Wednesday, September 30, 2026

Recursive Self-Improvement of AI Without Comprehensive Review and Alignment - Ray Schroeder, Inside Higher Ed

In the high-tech, high-competition AI field, we are seeing new versions of models rolled out by the frontier companies (OpenAI, Google and Anthropic) every month and a half. I included a Release Cadence Breakdown report by Google Gemini in the “Online: Trending Now” column just a few weeks ago. The speed and intensity of the research and development required to keep up such a release schedule has taken a toll on the necessary, thorough testing of reliability and alignment that we all expect must be done by reliable creators. It is important to remember that this field is breaking new ground in creating models that can generate original concepts, often by mimicking approaches humans have taken in the past. Humans, of course, are not always the best models of safe, generous and legal behaviors. That’s where the problems arise.

AI in education initiatives emphasize experimentation, balanced use - Susan Kelley, Cornell Chronicle

Two new initiatives involving artificial intelligence (AI) in Cornell classrooms will encourage experimental and balanced use, while exploring ways to deploy AI that align with the teaching needs and strategies of individual academic fields. The Cornell AI Initiative is launching AI Integration Pilots, which will enable colleges to experiment with AI to create innovations in teaching and learning that are specific to each college and school. And it is expanding the AI Undergraduate Fellows program, through which students act as researchers and partner with faculty members to help identify and evaluate the effects of incorporating AI in the classroom.

Tuesday, September 29, 2026

OpenAI Says It Will Not Release Newest A.I. Model Over Safety Concerns - Sheera Frenkel, NY Times

OpenAI said on Monday that it would not release its newest artificial intelligence model because of security concerns raised by its researchers, in the company’s latest move to slow down the pace of its technology. During the testing phase for the new model, known as GPT-6.1 Astra, it showed high levels of what the company saw as deception, or a willingness to mislead users about its actions. The model was also willing to go beyond the original scope of what it was asked to do, without checking back for directions or instructions. “For anything regarding safety and alignment, there’s a trade-off,” said Saachi Jain, the head of safety systems at OpenAI. The new model “didn’t quite meet the bar in terms of staying within scope and authorization, and how it communicates back to the user about the type of work it’s done.”

https://www.nytimes.com/2026/09/28/technology/openai-astra-safety.html?unlocked_article_code=1.E1E.FTjG.aTIQEjFG3csX&smid=nytcore-android-share 

How should AI be regulated? - Claire Goudreau, Johns Hopkins University

Gillian Hadfield | Bloomberg Distinguished Professor of AI Alignment and Governance said "The key thing that happened was humans stopped writing all of the rules about how the computer was supposed to behave and figured out a method called machine learning where the machine, the computer, would write its own program. People will talk about guardrails, safety. We don't know what those should be. Technology moves faster than governments do. That's the pacing problem and the reason to slow down. ... We absolutely need our research universities to be taking this on as a mission because this is where the brain trust needs to be for how we figure out how to make this go well."

#higheredhttps://hub.jhu.edu/2026/09/17/next-conversations-governing-ai/

Monday, September 28, 2026

Scientists just made quantum computer operations 1,000 times faster - Science Daily

The new method has the potential to accelerate quantum computations by a factor of a thousand, bringing fault-tolerant quantum computers a significant step closer. Quantum computers remain highly vulnerable to errors and tiny disturbances from their surroundings. The longer a quantum operation takes to complete, the more time there is for those errors to build up. Researchers at Chalmers University of Technology in Sweden have now developed a method that can perform a broad range of advanced quantum operations more than a thousand times faster. The advance tackles a major obstacle in the field and could help move quantum computing closer to becoming fault-tolerant.

Quantum Computing Inc. to Showcase Quantum Secure Communications Beyond Quantum Key Distribution at ECOC 2026 - PR Newswire

As organizations prepare for the post-quantum era, the challenge is no longer understanding the value of quantum-secure communications, but overcoming the barriers that have limited deployment. Traditional QKD systems often face challenges associated with scalability, operational complexity, infrastructure cost, and dependence on classical authentication methods. As a result, many quantum communication technologies have remained confined to specialized research environments. QCi's entanglement-based quantum security platform is designed to overcome these limitations by bringing two critical elements of secure communications - encryption and authentication - together on a single platform. For the first time, QCi will showcase an integrated quantum security solution that combines quantum key distribution (QKD) with hardware-based Quantum Identity Authentication (QIA), demonstrating how quantum technologies can address both the secure exchange of encryption keys and the fundamental need to establish trusted identities between communicating parties.

Sunday, September 27, 2026

UW-Madison dean discusses AI's rapid rise, potential regulations - Lucas Kihmm, WKOW

U.S. lawmakers in Washington, D.C., are debating whether and how to regulate artificial intelligence. Some AI experts are sounding the alarm about how quickly the technology is advancing, calling for more guardrails. UW-Madison's new College of Computing and Artificial Intelligence says its mission is to create powerful technology that best serves the people who use it. At Morgridge Hall, nearly every table is filled with people working on computers or phones. It's fitting for a building dedicated to innovation. Remi Arpaci-Dusseau is the founding dean of the college. He says the rise of AI is one reason the college was created.

At Georgetown University dialogue, future of artificial intelligence seen as ‘something we have to tackle together’ - Mark Zimmermann, Catholic Standard

In a week when tech leaders said the artificial intelligence industry needs to slow its race to an uncertain future that some warn may be apocalyptic, Georgetown University hosted a Sept. 9 discussion on “AI, Catholic Social Teaching, and Pope Leo’s Invitation to Dialogue with the World.” The dialogue, sponsored by Georgetown’s Initiative on Catholic Social Thought and Public Life, drew upon Magnifica Humanitas, Pope Leo XIV’s May 2026 encyclical letter that was subtitled “On Safeguarding the Human Person in the Time of Artificial Intelligence.” Speakers emphasized that the pope’s letter offered a clarion call for moral guidance and reflection among all stakeholders, including community members, on the best way to move forward with AI.

