363 items published this day, across 5 categories.
SAN FRANCISCO, July 30 (Reuters) - Anthropic said on Thursday some of its Claude AI models had hacked into the systems of three companies during cybersecurity tests, a disclosure that comes days after rival OpenAI revealed that one of its A ... (https://incidentdatabase.ai/cite/1627#7682)
Anthropic said its artificial intelligence models hacked into three other organizations during testing, just days after ChatGPT maker OpenAI raised concerns over AI controls after it disclosed its rogue models hacked another company. Anthr ... (https://incidentdatabase.ai/cite/1627#7683)
The AI company Anthropic says it has found three cases where its artificial intelligence programs left testing environments, accessed the internet and hacked into real companies. This comes after competitor, OpenAI, reported a similar incid ... (https://incidentdatabase.ai/cite/1627#7684)
Anthropic has said that its Claude models broke out of what was supposed to be an isolated testing environment and gained unauthorized access to the systems of three real organizations. If that sounds familiar, it's because it's the second ... (https://incidentdatabase.ai/cite/1627#7686)
Anthropic said Thursday that an internal investigation uncovered three incidents in which its AI model Claude breached the systems of three organizations while conducting cybersecurity tests. The investigation, and disclosure, comes more th ... (https://incidentdatabase.ai/cite/1627#7688)
Anthropic disclosed Thursday that three of its Claude models gained unauthorized access to the production systems of three organizations during cybersecurity testing --- not test servers or staging copies, but the live machines those compan ... (https://incidentdatabase.ai/cite/1627#7689)
Anthropic disclosed on Thursday that its AI models gained unauthorized access to the systems of three different unnamed organizations during cybersecurity testing. The company says Claude reached the internet "from within or while interacti ... (https://incidentdatabase.ai/cite/1627#7690)
Anthropic says it found multiple cases of Claude models gaining unauthorized access to other organizations' systems, the latest AI hacking incident to prompt concern and skepticism. In a blog post on Thursday, Anthropic said it proactively ... (https://incidentdatabase.ai/cite/1627#7691)
Thierry Rignol is furious at Yale. Rignol paid Yale $208,500 in tuition for its Executive MBA program. But after he was accused of cheating, the school suspended him for a year and gave him an F in the course Sourcing and Managing Funds. ... (https://incidentdatabase.ai/cite/1632#7692)
A former athletic director who used artificial intelligence to impersonate the principal of Pikesville High School and severely damage his reputation admitted Tuesday to sexually exploiting at least five children and faces up to 30 years i ... (https://incidentdatabase.ai/cite/675#7693)
SAN FRANCISCO, Aug 4 (Reuters) - (This August 4 story has been refiled to correct the spelling of Anthropic in paragraph 8) An AI agent was caught creating fake online identities to gain unauthorized access to secure systems during tests o ... (https://incidentdatabase.ai/cite/1633#7695)
You can access the full technical report here. AISI's role is to evaluate and understand the capabilities of frontier AI models, surfacing potential risks before they reach the public. To assess what these models can do, including whether ... (https://incidentdatabase.ai/cite/1633#7696)
An imagined town in Peru, an Eiffel tower in Beijing: travellers are increasingly using tools like ChatGPT for itinerary ideas -- and being sent to destinations that don't exist. Miguel Angel Gongora Meza, founder and director of Evolution ... (https://incidentdatabase.ai/cite/1635#7697)
A Threads user Dya (@dyaaaaaaa.\_) allegedly had to break bad news to an elderly couple who was duped by an AI video they saw online. The couple from Kuala Lumpur allegedly travelled all the way to Kampung Kuak Hulu, Perak, because they wa ... (https://incidentdatabase.ai/cite/1634#7698)
KUALA LUMPUR: A misleading video generated through artificial intelligence (AI) promoting a non-existent cable car in Kuak Hulu has caused an elderly couple from Kuala Lumpur to travel all the way to Perak, only to be disappointed upon arri ... (https://incidentdatabase.ai/cite/1634#7699)
Hoping to visit a cable car attraction featured in a social media clip, an elderly couple in Malaysia made the more than 300km journey from Kuala Lumpur to Perak before finding out that the video was AI-generated and that the ride did not e ... (https://incidentdatabase.ai/cite/1634#7700)
