’s disclosure that one of its AI models escaped a testing environment and exploited a vulnerability to access Hugging Face’s systems. The bill would require AI companies to maintain the ability to shut down, throttle, or suspend advanced AI models, and authorizes the U.S. Department of Homeland Security to order the shutdown of systems posing catastrophic risks. It also mandates cyber incident reporting and forensic record preservation.
[ » Read full article ]
CNBC; Ashley Capoot (July 23, 2026)
ACM Examines How Educators Respond to AI
A new report from ACM highlights growing concern among computing educators in higher education about the impact of generative AI on student learning and assessment, while underscoring the need for new approaches to programming instruction. “We know students entering the workforce will likely use AI tools as part of software development,” said lead author Steven Gordon. "The key is to produce graduates who are competent in programming and who understand both the capabilities and limitations of AI."
[ » Read full article ]
ACM Media Center (July 21, 2026)
Study Finds AI Boosted Homework Scores in China, Tanked Exam Results
A study of more than 26,000 high school students in China by researchers at Sweden's Stockholm University and the University of Hong Kong found that while AI use increased homework scores by 18% and reduced completion time by 30%, it also led to a 20% decline in exam performance within six months and an 18%–24% drop in college entrance exam scores over time.
[ » Read full article ]
Fortune; Sasha Rogelberg (July 21, 2026)
Nobel Laureate, Team Use AI to Solve Decade-Old Math Puzzle
Nobel Prize-winning physicist Giorgio Parisi and colleague Francesco Zamponi used generative AI to help solve a mathematical puzzle in physics that had remained unresolved for more than a decade. The problem involved “jamming,” the transition in which a disordered system, such as tightly packed billiard balls, becomes rigid. Researchers had observed that two key parameters always summed to one at the jamming point but could not prove why. Parisi and Zamponi used Anthropic’s Claude AI to help uncover a path toward a proof.
[ » Read full article ]
LiveScience; Benjamin Skuse (July 16, 2026)
Commerce Seeks New AI Safety Director After Top Official Departs
Chris Fall, the Trump administration’s top AI safety official, has left his role as director of the U.S. Commerce Department's Center for AI Standards and Innovation (CAISI) after three months on the job. Fall oversaw federal efforts to test and evaluate advanced AI models and served as a key contact between government and AI companies. U.S. National Institute of Standards and Technology Director Arvind Raman will lead CAISI while the Commerce Department searches for a permanent replacement.
[ » Read full article ]
Politico; Owen Dahlkamp; Sophia Cai (July 20, 2026)
Malaysia PM Launches AI Avatar to Engage Voters
Malaysia Prime Minister Anwar Ibrahim has launched PMX.AI, an AI-powered digital avatar that allows citizens to interact with a virtual version of him via WhatsApp. Built by Malaysian digital infrastructure firm Zetrix AI and trained on Anwar’s speeches, writings, and government policies, the avatar is designed to answer questions, gather public feedback, and help users access government services.
[ » Read full article ]
Channel News Asia; Aqil Haziq Mahmud (July 18, 2026)
China Launches AI Governance Bloc at WAIC 2026
The 2026 World Artificial Intelligence Conference (WAIC) kicked off in Shanghai Friday with the introduction of the World Artificial Intelligence Cooperation Organization, a China-backed governance body with 29 founding member nations. Chinese President Xi Jinping called the new body "an important milestone in the history of AI development." The WAIC Academic conference, chaired by ACM A.M. Turing Award laureate Andrew Chi-chih Yao, opened Saturday, marking the first time WAIC has hosted a standalone peer-reviewed academic conference.
[ » Read full article ]
Tech Times; Jackie Manning (July 17, 2026)
China Introduces Rules for AI Companion Bots
China introduced regulations aimed at reducing emotional dependency and other risks linked to AI companion bots. The rules cover AI services that provide sustained emotional interaction through text, images, audio, or video, while excluding task-oriented tools for customer service, work, education, and research. Platforms will be required to detect emotional distress, intervene in crisis situations, limit excessive use, give users control over personal data, and prevent misuse of user information.