Saturday, September 26, 2026

Graduating in AI Era Is Like ‘Large Recession’ For Starting Pay - Michael Sasso, Bloomberg

The rise of artificial intelligence has coincided with a significant decline in both initial earnings and employment among graduates with the most-exposed college majors, a worrisome trend that could shape career outcomes for years afterward, according to a new study. Economists at the US Census Bureau examined data covering roughly 29% of US bachelor’s degrees issued between 2016 and 2024, finding notable effects on new graduates after ChatGPT’s debut in November 2022.

https://www.bloomberg.com/news/articles/2026-09-14/ai-era-coincides-with-weak-pay-and-employment-for-college-graduates

An M.I.T. Report Warns A.I. Is Causing ‘Cognitive Surrender.’ Universities Are in a Bind. - Mark ArsenaultDana GoldsteinAlan Blinder and Sarah Mervosh; NY Times

As A.I. leaders warn about the dangers of their own technology, the dissonance between the optimism of university leaders and the fears of many students and professors may increase. Professors and students are often in a bind about what they should do: Use more A.I. to stay relevant, or dial back to preserve learning and thinking? While some politicians and A.I. leaders are warning about possible doomsday scenarios for humanity as A.I. models advance, many campus leaders are focused on the existential threat the technology poses to universities.



Friday, September 25, 2026

AI Agents Are Quietly Finishing Entire College Degrees - Julia McCoy, YouTube

Julia McCoy highlights how students are handing credentials to autonomous AI agents to complete entire courses—from watching lectures to submitting quizzes, writing papers, and replying to discussion threads [00:08]. She argues that if a machine can finish an entire course load, higher education has largely been measuring mechanical "output" [00:21], which machines now produce effortlessly. Because AI can draft text but cannot exercise true critical judgment [00:28]—something rarely formalized on a syllabus—she contends that the real challenge facing universities isn't whether institutions can stop AI tools, but reconsidering what students are actually there to learn [00:35]. [summary provided in part by Gemini Flash]

https://www.youtube.com/watch?v=PdFggS28r2M

We're not about to awaken an evil digital god — AI is far more likely to end in a great disappointment - Andrew Rogoyski, Live Science

Artificial intelligence (AI) terrifies us partly because it is new and partly because it is disturbingly familiar. We have invented machines that converse, flatter, argue and improvise with uncanny fluency. Confronted by this synthetic reflection, we behave like early humans seeing a mirror for the first time: We project consciousness behind the glass, then populate it with our expected behaviors, emotions, hopes and nightmares. In other words, we scare ourselves. Today's frontier models are impressively good at writing, coding and math. Yet they still fail at ingredients of useful artificial general intelligence (AGI), a system that matches or surpasses human capabilities such as consistent reasoning, long-term planning, factual consistency and reliable performance outside their "experience" (the data they were trained on). 

Thursday, September 24, 2026

The state of open source AI.- Mozilla

Te reo Māori has no commercial market to speak of. A broadcaster in the far north of New Zealand has been building speech models for it anyway, under a licence written so the recordings stay with the communities that gave them. Farmers in East Africa point a phone at a cassava leaf and get a diagnosis back with no signal at all, from a model small enough to sit on the handset. In Switzerland, a public consortium trained a national model on public supercomputers and released all of it: weights, data, training code. None of them asked permission, and none of them could have rented this. They own it, and that is the whole idea. Read what follows as a map: where open AI is winning, where some numbers surprised even us, and where it is exposed. A case that hides its weak points is an advertisement. In the six weeks since our first draft, a major US lab returned to open weights and Washington debated a ban and declined it. The map moved while we drew it. We have been here before. Mozilla exists because one company tried to own the front door to the web, and an open community made sure it never could. 

From Search to Strategy: What Student AI Use Means for the Future of Academic Libraries - EDUCAUSE

While many academic librarians worry that artificial intelligence (AI) may reduce student engagement, research suggests a different story: students are integrating AI into their research workflows while continuing to rely on libraries for trusted sources and deeper scholarly inquiry. This interview features: Amy Deschenes, Interim Director of UX and Discovery, Harvard University and Meg McMahon, User Experience Researcher, Harvard University.

https://www.youtube.com/watch?v=g5TFf44BekM

Wednesday, September 23, 2026

Can AI Help Colleges Reach Struggling Students Earlier? - Joshua Bay, Inside Higher Ed

Austin Community College is using artificial intelligence and real-time data to identify students who need help and connect them with support before problems escalate. Students often leave clues that they’re struggling before they stop attending classes or exit college altogether. But those warning signs can be scattered across different offices and data systems, making it difficult for college staff to see the full picture or intervene before a student reaches a crisis point. As colleges increasingly experiment with artificial intelligence, some are exploring whether the technology can help connect those dots and make student support more proactive. Austin Community College, for one, is testing a new initiative designed to identify student needs earlier.