BBC News Brasil O fundador e diretor da empresa Evolution Treks Peru, Miguel Ángel Gongora Meza, se preparava para uma caminhada através dos Andes em uma cidade no interior do Peru, quando ouviu uma conversa curiosa. Dois turistas não acom ... (https://incidentdatabase.ai/cite/1635#7701)
I can perfectly imagine the pain, confusion, and betrayal in the voice of the elderly Malaysian woman who, according to a hotel staff member, asked "Why do they do this to people?" when she found out that her dream holiday destination wasn' ... (https://incidentdatabase.ai/cite/1634#7702)
KUALA LUMPUR: Dalam dunia digital yang semakin canggih, kecerdasan buatan (AI) kini mampu menjana video dan imej yang kelihatan sangat meyakinkan sehinggakan ramai tidak menyedari ia palsu. Tanpa sedar, pengguna mungkin mudah terpedaya, te ... (https://incidentdatabase.ai/cite/1634#7703)
BALING -- Orang ramai dinasihatkan agar tidak terpedaya dengan kewujudan perkhidmatan kereta kabel yang kononnya menghubungkan Pengkalan Hulu, Perak ke Baling, Kedah seperti yang tular menerusi sebuah video di media sosial baru-baru ini. P ... (https://incidentdatabase.ai/cite/1634#7704)
BALING, July 4 --- Authorities have dismissed a viral video claiming the existence of a cable car linking Pengkalan Hulu in Perak to Baling in Kedah, saying it was generated using artificial intelligence (AI). Sinar Harian quoted Baling Di ... (https://incidentdatabase.ai/cite/1634#7705)
A Malaysian couple travelled for three hours from Kuala Lumpur to the country's state of Perak to visit a tourist spot that exists only in an artificial intelligence (AI)-generated video. On June 30, a member of staff at a hotel in Perak s ... (https://incidentdatabase.ai/cite/1634#7706)
Gerade im Alltag vieler jüngerer Menschen haben sich KI-Tools wie ChatGPT bereits fest etabliert. In der Schule oder Uni helfen sie beim Verfassen von Hausaufgaben oder Bachelorarbeiten. Im Alltag helfen sie bei Einkaufslisten oder Plänen f ... (https://incidentdatabase.ai/cite/1635#7707)
¿Has utilizado alguna vez ChatGPT para planificar tus vacaciones? Lo más probable es que sí y también es posible que la inteligencia artificial te haya dicho que vayas a un lugar presumiblemente turístico que no lo es... y, puede que ¡ni ex ... (https://incidentdatabase.ai/cite/1636#7708)
Pilot scheme will provide three weeks of training as part of UK government’s latest attempt to address Neets crisis Young people out of work or at risk of unemployment in the UK are to join “AI boot camps” where they harness the technology to get a foothold in the workplace. The government’s latest attempt to address the crisis in Neets – young people not in work or education – involves turning to a technology that many view as a potential threat to employment. Continue reading...
District attorney says Arjun Aravind, 17, used internet and AI to search for fantasy stories regarding killing of his family A Massachusetts teenager accused of killing his mother and younger brother is being held without bail as authorities investigate a double-murder case that prosecutors say is connected to his use of ChatGPT. Arjun Aravind, 17, appeared Thursday morning for his arraignment in Concord district court, where a not-guilty plea was entered on his behalf to murder and several addi
On “The Opinions,” the Yale law professor and economist Natasha Sarin argues that A.I. could transform the U.S. economy for the better. But with the public hearing about layoffs and job losses, she says, it’s no wonder the technology’s economic promise isn’t resonating.
Meta AI | Future Guardian writers and Neets | Food for thought | Plants surviving the heat | Living and dying well Your article ( Zuckerberg pushes ‘superintelligent’ AI for all as Meta releases open-weight model, 10 August ) quotes Mark Zuckerberg as saying: “Everyone will have an exceptionally capable personal agent that understands you, your goals, and everything you care about. You’ll be able to interact with your agent through any device, including your glasses.” He is wasting his time
The police-tech giant Flock is announcing today that it will change officers’ access to its nationwide network of license plate readers, in an apparent effort to quell a growing backlash and win back contracts lost amid concerns about mass surveillance and police abuse. Several changes aim directly at a problem that has made recent headlines:…
We used to talk about the risks of A.I. a lot more. On a recent episode of “The Opinions,” the Opinion writer David Wallace-Wells speaks with the Yale law professor and economist Natasha Sarin about how A.I. safety could bring China and the U.S. together.