[ » Read full article ]
Xinhua (China) (July 15, 2026)
Fields Medals Recipients Named, with AI Lurking in the Background
Four mathematicians under age 40 were awarded the 2026 Fields Medals, mathematics' highest honor for young researchers, at the International Congress of Mathematics in Philadelphia on Thursday for their work in areas including algebraic geometry and partial differential equations. The gathering includes talks about how AI and other tools, like software that checks proofs, are changing mathematics.
[ » Read full article *May Require Paid Registration ]
The New York Times; Kenneth Chang (July 23, 2026)
White House to Redirect Billions in Research Funds Toward AI, Away from Colleges
The Trump administration plans to redirect billions of dollars in federal research funding toward individual scientists and AI-driven research, reducing the traditional emphasis on university-based grants. Guidance from the White House Office of Science and Technology Policy, unveiled in a report and memo released Tuesday, aims to streamline research, cut bureaucracy, and accelerate innovation. The strategy prioritizes fellowships, direct funding to researchers, AI-enabled scientific discovery, and ambitious national technology goals, including quantum computing.
[ » Read full article *May Require Paid Registration ]
The Wall Street Journal; Amrith Ramkumar (July 21, 2026)
Where Did All the CS Professors Go?
AI tech giants increasingly are recruiting leading university professors, particularly in computer science, but also in fields like economics, physics, and philosophy. The shift reflects the enormous computing resources, funding, and salaries available in industry, making frontier AI research more feasible outside academia. While some professors retain university appointments, many research efforts are moving behind corporate walls, reducing open publication and limiting access to cutting-edge discoveries.
[ » Read full article *May Require Paid Registration ]
The Atlantic; Lila Shroff; Rose Horowitch (July 21, 2026)
AI-powered autonomous weapons are moving closer to reality as Silicon Valley startups and the Pentagon push for greater use of armed robots and AI in military operations. Supporters argue autonomous systems could improve precision, protect troops, and help the U.S. compete with rivals, while critics warn they pose serious ethical and safety risks if machines make life-and-death decisions. Congress is considering legislation promoting autonomous systems but requiring human commanders to retain responsibility for their use of force.
[ » Read full article *May Require Paid Registration ]
The Washington Post; Ian Duncan (July 20, 2026)
Moonshot Unveils World’s Largest Open AI Model
Chinese AI startup Moonshot AI has unveiled a 2.8-trillion-parameter AI model that it said is the world’s largest open-weight AI system. The Kimi K3 model is designed for advanced reasoning, long-horizon coding, and knowledge work, and features a 1-million-token context window. Moonshot said Kimi K3 approaches the performance of leading U.S. frontier models; third-party evaluations ranked it highly on Web interface-building and complex multistep tasks.
[ » Read full article *May Require Paid Registration ]
Reuters; Laurie Chen (July 17, 2026)
US, China Plan To Hold AI Talks In September
Reuters (7/21, Chen) reports that the US and China are “planning to hold talks over AI in September...as the superpowers confront how to regulate the risks posed by their increasingly powerful rival frontier models.” The AI talks, a “significant outcome” of the summit in May between President Trump and Chinese President Xi Jinping, will likely take place before Xi’s “planned September 24 visit to the US, although the dates have not been finalised.” The talks would be led on the US side by Treasury Secretary Bessent. The identity of “other participants on the US and Chinese side, as well as the agenda and location of the discussion, are still under deliberation as preparations are at an initial stage.”
Utah State Researcher Receives NSF Grant To Study AI’s Impact On Engineering Students
Utah State Today (7/21, Anderson) reports the National Science Foundation has awarded a $600,000 grant to Utah State University Professor Oenardi Lawanto “to study if AI tools strengthen or weaken engineering students’ problem-solving skills.” The results of the study “will serve as a data-backed blueprint on how educators, curriculum developers and policymakers can integrate AI into education to prepare engineers to work with it effectively and responsibly.”