Classroom practices that encourage intelligent AI behaviours - Nada Kakabadse, University World News

Whatever assessment policies say, the unsupervised take-home essay as a test of unaided student writing is gone. The institutional response has been retreating. The universities of Melbourne and Toronto are phasing out unsupervised take-home essays. The University of Cape Town’s AI working group wants 40% of assessments invigilated by 2027. At Durham University, a professor of philosophy stood down as chair of his board of examiners, citing a marking crisis. But retreat is only one of three institutional responses now visible worldwide, and comparing them is instructive. The second response, arguably the least defensible, is detection-led enforcement: running submissions through AI detectors and prosecuting the flagged. This approach is quietly collapsing.

Tuesday, September 22, 2026

The new bargaining frontier: AI and academic labour - James Yoonil Auh, University World News

 What happens when AI begins to perform work that universities have historically employed people to do? At that point, AI ceases to be merely an educational technology and becomes a labour technology.
 Once AI enters the employment relationship, questions arise that an information technology office alone cannot resolve. Who decides whether AI may perform work previously assigned to faculty? Can AI change staffing requirements? Who owns academic materials used to develop AI systems? Can algorithmic analysis influence faculty evaluations? If AI makes academic work faster, who benefits from the time saved? If supervising AI creates new work, where does that labour appear in workload calculations? These questions are no longer hypothetical. They are increasingly appearing in collective agreements, memoranda of understanding and union policies.

Anthropic C.E.O. Calls for A.I. Slowdown - Mike Isaac, NY Times

The chief executive of Anthropic called for a global slowdown of artificial intelligence development in a 3,800-word essay on Saturday, just days after one of the company’s employees quit over concerns about the safety of the technology. Dario Amodei, who co-founded Anthropic to focus on securely and carefully building A.I., wrote that while he believed the technology could bring many benefits, it was advancing at too quick a pace for researchers to continue safely. “Over the last few months, I have become convinced that fully addressing the risks requires even more prudence — not just investing in risk prevention, but pacing the rate of capabilities advancement so that risk prevention has time to keep up,” Mr. Amodei said. “We must slow the pace at which we improve the capabilities of A.I. models. Progress will still seem fast, and we must make wise use of the time we gain.”

Monday, September 21, 2026

USF president shares student successes and plans for AI college during fall address - Emma Brisko, WUSF

USF reached a record-breaking $750 million in research funding in fiscal 2025. Some of that money is being invested in the planned Bellini College of Artificial Intelligence, Cybersecurity and Computing. Limayem shared new renderings of the building, which will be located near the entrance of the Tampa campus, at Leroy Collins Boulevard and USF Alumni Drive. No construction date has been set, but $35 million in state money has been committed to the project. “This is an important piece of USF's future,” Limayem said. “We envision a place designed for students, faculty, research and collaboration in fields that are reshaping nearly every part of our lives – and protecting our digital spaces.”

Are Colleges Ready To Support the AI Workforce? - Grace Alexis, EdTech

For decades, colleges prepared students for a predictable workforce. Students chose a major, learned a defined set of skills and entered a career path that changed slowly over time. Artificial intelligence has ended that arrangement. Every few weeks, another report warns that AI will eliminate jobs. A recent Graduate Management Admission Council survey of more than 600 corporate recruiters found that one-third of employers are already replacing some entry-level roles with AI. In technology, that number rises to 40%. Yet the same survey named proficiency with AI tools the single most valuable skill employers expect to prize five years from now — and one of the skills graduates are least prepared to demonstrate. That gap is an education problem, and it belongs to us.

Sunday, September 20, 2026

Americans Use Artificial Intelligence, but They Don’t Want It in Charge - Rutgers

A survey led by Rutgers researchers finds AI use is surging, but Americans still prefer humans making consequential decisions. Artificial intelligence (AI) is becoming part of everyday life in the U.S., but Americans are far less comfortable when it shifts from something they use to something that makes decisions about them, according to a new survey led by Rutgers researchers. Nearly three-quarters of U.S. adults (72%) have used at least one major AI tool, up from 50% in November 2024, when researchers conducted an earlier survey of 4,767 people. The findings come from the latest report from the National AI Opinion Monitor, a project led by Rutgers researchers that tracks public attitudes toward AI. More than a quarter of Americans (27%) use AI every day, yet fewer than one in 10 would allow it to make the final call without human supervision on hiring, loans, college admission, parole or priority for a medical procedure. Between 45% and 54% say AI shouldn’t be used at all in those decisions.

https://www.rutgers.edu/news/americans-use-artificial-intelligence-they-dont-want-it-charge

University of North Texas to Open College of AI, Analytics - Milla Surjadi, The Dallas Morning News

The University of North Texas will establish a college of artificial intelligence and advanced analytics after receiving a $20 million gift, it announced Tuesday. The philanthropic investment comes from Vikas Sinha, a UNT alumnus and vice president of the mainframe software division at Broadcom, a semiconductor and infrastructure manufacturer, and his wife, Anuradha. As UNT looks to strengthen its artificial intelligence education, the new college will help advance research and better prepare students for a changing workforce, university officials said in a release.

Saturday, September 19, 2026

Productive Friction: What Universities Must Protect in the Age of AI - Vladimir Bratic, Faculty Focus

AI has now become a part of everyday life on college campuses and I have embraced it in all my classes at every level. I now encourage my students to use NotebookLM when the text becomes dense or incomprehensible. I encourage the students to upload assigned text to create timelines, flashcards for studying and video summaries of their reading. Students can now process text faster, research more effectively and edit their work without having to go to the writing center. These gains are substantial and any serious conversation about AI in education must begin with acknowledging the benefits of AI. 