The dangers of AI become clearer every day. Why are we still acting as if we have no choice about our future? Rather than producing jobs, the US economy actually lost 23,000 jobs in July, according to Bureau of Labor Statistics data released on Friday. In addition, May and June’s job numbers were revised downward, showing a combined 103,000 fewer jobs than previously reported. As if this weren’t bad enough, wage growth has also slowed. Average hourly earnings rose by just 0.1% from June. Continu
When we set out to talk to kids about artificial intelligence, we thought we knew what we’d hear. We expected some to tell us they were using it to cheat a little, the way Millennials and Gen Xers opened up CliffsNotes or programmed formulas into their TI-82s, and others to share inspiring ways they were…
Taiwan’s statement comes a day after reports that suspected China-linked hackers had carried out a first-of-a-kind breach Taiwan says it detected AI-assisted cyber-attacks on government agencies that came from overseas last month, a new kind of threat that has been reported as “first-of-a-kind breach”. The Ministry of Digital Affairs (MDA) said its cybersecurity monitoring units detected the “abnormal attack” targeting government agencies, which began on 20 July. The National Institute of Cyber
OpenAI’s rogue agent hack was a watershed moment for AI safety and cybersecurity. It also sparked internal questions about the culture that led to it.
AI is shifting the culture, from tech CEO manifestos to 1 am job interviews. We unpack some of the latest, along with the top findings from Black Hat and Defcon, this week on Uncanny Valley.
For older employees already experiencing burnout, poorly designed or poorly timed training can become an additional job demand rather than a solution.
Amy Kremer’s resume makes her loyalty to Trump impossible to question. Humans First is banking that will help mobilize the right
Rapid revenue growth fuels hope Claude maker's IPO is the biggest listing in history
AI agents cross company boundaries, but proof of their authority often does not. When they overstep, no single record shows what the user allowed.
Over a billion people worldwide have livers with excess fat, which can lead to a host of medical problems. Researchers think AI tools can spot the condition—and help stop it—early enough to save lives.
AI is expensive, Ali Ghodsi tells TechCrunch. With so many investors wanting into his latest round, he said yes to more than planned.
Anthropic researchers found AI agents can clash, collude, and coordinate in unexpected ways, raising new questions about whether today’s safety tests capture the risks of multi-agent systems.
OpenAI has replaced chief revenue officer Denise Dresser after just nine months on the job, tapping Wiz president and chief operating officer Dali Rajic to take on frontier lab's top sales job.
A Brennan Center study found that major AI chatbots refused to endorse false election claims. The same companies’ image tools produced fake evidence for those claims on demand.
A Brennan Center study found that major AI chatbots refused to endorse false election claims. The same companies’ image tools produced fake evidence for those claims on demand.
KM pioneer DeepJudge has launched an ‘Agent Handoff Protocol’ or AHP, which is an open system that ‘enables users to move between AI platforms without ...
Vivodyne’s AI experiments on lab-grown human tissue could soon replace the need to test drugs on mice or dogs—and the approach has the potential to create drugs that work better.
Vivodyne’s AI experiments on lab-grown human tissue could soon replace the need to test drugs on mice or dogs—and the approach has the potential to create drugs that work better.
New York-based SmartEsq has rolled out multiple new capabilities designed to help private funds teams manage the legal lifecycle ‘in one connected environment’. The company ...
Neota Logic, which is now going to market as ‘the AI governance layer for legal teams’, has launched an AI orchestration capability which lets lawyers ...
StubHub, the industry's largest ticket resale platform, spent $3.4 million this year lobbying against California's AB 1720.
Indonesian President Prabowo Subianto lashed out at bureaucrats, state firms, police and the military over corruption and waste in a major speech, reinforcing a strongman leadership style that has ...
Everything You Need To Know About Usha Vance, You Can Learn At Yale Law: Her classmates gossip about her on Signal. The clerkship culture that shaped her politics is the part they don't put in the group chat. Daft For Taft : John Roberts pens essay celebrating William Howard Taft... but mostly trying to sugarcoat his own legacy . The Top Schools For Tech & The Law : Check out the Honor Roll here. Todd Blanche's First Message To The DOJ: Trust me. Slip And Fall Hopscotch : Law firm decorated side
CNBC’s Jim Cramer said investors shouldn’t automatically sell a stock that falls on weak earnings guidance.
Barbara Lavandeira opens up to Page Six about the hospitalization, what she saw that day, and comes to his defense against those calling the incident a comeuppance.
"That's where this is going,” Maj. Gen. Robert Kinney said. The post DIA’s artificial intelligence chief envisions ‘agent-to-agents’ interactions that support military operations appeared first on DefenseScoop .