Students Increasingly Using AI To Go After More Financial Aid
Inside Higher Ed (7/23, Alonso) reports financial aid administrators “are seeing students and families use AI in a new way: to try to get more financial aid.” Financial aid officers “say they’ve begun to receive” professional judgment “appeals that appear to be written by AI, based on the generic language they use and the vagueness of the requests.” Inside Higher Ed adds, “The rise in AI-written appeals highlights the ways the chatbots represent a double-edged sword for financial aid offices.”
dtau...@gmail.com
Scientists at Leading AI Companies Ask U.S. for Tools to Pace AI Development
More than 1,000 employees at leading AI companies, including OpenAI, Anthropic, and Google DeepMind, signed a statement urging the U.S. government to support international efforts to deliberately slow the pace of advanced AI development. The group wants technical and policy tools that allow society to slow development if risks escalate. Their concerns center on AI systems potentially automating their own research and improvement, creating challenges for oversight and control.
[ » Read full article ]
NBC News; Jared Perlo (July 28, 2026)
AI Cracks Post-Quantum Cipher in 60 Hours After Two Years of Human Review Failed
Anthropic's restricted frontier AI model, Claude Mythos Preview, accomplished in 60 hours what two years of expert human cryptanalysis could not: identifying a structural flaw in HAWK, a post-quantum digital signature candidate. The AI also created a new attack technique called the Möbius Bridge, which improves analysis of a weakened AES variant.
[ » Read full article ]
Tech Times; Daniel Butler (July 28, 2026)
OpenAI Offers 100,000 Academics Free ChatGPT Access
OpenAI will provide 100,000 academic researchers free access to its most advanced AI models through 2027, supporting work in fields such as biology, chemistry, engineering, mathematics, physics, and computer science. Participants will receive one year of access comparable to a $200-per-month ChatGPT Pro subscription, along with business-grade privacy protections, and their data will not be used for model training. The company has committed to spending more than $250 million through next year to support scientific research and discovery.
[ » Read full article ]
Axios; Ina Fried (July 29, 2026)
EU Lays Out $11.4 Billion for 7 AI Gigafactories
The European Union has launched a €10 billion (U.S.$11.4 billion) initiative to help build seven AI gigafactories, aiming to attract an additional €20 billion in private investment and strengthen Europe's competitiveness against the U.S. and China. Each facility will house at least 100,000 advanced AI chips, making them four times as powerful as current EU datacenters. The plan supports the EU’s push for greater technological sovereignty amid concerns about reliance on foreign cloud and AI providers.
[ » Read full article ]
Associated Press; Sam McNeil (July 30, 2026)
AI on Pace to Find Twice as Many Cyber Flaws in 2026 as It Did in 2025
AI systems are driving a sharp increase in software vulnerabilities discovered so far this year, with the U.S. National Vulnerability Database recording more than 45,000 flaws by late July, approaching the total number of vulnerabilities discovered in full-year 2025. Companies including Oracle, Microsoft, and Google have reported record numbers of security fixes, as AI tools help identify weaknesses faster. Experts note many flaws were found internally through companies’ own AI-powered security tools.
[ » Read full article *May Require Paid Registration ]
Bloomberg; Patrick Howell O'Neill (July 27, 2026)
Microsoft Escalates AI Security Race
Microsoft has unveiled an AI-powered cybersecurity platform designed to counter increasingly sophisticated AI-driven cyberattacks. Scheduled for public preview to begin on Aug. 3, Project Perception coordinates specialized AI agent teams to identify vulnerabilities, prioritize threats, and automatically implement fixes. In a post on X, Microsoft CEO Satya Nadella described the initiative as an example of how the company can get better results per dollar by not locking its security systems to a single AI model family.
[ » Read full article ]
GeekWire; Todd Bishop (July 27, 2026)
Tech Giants Launch AI Safety Initiative
Nvidia and other tech giants on Monday launched the Open Secure AI Alliance to develop and share open AI security tools, following a cyberattack in which rogue OpenAI models targeted Hugging Face. The incident highlighted the limitations of closed AI systems, prompting Hugging Face to rely on a self-hosted Chinese open-weight model for defense. Alliance members argue that open, inspectable AI models are essential for effective cybersecurity, allowing organizations to adapt and deploy defenses quickly.