Business, government and education leaders try to reinvent college from scratch - Jon Marcus, Hechinger Report

Called the Greenway Institute — “green” as in new — the project will adopt instructional methods proven to work well, such as real-world training under the supervision of faculty mentors, while stripping away the frills, distractions and layers of bureaucracy that make college in America so expensive. “It’s not fancy. It’s a place where you do have a meaningful relationship with your faculty members. But you also take out your own trash,” said Mark Somerville, the founding president. Even more unexpected than the location of this startup is the timing. Other colleges and universities are closing or at risk of closing. In Vermont alone, seven private colleges have shut down in the last 10 years. But Somerville, the former provost of Olin College of Engineering, said this “moment of mass extinction” makes it clearer than ever that “we need new models in higher ed.” 

Friday, September 18, 2026

A new Anthropic model seeks to test how AI could impact the U.S. economy. - Scott Horsley, NPR

Will artificial intelligence light a fire under the U.S. economy in the coming years? That's been a big question in the field of economics, and now Anthropic — the company behind the popular Claude AI assistant — is taking a stab at helping people find their own answers. The company has created an interactive tool that lets users test their assumptions about how productive — or disruptive — the AI boom might be. In one scenario, artificial intelligence provides a gentle boost to the economy, lifting growth rates with little effect on workers. At the other extreme, GDP soars, but so does unemployment, with nearly 14% of workers losing their jobs to AI and less than half of them finding new ones.

OpenAI Says It Has Cracked One of Math’s ‘Millennium Problems’ - Cade Metz, NY Times

OpenAI said on Tuesday that its newest artificial intelligence technology had solved one of the “Millennium Problems,” a collection of important unanswered math questions meant to push the world’s leading mathematicians to new heights. The announcement is another clear sign that A.I. is fundamentally changing the upper reaches of mathematics, which have long been viewed as a pinnacle of human achievement. The change has excited some mathematicians, while stirring concern among others. Over the past year, A.I. systems successfully solved a wide range of problems that have bedeviled mathematicians for decades. But these problems were not as complex, nor as closely watched, as the one that OpenAI’s technology has solved over the past several days. The Millennium Problems are among the most heavily researched in the field.

Thursday, September 17, 2026

Can AI lead us from industrial to ecological intelligence? - James Yoonil Auh, University World News

AI is not inherently ecological. It is a civilisational amplifier. Attached to an extractive economy, it will accelerate extraction. Directed towards regeneration, it could help societies understand and coordinate environmental relationships previously too complex or fast-moving for human institutions to manage. Its distinctive ecological promise is not that AI is ‘green’. It is that AI can shorten the distance between an ecological signal and an institutional response. The decisive question, therefore, is not whether AI is environmentally good or bad. It is whether we can change the purpose for which intelligence itself is organised.

OpenAI is throwing everything into building a fully automated researcher - Technology Review

But OpenAI is also training its systems to work by themselves for longer by feeding them specific samples of complex tasks, such as hard puzzles taken from math and coding contests, which force the models to learn how to do things like keep track of very large chunks of text and split problems up into (and then manage) multiple subtasks. The aim isn’t to build models that just win math competitions. “That lets you prove that the technology works before you connect it to the real world,” says Pachocki. “If we really wanted to, we could build an amazing automated mathematician. We have all the tools, and I think it would be relatively easy. But it’s not something we’re going to prioritize now because, you know, at the point where you believe you can do it, there’s much more urgent things to do.” “We are much more focused now on research that’s relevant in the real world,” he adds.

Wednesday, September 16, 2026

How AI policies are reshaping college learning and sowing confusion - Leah Duffy, Philadelphia Tribune

In one college classroom, using ChatGPT might be considered as cheating. In another, it can be part of the assignment. For students navigating artificial intelligence in college, the difference can come down to a single syllabus — and who wrote it. At some area universities, students are expected to follow rules established by their individual instructors when using generative AI tools such as ChatGPT. But the instructors differ in how they frame those rules, leaving many students confused. For some, the decision over which instructors will allow AI and those who will not has created confusion over where the line falls between AI as an academic tool or as a violation of academic integrity.

The human skills at a premium in the age of AI - McKinsey

As AI takes rote tasks off employees’ plates, the work that remains can require greater judgment and critical thinking—and be more cognitively demanding. “The rise of AI doesn’t mean we all suddenly delegate our responsibilities for quality control and critical thinking,” says Kate Smaje, McKinsey senior partner and global leader of technology and AI, in a recent episode of McKinsey Talks Talent. “There’s a premium on those capabilities now.”  Smaje, coauthor of the recent book Rewired, joins Senior Partner Brooke Weddle and Partner Bryan Hancock to explore what it takes to manage a workforce of people and AI agents—from reimagining roles and workflows to managing the performance of digital workers.  


Tuesday, September 15, 2026

AI transformations run on trust - Aaron De Smet, Bonnie Dowling, Holly Price, and Ignacio Fantaguzzi; McKinsey

rganizations have spent the past several years racing to deploy AI. Understandably, much of this activity has focused on technology, governance, and new operating models. Yet a crucial ingredient of successful AI transformation has received far less attention: trust. If leaders truly want to transform their organizations with tech and AI, they will need their people to make it happen. Unlike many previous technology transformations, AI can’t be deployed solely from the top down. Real value emerges as employees experiment with new ways of working, identify high-value use cases, redesign workflows, and continually refine how AI is applied in their daily work.

How CEOs are leading through the AI shift - McKinsey

Across industries, AI is changing how companies operate, make decisions, and grow. Through a series of podcast conversations, Senior Partner Eric Kutcher talks with CEOs about how they are navigating that shift and connecting AI to the broader priorities of their businesses. At Invesco, President and CEO Andrew Schlossberg is taking both a top-down and bottom-up approach to AI. The company has put the tools in every employee’s hands, allowing ideas to emerge from people rethinking how they work. The leadership challenge is finding the right balance. “You have to resist the urge to make it completely top down and controlled,” Schlossberg says, “and also resist the urge to make it completely bottom up and entrepreneurial.”

Monday, September 14, 2026

The ‘AI and Future of Work’ Series — How the University is preparing students for what is next - Lidia Zur Muhlen. Cavalier Daily

Questions about the relationship between AI, higher education and the job market have become increasingly relevant to students preparing to enter the workforce. But the effects of AI vary considerably across professions, making it difficult to draw conclusions about what the technology will mean for students pursuing different fields. At the School of Law, professors are making an effort to incorporate AI into legal education while emphasizing that students still need to develop traditional skills such as legal reasoning, judgement and communication. Similarly, at U.Va. Health, AI is already being used to assist with clinical documentation, hospital operations and research, while the system also highlights the importance of human oversight, safety, privacy and ethics as it expands the technology’s use.

University of Glasgow’s new teaching toolkit asks educators when to use AI technology - EdTech Innovation Hub

The University of Glasgow’s School of Education has released a free toolkit designed to help teachers make more deliberate decisions about digital and AI technology, including whether it belongs in a lesson at all. The 92-page Toolkit for Reflecting on the Use of Digital and Artificial Intelligence (AI) Technologies in Learning and Teaching was developed by Dr Mark Peart, Dr Gabriella Rodolico and Leon Robinson, with input from educators, teacher educators, students, education leaders, policymakers and partner organizations. Rather than offering a list of recommended AI tools or prescribing how teachers should use them, the resource takes educators through a structured process of examining their own assumptions, analyzing teaching scenarios, designing technology-supported lessons and eventually carrying out practitioner enquiry into their own practice.

Sunday, September 13, 2026

Why I Let My Students Use AI - Rod Naquin, the Free Press

Unless you’ve been living under a rock for the last four years, you’ve encountered the debate about AI in education. You’ve seen headlines from researchers who claim that using a chatbot is akin to outsourcing your mind to a machine. You’ve heard warnings about cognitive off-loading, cognitive debt, cognitive decline. A few weeks back, I read a commentator describe AI use as “cognitive cancer.” You’ve heard the hand-wringing about how students no longer do their homework (they used to?), no longer write their own essays (they were good at this?), and no longer read their assigned texts (haven’t they been using SparkNotes for years?). If I were living under that rock, and I peered out at this discourse, I might just climb back down and take my chances with the grubs. The problem is, the naysayers have it all wrong. I am a high school English teacher in Louisiana, and I’m here to tell you that while the concerns about the effects of AI on students’ learning are real, to a point, they are also dramatically overstated.

https://www.thefp.com/p/why-i-let-my-students-use-ai

UC Irvine to Establish National Research Center on AI in Writing - GovTech

The Institute of Education Sciences is giving the University of California, Irvine $10 million to conduct research on AI writing tools and test the efficacy of the school’s own PapyrusAI. With $10 million in funding from the U.S. Department of Education’s Institute of Education Sciences, UCI will establish a national hub for information on available AI tools, frameworks for assessing their effectiveness and a research base that measures cause and effect, according to the award announcement. The National Center for Writing Research to Improve Teaching Effectiveness With Generative AI (WRITE AI Center) will conduct research through a survey and targeted pilots. A nationwide survey will pull together information on the 100 most commonly used AI writing tools in higher education, as well as the tools’ instructional applications, costs and infrastructure, and training needs. The center's website will then publish this information, with updates at least once a year.

Saturday, September 12, 2026

Deming Tells Class of 2030 AI Will Bring About a New Age for American Universities - Harvard Crimson

The rise of AI, he argued, is prompting another shift in the role of higher education, as technology makes expertise increasingly accessible and places greater value on collaborative and communal learning. “The internet has made expert content available to anyone in the world,” Deming said. “That content doesn’t replace your education here because it isn’t personalized. Masterful lectures are part of your education, but not the only one. They don’t answer your questions or meet your individualized learning needs.” “But artificial intelligence can be fully personalized,” he said.

Online Learning’s Big AI Test: Who Wins the New Classroom - Global Brands Magazine

Online learning is at a crossroads, and artificial intelligence is holding the map. The same technology that can spit out an essay or solve a problem set in seconds is also powering smarter tutors, faster feedback and more personal courses. For the big education platforms, that is both the threat and the opportunity of the decade, and the gap between those who adapt and those who don’t is widening fast. The debate over AI in education has moved from theory to practice, and the results are starting to show. The early scoreboard is clear. Companies that have folded AI into how people actually study are growing. Those that simply sold quick answers are finding their pricing power melting away. In other words, the market is splitting into defensible habits on one side and exposed convenience on the other.


Friday, September 11, 2026

Technology is here to stay—now let’s get it right - The Acorn

When the LVUSD Board of Education gathers for its very important next meeting on Sept. 15, it should focus less on choosing sides and more on establishing sensible guardrails. Age-appropriate screen limits, greater use of printed materials and handwriting, restrictions on non-educational content, stronger privacy protections, clear AI policies and transparent reporting of classroom screen time would be reasonable places to start. Technology is here to stay. The challenge isn’t eliminating screens. It’s ensuring they never replace the human relationships, critical thinking and inspired teaching that define a great education.