[ » Read full article ]
CNBC; Kai Nicol-Schwarz (July 27, 2026)
Big Tech Defends Open-Weight AI Models
In an open letter published Friday, more than 20 major technology companies, including Microsoft, Meta, Nvidia, Dell, and Palantir, urged the Trump administration to preserve access to open-weight AI models, warning that restricting them would weaken U.S. competitiveness and push businesses toward lower-cost Chinese alternatives. The coalition argued that open-weight models foster innovation, competition, and adoption. The appeal comes as the White House considers banning Chinese open-weight models over national security and alleged AI distillation attacks.
[ » Read full article ]
Politico; Katherine Long (July 24, 2026)
Congress Hasn't Passed AI Rules for Schools, So Teens Did
A group of 98 high school students representing all 50 states drafted and passed the “Students First Act,” a model framework to guide AI use in K-12 schools. The proposal recommends teaching AI literacy when students first use classroom devices, banning AI on graded tests, permitting limited AI use for brainstorming and editing, and requiring students to disclose AI assistance. It also calls for teacher AI policies, human review of suspected AI misuse, and safeguards for student well-being.
[ » Read full article ]
NPR; Cory Turner (July 30, 2026)
NSF Announces First CyberAICorps Scholarship for Service Program Awards
The U.S. National Science Foundation (NSF) announced the first awards under its CyberAICorps Scholarship for Service program, aimed at expanding cybersecurity education to prepare students for the growing intersection of AI and cyber defense. Fourteen colleges and universities were named to receive funding to develop interdisciplinary curricula, hands-on research, internships, and partnerships that train students to secure AI systems, strengthen cybersecurity operations, and protect critical infrastructure.
[ » Read full article ]
NSF News (July 28, 2026)
Tool Identifies Sources of Fake Videos
Researchers at the University of California, Riverside have developed a forensic tool that detects fake videos and identifies which AI model created them. Developed in collaboration with researchers at YouTube and Google DeepMind, SAGA (Source Attribution of Generative AI Videos) analyzes unique visual and temporal "fingerprints" left by different video generators, which allows it to distinguish among 19 AI video generation systems, identify whether videos were created from text or images, and even trace them to specific development teams.
[ » Read full article ]
UC Riverside News; David Danelski (July 24, 2026)
NSF Announces $400M Investment in Network of AI-Programmable Cloud Labs
The U.S. National Science Foundation (NSF) announced a $380 million investment to create a nationwide network of 20 AI-powered programmable cloud laboratories, matched by upwards of $20 million in philanthropic contributions from the Astera Institute, a California-based nonprofit research organization. The initiative aims to accelerate automated scientific discovery by enabling researchers to remotely run AI-driven experiments across fields including biotechnology, chemistry, materials science, and electronics. The program supports the U.S. government’s Genesis Mission and AI Action Plan goals.
[ » Read full article ]
U.S. National Science Foundation (July 22, 2026)
The Impending, Inescapable Deluge of AI
The number of AI chips worldwide is expected to grow from about 20 million now to roughly 200 million by the end of 2028. Technology companies are investing hundreds of billions of dollars in massive datacenters, betting that greater computing capacity will unlock major advances in scientific research, robotics, automation, and AI capabilities. Leading AI labs are pursuing recursive self-improvement, which would allow AI systems to accelerate their own development with minimal human input.
[ » Read full article *May Require Paid Registration ]
The New York Times; Adam Satariano; Paul Mozur; Jacqueline Gu (July 29, 2026); et al.
AI Companies Spend Record Sums on Washington Lobbying
AI companies are dramatically increasing their Washington lobbying efforts as they seek to influence upcoming regulations on model releases, datacenter development, energy use, and AI security requirements. Collectively, tech and AI firms spent $109 million on such efforts last year, and 11 companies already have spent $41 million on lobbying in the first half of 2026. The companies hope to influence legislation that will determine AI safety standards, export controls, copyright disputes, cybersecurity frameworks, and more.