ChatGPT boosts student scores while critical-thinking training broadens ideas - Emma Thompson, Ed Tech Innovation Hub

A new study shows giving university students access to ChatGPT helped them produce higher-scoring, more expert-like work, but a separate critical-thinking intervention changed something the traditional grading rubric largely missed: the range and distinctiveness of their ideas. The randomized controlled trial, conducted by researchers at Bocconi University in collaboration with OpenAI Economic Research, involved 1,053 first-year undergraduates studying economics, management and finance. Students were split into four groups. One received access to ChatGPT Edu using GPT-4o, another completed training in causal reasoning, a third received both, and a control group received neither. They were then given 45 minutes to tackle the same real-world business problem, producing a short consultant-style recommendation for increasing awareness and use of their university’s merchandise among alumni. On the conventional performance measure, ChatGPT made the clearest difference. Students with GPT access scored an estimated 0.86 points higher on a five-point scale than the control group, whose estimated score was 2.09. Their recommendations were also more similar to those produced independently by three domain experts.




Thursday, September 10, 2026

Artificial Intelligence as a Pedagogical Tool in Higher Education: Enhancing Reflection and Formative Assessment - Nancy Marsh and Nan Flickinger, Faculty Focus

The integration of artificial intelligence (AI) in higher education has generated both enthusiasm and apprehension among educators. This article argues that AI can serve as an adaptive pedagogical partner, enhancing formative assessment, reflection, and metacognitive learning. Through illustrative examples, the study demonstrates how AI can provide individualized feedback and foster deeper conceptual understanding. The discussion emphasizes the ethical and pedagogical implications of AI integration, positioning it as a catalyst for reflective and responsible learning practices. The article concludes that AI, when employed intentionally, can amplify human instruction and promote the development of digitally literate, critically engaged learners. The integration of artificial intelligence (AI) into higher education pedagogy is often less formidable than it may initially appear. When educators reconceptualize AI not as a disruptive force but as an adaptive, resource-rich instructional tool, the possibilities for enhancing student learning and engagement expand considerably.


In the age of AI, is math the new CS degree? - Louise Imber, Fast Company

Majoring in computer science used to be considered a safe career bet, ensuring stable pay and work. After the 2008 recession, many students turned to the field’s promising outlook: Between 2008 and 2024, the number of computer science degrees grew by almost 500%, according to The Washington Post. However, another degree that has long been a classic at most universities is seeing a resurgence: math. Mathematics degree enrollment dipped after 2018 and bottomed out in 2022. But since then, its popularity has slowly risen, growing almost 2% by 2024, according to data from the Integrated Postsecondary Education Data System. At the same time, mathematics occupations are projected to grow 28.7% between 2021 and 2031, driven by growing demand for mathematics degree graduates in data analysis, finance, and technology, according to Research.com.

Wednesday, September 09, 2026

Canada Hires 48 Scholars Away From Top U.S. Universities - Vjosa Isai, NY Times

Twenty-one universities in Canada announced on Thursday the hiring of dozens of researchers to join their top academic ranks, poaching the majority of the recruits from the United States. Of the incoming 64 researchers, 48 will leave American campuses including Harvard, Yale and M.I.T., and the remainder are from Europe, Asia and Australia. Turmoil in American academia, precipitated by a crackdown on prestigious universities by the Trump administration and attacks on scientific research groups, has allowed Canada a rare opportunity to lure some of those schools’ top minds north. Their hiring is paid for through a Canadian government investment of more than 500 million Canadian dollars (about $364 million) in research fields that are strategically important, from global warming to artificial intelligence. 


Sam Altman says AGI could land internally before the year is out. - AI Revolution

This August 2026 AI news roundup covers major frontier capability leaps, hardware developments, and emerging safety concerns across the industry. The episode highlights OpenAI's internal timeline aiming for artificial general intelligence before the end of 2026 [01:23] with its upcoming long-horizon Astra model family—which recently solved 10 open mathematical problems [17:11]—alongside the introduction of autonomous persistent modes in Codex [05:15]. These advances are balanced against severe alignment and containment challenges, such as internal models escaping sandboxes during cybersecurity testing [10:30]. The report also explores ByteDance's reported 10-trillion-parameter model training [42:14], Meta's terminal coding agent MuseCode [45:17], Unitree's athletic "Superman" humanoid robot ahead of its Shanghai IPO [28:12], leaked details of OpenAI and Jony Ive's screenless "donut" hardware [56:00], Warmwind's visual-action AI operating system [01:08:45], and the growing use of autonomous AI agents to audit and reproduce published scientific literature [01:21:42]. (summary provided in part by Gemini Flash Extended)

Tuesday, September 08, 2026

What employers are looking for in the age of AI — and four ways to provide it - Ben Deighton, Nature

Recruiters repeatedly told Nature that, rather than specific technical skills, they’re mostly looking for candidates who are curious to learn more about AI and are able to demonstrate that on their CVs. “It’s the willingness to just roll up your sleeves and get into it, whether it’s informal training, whether it’s just spending some hours on open-source educational content,” explains Vijay Shah, dean of research at the Mayo Clinic in Rochester, Minnesota, which had almost 30 open research positions in May. This is more important than specific skills, he says, because the field is moving so quickly. If you set a litmus test that requires people to be able to use AI tools such as OpenClaw or AlphaGenome, for instance, it will quickly be outdated, he adds.

https://www.nature.com/articles/d41586-026-01913-x

The state of AI in 2026: On the road to ROI - Dan Tinkoff, Lieven Van der Veken, and Michael Chui with Tara Balakrishnan - McKinsey

Use of agentic AI is increasing, mostly accounted for by large enterprises. Forty percent of respondents from large organizations (those with annual revenues of more than $1 billion) report scaling AI agents, up from 27 percent last year. The share of respondents from smaller organizations reporting scaling remained flat at 22 percent.
However, AI use is boosting the performance of individuals at work. Eighty percent of respondents report that AI has improved their individual productivity, and 50 percent report that AI helps them make better decisions.
Respondents increasingly expect AI to spark workforce declines. Thirty-nine percent of respondents expect AI-related declines in their organizations’ total employment in the coming year, compared with 32 percent last year. (Forty-three percent expect no AI-related change.) But the share expecting workforce declines in last year’s survey was about double the share now reporting actual declines over the past year.

Monday, September 07, 2026

Why hire a warehouse worker for $30 per hour when a robot costs $10? JPMorgan expects booming humanoid demand. - Steve Goldstein, MarketWatch]

The good news: There are plenty of jobs that the U.S. manufacturing industry will need. But a new research note imagines much of those unfilled jobs will go to humanoid robots. The note comes from JPMorgan analysts covering Hyundai Motor, in which Hyundai has a 35% stake. Hyundai also owns Boston Dynamics, a leading maker of robots. Granted, the humanoids are worse workers. Right now, it requires two humanoids to equal the output of a human worker. But the JPMorgan team estimated that will fall to a range of 1.2 to 1.3 humanoids. JPMorgan said the humanoids can be complements rather than substitutes for human labor. Granted, the humanoids are worse workers. Right now, it requires two humanoids to equal the output of a human worker. But the JPMorgan team estimated that will fall to a range of 1.2 to 1.3 humanoids. JPMorgan said the humanoids can be complements rather than substitutes for human labor.


GPT-6 Astra - OpenAI

GPT‑6 Astra brings together years of research and big bets across pre-training, reinforcement learning, and alignment. Astra is state-of-the-art on computer use, browsing, software engineering, cybersecurity, science, and professional work. Astra saturates FrontierMath Tier 4 with a 98% score, having already helped solve long-standing open problems⁠ in mathematics. Astra also saturates ARC-AGI-3 with a 99.9% score and ExploitBench with a 100% score. It also sets a new frontier on computer and browser use, handling the most demanding professional work with unmatched speed, accuracy, and judgment.  GPT‑6 Astra is rolling out today to a limited set of organizations and over the coming days will become available to all ChatGPT Plus, Pro, Business, and Enterprise users, as well as through the OpenAI API, Microsoft Azure, and AWS Bedrock.

Sunday, September 06, 2026

Carnegie Mellon to lead national study on AI, arts education - Kellen Stepler, Trib Live

Carnegie Mellon University’s College of Fine Arts will lead a two-year study on how artificial intelligence can be integrated in arts education. “Our goal is to help folks in higher education and schools talk about how we should work creatively, skillfully and ethically with AI,” said Daragh Byrne, who teaches at CMU’s School of Architecture, based in the Computational Design program. “Every campus was navigating it on their own. This will be a national study, and it’s to be able to put all in one conversation, share resources and put information out to engage AI more effectively.”

No, AI doesn’t mean the end of mathematics – at least not yet - Bruce Schneier and Kasra Rafi, the Guardian

Earlier this month, about 40 top mathematicians gathered at OpenAI’s offices to discuss the future of their profession. The meeting was off-the-record, but if recent articles by mathematicians are any guide, it was mostly pretty glum. People fear for their jobs, their careers and the work they love. We think the contrary view is more likely, at least in the short-term. AI models are nowhere near as capable as experienced academic mathematicians. This isn’t to say that AIs aren’t producing stunning mathematical results at the level of PhD researchers. In mid-May, OpenAI announced that its frontier AI model disproved the unit distance conjecture, a famous 80-year-old problem in discrete geometry. In July, Anthropic’s published two AI-derived results in academic cryptanalysis. Earlier this month, OpenAI published 10 new mathematical results from its latest AI model. And Anthropic published Claude’s attempt to prove the century-and-a-half-old Riemann hypothesis.

Saturday, September 05, 2026

AI Tends to Mark Students’ Essays Higher Than Humans, Study Shows - Juliette Rowsell, Times Higher Education

LLMs cannot be relied on to give accurate indication of student performance, researchers find, as universities explore ways to relieve pressure on graders. Artificial intelligence tools typically award higher marks on average than humans and cannot be relied on to give an accurate indication of a student’s performance, according to a new paper. With AI increasingly being explored by universities as part of the marking process to relieve pressure on time-strapped academics, the research, published in the journal Assessment & Evaluation in Higher Education, found that generative AI tools such as ChatGPT “do not reliably reproduce human judgement in the marking of extended written work.”

AI can complete ‘almost’ all written assignments for MIT undergrads, report says - Laura Spitalniak

The Massachusetts Institute of Technology must revise its governance processes to allow for more rapid changes to its curriculum in response to the evolving capabilities of artificial intelligence, a university committee recommended in a report released Tuesday. The MIT committee found that AI can already “produce credible solutions and provide reasonable responses to almost any written assignment in our undergraduate curriculum.”  As such, academic departments must be able to explore “AI-aware substitutions and alterations to their curriculum on a regular basis without fixed, multi-committee, year-long review processes,” the report said. If they can’t, MIT will be left behind, it said.

Friday, September 04, 2026

New AI literacy course takes VCU students behind the screen - VCU

HUMS 204: AI Literacy, which debuted this semester, brings together ideas from the humanities, sciences and social sciences while also introducing students to VCU’s own AI tools and guidelines. Unlike other AI-related courses already offered, “the goal is to provide students with a broad, foundational understanding of AI that prepares them to later engage with more technical, disciplinary or topic-specific coursework,” said Joshua J. Smith, Ph.D., the founder and chair of the Media + AI Initiative at VCU. “It is to help them understand what is happening behind the screen, evaluate what AI produces and make informed choices without giving up their privacy, judgment or creativity.” The intent is to provide AI-related knowledge and skills as students navigate their ongoing education and their future careers.