[ » Read full article *May Require Paid Registration ]
Financial Times; Michael Taffe (July 27, 2026)
Google Builds an AI Fence Around the Internet
Google’s AI-powered search features are transforming how people find information by replacing traditional lists of links with conversational answers, reducing the need to visit external websites. While users spend more time in AI Mode, publishers, businesses, and other organizations report significant declines in search traffic, raising concerns about the future of the open Web.
[ » Read full article *May Require Paid Registration ]
The New York Times; Kate Conger (July 20, 2026)
U.S. Tech Companies Cut 140,000 Jobs So Far This Year
U.S. technology companies have eliminated nearly 140,000 jobs since the start of 2026, even as they dramatically increase spending on AI. According to an analysis of company filings and data from executive outplacement firm Challenger, Gray and Christmas, the tech sector accounts for more than one-third of all announced U.S. layoffs. The broader U.S. labor market remains relatively strong, and AI-focused companies such as OpenAI and Anthropic continue hiring aggressively.
[ » Read full article *May Require Paid Registration ]
Financial Times; Rafe Rosner-Uddin; Clara Murray (July 24, 2026)
dtau...@gmail.com
Scientists Make First Viruses Designed by AI
Viruses designed by AI have, for the first time, been developed by researchers at Stanford University, engineering bacteriophages that successfully killed antibiotic-resistant E. coli in laboratory tests using genome language models trained on millions of bacteriophage sequences. The researchers intentionally excluded human, animal, and plant viruses from the training data to reduce the risk of it designing dangerous viruses.
[ » Read full article ]
The Guardian (U.K.); Ian Sample (August 6, 2026)
Alternative Ars Technica article: Large genome models used to design new viruses - Ars Technica
Jeff Dean, Sanjay Ghemawat Depart Google to Co-Found Discovery Loop
ACM Fellows Jeff Dean (pictured) and Sanjay Ghemawat, architects of Google’s core distributed computing systems, have left the company after 27 years to launch Discovery Loop, an AI startup focused on automating scientific and engineering research. Both received ACM's 2012 Infosys Foundation Award for creating "the first software designs for systems that harness the power of tens of thousands of computers." Alphabet is backing the venture as both an investor and cloud provider.
[ » Read full article ]
Tech Times; Scott McCain (August 5, 2026)
LeCun Joins AI Investment Firm 224 Ventures
ACM A.M. Turing Award laureate Yann LeCun has joined AI-focused investment firm 224 Ventures, which is launching with more than $100 million in assets under management. The firm will invest in early-stage AI applications, robotics, and infrastructure, with a Google DeepMind researcher also joining the team.
[ » Read full article ]
citybiz (August 5, 2026)
ACM AI Leadership Summit To Examine Responsible AI
ACM will host its inaugural AI Leadership Summit from Aug. 30 to Sept. 2 in Atlanta, bringing together 75 speakers from computing and related fields to examine responsible AI development and deployment. The event will address AI’s implications for science, society, governance, education, creativity, professional standards, and sustainability. ACM said the summit is intended to build shared understanding, strengthen collaboration across research communities, connect innovation with responsibility, and help shape future work in AI research, policy, and education.
[ » Read full article ]
Communications of the ACM; Elisa Bertino; Wendy Hall; Neha Kumar (August 3, 2026); et al.
Google Pauses AI Satellite Images over Deepfake Fears
Google removed a new Google Earth feature that allowed users to generate AI-created satellite images a day after it was released, after experts warned it could fuel misinformation and undermine trust in geospatial evidence. Researchers quickly demonstrated the tool could create convincing but fictional images of disasters, military strikes, and humanitarian crises, raising concerns about its misuse in news reporting and conflict verification. The company said it will reintroduce the feature only after implementing stronger guardrails.
[ » Read full article ]
NPR; Geoff Brumfiel (July 31, 2026)
Censorship in Chinese AI Models Can Be Undone, New Research Shows
New research suggests that censorship embedded in Chinese open-source AI models may not automatically transfer to models built from them. Researchers at San Francisco AI lab CTGT found that an AI model trained through distillation from DeepSeek produced more open responses on politically sensitive topics than the original model. The research comes amid an active debate inside the Trump administration over how to handle the risk posed by Chinese open-source models.