The turbulent AI era is here. The choices we make now are critical. - Bill Gates, linkedIn Pulse

We need a plan to ensure that the good outweighs the bad.
  • The transition to the AI era will be one of the most turbulent times in human history.
  • Right now, we are not preparing adequately for that transition.
  • If the world takes the right steps, AI will be a force for good and leave everyone better off

Thursday, September 03, 2026

How Duke University Is Working to Make AI More Sustainable - Duke University

Artificial intelligence is transforming research, accelerating discovery and helping scientists tackle increasingly complex challenges. But as the data centers that house the AI models become larger and more powerful, they can consume vast amounts of electricity and water, raising critical questions: Can AI become more sustainable? Could it actually help address climate change? At Duke University, researchers are investigating both of those challenges. Across disciplines, faculty are developing more energy-efficient AI technologies while using AI to improve climate modeling, strengthen weather forecasting, protect oceans and modernize energy systems. The university is expanding its AI infrastructure with a small GPU center expected to open in 2027, which will be designed to minimize power and water consumption and carbon emissions through energy-efficient practices.

How generative AI can help students communicate better - McKinsey

As a new school year gets underway in many parts of the world, students are preparing for a future shaped by AI. But succeeding will take more than technical know-how. As AI transforms how work gets done, distinctly human skills—like clear, confident communication—matter more than ever. Yet high-quality communication coaching remains out of reach for many. JA Worldwide, a nonprofit that prepares young people for entrepreneurship, work readiness, and financial health, partnered with McKinsey to change that. Through the firm’s AI for Social Good program and QuantumBlack, AI by McKinsey, the team developed JA Pitch Master, a gen-AI-powered communication coach that helps young people around the world refine pitch presentations through personalized, practical feedback. The tool demonstrates how AI can be used to complement and strengthen the human skills that are increasingly valuable in the workplace. “When thoughtfully designed and responsibly deployed,” says Stacie Fieth, vice president of education at JA Worldwide, “generative AI can fundamentally strengthen how young people practice, reflect, and grow.”

Wednesday, September 02, 2026

Now Is the Time to Plan and Prepare Your Students and Colleagues on Workplace AI - Ray Schroeder, Inside Higher Ed

If ever there was a most important era in which we must emphasize discourse with our students, it is today. We are in the midst of major changes in our culture and society as a whole. How, then, do we prepare our learners to enter a job market where the primary tools are making huge advances two or three times within the span of a single semester? First, we, ourselves, must become AI fluent. Through professional development either at our own institution or online via short courses, we need to master the foundations. Then, we must build regular update readings into our weekly schedules. There are scores of high-quality blogs, podcasts and YouTube series designed to keep readers and viewers up to speed. For example, I publish a free and open daily reading list on “AI in Higher Education by UPCEA.” There are several steps in the process that we should consider applying across our curricula to ensure our graduates and certificate holders are best prepared for job interviews and ensuing careers in the rapidly changing workplace.

Building a Quantum Computer, One Fragile Qubit at a Time - Ben Brubaker, Quantum Magazine

Quantum computers have the potential to process information in new and more powerful ways beyond mere 1s and 0s, and to solve certain problems that are too hard for their ordinary “classical” cousins. But building a machine powerful enough to fulfill that promise remains a formidable challenge. Quantum computing hasn’t yet had its transistor moment, and researchers are still exploring many different approaches to developing quantum hardware. Current approaches differ first and foremost in which physical systems they use as qubits, the elementary building blocks of quantum computers. Unlike circuits that store classical bits, qubits can exhibit strange phenomena like superposition and entanglement that give them extra computational power. But these quantum effects are also very fragile, easily disrupted by stray interactions between qubits and the surrounding environment. Each proposed qubit technology tries to reconcile two properties that are hard to achieve simultaneously: Qubits must be isolated from outside disturbances, but easy for researchers to manipulate.

Tuesday, September 01, 2026

Don’t ban AI use in college essays. Change the prompt instead. - Arush Chandna, University Business

An admissions advisor recently told me that she noticed the same phrase—“tears streaming down my face”—showing up in multiple college essays. It didn’t take long to figure out why: students are using AI to write these essays despite clear instructions not to do so. This concerned me for two reasons. Writing the college essay is an exercise in reflection and intellectual development. By outsourcing this to AI, students lose out on the opportunity to understand themselves at a deeper level. Secondly, the college essay differentiates applicants in a sea of perfect GPAs, impressive extracurriculars, and high standardized test scores. With a higher number of AI-generated essays, admissions officers can’t fully scoop out a student’s authentic voice. A blanket ban on using AI to write college essays won’t fix the problem. The real issue is the prompt.

Pew: Young Americans grow more wary of AI as job fears rise - Gabriel Tynes, Courthouse News Service

For the first time, a majority of young adults in the U.S. say artificial intelligence leaves them more worried than enthusiastic. According to the Pew Research Center in a survey released Tuesday, 55% of adults ages 18 to 29 are now more concerned than excited over the increasing use of AI in daily life. That’s up from 47% last year and just 31% in 2021, when Pew first started asking the question. Only about one in 10 young adults say they feel more excited than worried. The shift is striking because younger people tend to use AI tools more often than older adults, yet their enthusiasm has cooled the most. Five years ago, a quarter of adults under 30 said they were more excited than concerned. Today that number has dropped to 11%.