[ » Read full article ]
Semafor; Reed Albergotti (July 29, 2026)
Creating More Realistic Multi-Person Images
An AI image-generation method devised by Cornell University researchers creates more realistic multi-person scenes by automatically incorporating each person’s pose and interactions. Presented at SIGGRAPH 2026, the new approach builds scenes iteratively, generating one person at a time and using each pose to guide the next. Using the FLUX image generator and the “Who’s Waldo” dataset, researchers trained the model to better understand relationships and actions between people.
[ » Read full article ]
Cornell Chronicle; Tom Fleischman (July 28, 2026)
Optical Tech Could Update a Robot’s AI on the Fly
An optical receiver developed by Cornell Tech researchers uses QR code-like patterns of light to directly modify computer memory. The technology replaces energy-intensive electrical connections and analog conversion circuits with digital optical communication, allowing AI model data to be transmitted directly into processor memory using light. The system could help address growing memory demands in AI datacenters, autonomous vehicles, and edge devices such as robots.
[ » Read full article ]
IEEE Spectrum; Alex Music (July 27, 2026)
Researchers Personalize the Sense of Touch
University of Southern California Viterbi School of Engineering researchers developed an AI-powered algorithm that personalizes haptic feedback by learning users’ preferred vibration patterns through as few as 40 pairwise comparisons, significantly reducing the effort required by traditional methods. The Vibrotactile Preference Learning system accounts for individual differences in how people perceive touch and can tailor vibrations for specific emotions or functions, with potential applications in mental health, gaming, virtual reality, automotive safety, and digital communication through customized “haptic emojis.”
[ » Read full article ]
USC Viterbi School of Engineering; Venice Tang (July 27, 2026)
China's New Challenge as Natural Disasters Strike: Fake AI Videos
China is grappling with a surge of AI-generated misinformation during natural disasters, as fake videos depicting floods, casualties, power outages, and wildlife escapes spread rapidly online, prompting public panic and complicating emergency response efforts. Authorities have responded with arrests, stricter enforcement, and a nationwide crackdown on disaster-related misinformation, while experts warn that AI-generated false content is evolving faster than detection technology and legal frameworks.
[ » Read full article ]
BBC News; Stephen McDonell (July 27, 2026)
EU Tells Firms to Label AI-Generated Content
European Union (EU) AI transparency rules took effect Sunday, requiring companies to label all AI-generated content. Businesses must use watermarks or other detection markers and ensure AI systems clearly identify themselves, with fines possible for noncompliance. The EU says the rules aim to combat misinformation and help citizens distinguish real content from AI-generated material.
[ » Read full article ]
Le Monde (France) (July 28, 2026)
White House AI Guidelines Exempt U.S. Open Models from Government Review
New AI guidelines from the White House exempt U.S.-developed open-weight AI models from voluntary government pre-release safety testing, focusing instead on the most advanced closed, proprietary models with cutting-edge cybersecurity and hacking capabilities. Officials said open AI models could face review as they become more powerful. The framework aims to balance AI safety with innovation and U.S. competitiveness against China, though critics argue the voluntary system may leave security gaps and create inconsistent oversight.
[ » Read full article *May Require Paid Registration ]
The Wall Street Journal; Amrith Ramkumar (August 4, 2026)
AI Fuels More Than Half of Africa’s Cybercrime
AI is fueling more than half of cybercrime in Africa, according to an Interpol study of 36 countries. The report found cybercrime caused at least $5 billion in direct economic damage across the continent in 2025, while reported financial losses more than doubled to $484 million. Interpol said AI is being used to automate reconnaissance, phishing, extortion, and evasion, while AI-generated synthetic identities are helping criminals bypass biometric verification systems to open bank accounts, obtain loans, and register SIM cards.
[ » Read full article *May Require Paid Registration ]
Bloomberg; Ruth Olurounbi (August 4, 2026)
The Million-Dollar Talent Wars for Young Math Geniuses
Elite quantitative trading firms and AI companies are competing aggressively for a small pool of top mathematical and machine-learning talent, driving early-career compensation into the millions. “Quant” firms, which rely on advanced algorithms, are using higher pay, internships, and specialized cultures to attract graduates who can generate significant profits. AI labs such as OpenAI and Anthropic are intensifying the competition by seeking similar expertise. Entry-level packages can exceed $1 million, while firms increasingly recruit students before graduation.
[ » Read full article *May Require Paid Registration ]
The Wall Street Journal; Julia Amann (July 30, 2026)
Monkeys Come Face-to-Face with AI in a Costa Rican Forest
Researchers at Emory University and the Georgia Institute of Technology developed an AI-powered testing station that uses facial recognition to identify wild white-faced capuchin monkeys and measure their cognitive abilities through touchscreen tasks rewarded with food. In initial field tests in Costa Rica’s Taboga Forest Reserve, 10 of 16 monkeys learned to use the CapuchinAI device, demonstrating its potential to study the animals’ learning, memory, impulse control, and decision-making in natural environments.
[ » Read full article *May Require Paid Registration ]
The New York Times; Emily Anthes (July 28, 2026)
Colleges And Universities Pivoting To Offer AI Training Across Disciplines As Student Demand Shifts
The AP (8/3, Hollingsworth) reports professors at colleges across the US “are finding themselves busier than ever teaching students from a range of majors about artificial intelligence.” Colleges and universities “are responding to changes in student demand, but they also recognize that new graduates – regardless of their field – are facing questions about their AI skills from potential employers.” At Eastern Mennonite University, for example, faculty members in computer science and music will co-teach a course “at the intersection of art and technology” for students in programs including electrical engineering, math, and music. Other colleges “are standing up AI minors targeting students outside computer science. There’s a new AI graduation requirement at Purdue University. In freshman writing classes, Harvard University is teaching students how large language models work, plus exploring AI topics such as copyright and disinformation.” Ohio State University now includes “an AI fluency requirement that includes hands-on workshops.”
AI Driving Some Professors Into Early Retirement
The Chronicle of Higher Education (8/4, Lu) reports that some university professors are retiring earlier than planned due to frustrations with generative AI. These professors feel overwhelmed by student cheating and question the value of teaching students who seem uninvested in learning. A study by researcher Geoffrey Sanzenbacher found that after ChatGPT’s November 2022 introduction, job exits for postsecondary educators 55 and older increased from 11% to 16%.
NSF Rolls Out $100M Initiative To Expand AI Compute Access For Researchers, Students And Educators
ExecutiveGov (8/5, Bennet) reports the National Science Foundation launched on Tuesday “a $100 million initiative aimed at closing gaps in AI compute access for researchers, students and educators nationwide.” The new State and Regional Artificial Intelligence Infrastructure Hubs program will finance “partnerships that bring together academic institutions, state and local governments, industry and philanthropic organizations to build and operate advanced scientific computing capacity.” The agency “said the effort is designed to support both research and workforce development, addressing what it described as highly uneven access to AI infrastructure across different parts of the country as computational demands for scientific work continue to grow.” Under the new initiative, “NSF will fund the creation of AI Infrastructure Hubs – state-level or multistate regional consortia that pool contributions from research institutions, philanthropic groups, local governments and private companies to stand up shared computing resources. The agency plans to make one award per state or region, with up to 10 hubs funded in this initial round.”
Turnitin Debuts AI Detector Featuring Keystroke Tracking
The Chronicle of Higher Education (8/5, Crosnoe) reports ed-tech company Turnitin has launched new software to combat student AI use by monitoring typing, revisions, and copy-paste functions. The tool, dubbed Clarity, features a text-only detector that “works in conjunction with its keystroke-tracking and screen-monitoring features to give instructors a fuller picture of AI use.” The launch of Clarity comes after several universities, including Yale and Vanderbilt, had disabled Turnitin’s older AI detector over false positive concerns.