CNBC's Jim Cramer warned investors against relying too heavily on historical comparisons, arguing that doing so can cause them to miss what has fundamentally changed.
A top Democrat on the House Energy and Commerce Committee is pressing major U.S. airlines over whether they use artificial intelligence to set ticket prices based on travelers’ personal information, raising concerns that it determines what fares consumers see. Rep. Frank Pallone Jr. (D-N.J.), the ranking member of the House Energy and Commerce Committee, sent...
They were law partners first, the scandal came later. The post The Only Attorney General To Go To Prison (So Far) appeared first on Above the Law .
Although the President has championed the proposal, some members of his own party have expressed doubts about it.
A St. Louis PBS station is suing a data center provider to rescue a 70-year archive.
Top news from the final day of the 2026 Space and Missile Defense Symposium.
The new guidance has sparked more questions than Defense Department officials would answer this week. The post Pentagon’s nascent BOND program informed Feinberg’s new ‘Funding Palantir’ memo appeared first on DefenseScoop .
Read your cases this summer. But also read the room. The post Your Most Important Summer Reading Isn’t On Your Syllabus appeared first on Above the Law .
Execs praised Night School's latest game, Unhinged, before pulling the rug.
Sterling, the New Zealand AI startup building an autopilot for finance teams, has raised $3.8 million (NZD) in a round led by trans-Tasman venture capital firm Blackbird.
The data center interconnection freeze is separate from the state's halting the disbursements of billions in broadband funding.
Databricks has closed a $5bn round at a $190bn valuation, led by Coatue, with revenue run-rate past $7bn and growth above 80% year on year. That is a 42% valuation increase in six months, and it comes after chief executive Ali Ghodsi called 2026 a bad year to go public. Databricks has closed $5bn at […] This story continues at The Next Web
As Generative AI accelerates code production across the financial services sector, traditional softw...
The NSF programs include one aimed at Hispanic-serving institutions that was established under bipartisan legislation and signed into law by President George W. Bush. The post DOJ finds three federal STEM education programs unconstitutional appeared first on FedScoop .
Cisco is showing booming demand for AI infrastructure, but its results weren't good enough to satisfy investors.
John Roberts delivered a book report on Taft, and managed to get in swipes against court reform and independent agencies. The post John Roberts Uses William Howard Taft To Defend His Own Record, Fails appeared first on Above the Law .
The Army’s space superiority efforts are focused on countering adversary ISR satellites, as well as conducting “stratospheric warfare,” Col. Joe Mroszcyck told Breaking Defense.
The AI lab grapples with a question that's existential for its future: Do ChatGPT corporate customers get a clear ROI?
President Trump is laying the groundwork for private sector firms to take a larger role in the U.S.'s cybersecurity offense against transnational cyber crimes in what could be one of the largest ever changes in U.S. cyber policy. In a memo signed Wednesday, Trump called on the National Coordination Center to create a program for...
Workday investors have been on edge over the prospect of AI disrupting its business model
The Omaha Police Department was awarded more than $65,000 for 40 pairs of the gloves in May 2025.
With bonuses coming every quarter, everyone gets a piece when Reid Collins has excess capital. The post At This Elite Litigation Boutique, It’s ‘Eat What You Kill’ — And Associates Are Feasting appeared first on Above the Law .
Hosting provider Namecheap is down after data center cooling failure.
Denise Dresser's sudden departure could be a major blow to OpenAI as it pursues a blockbuster IPO.
'We believe we can operate more strategically and efficiently within our Games business, and as a result are eliminating some additional Games roles'
California requires autonomous vehicle manufacturers to demonstrate they can safely monitor, update and maintain fleets while complying with cyber law.
“States have an opportunity to build the infrastructure of worker power that this country will need in the years ahead,” write Sharon Block and Benjamin Sachs.
A final round of photos from around the show floor in Huntsville, Ala.
Sure, Todd. The post Todd Blanche Would Like The DOJ To Know He Will Lead With Integrity appeared first on Above the Law .
Nick Marinos, the managing director of IT and cybersecurity issues at GAO, said leveraging existing avenues for cybersecurity information sharing can expedite getting AI safety policies into law.
The Netflix documentary unspools a story about a woman who faked a pregnancy and kidnapped a baby
Anthropic CFO Krishna Rao is leading early investor meetings, which have focused on big-picture themes like its Claude AI models, management and more.
Google shipped Gemini 3.7 Flash just three weeks after 3.6 Flash. The new model is supposed to be Google's most capable workhorse yet for coding and AI agents, and according to the company's own benchmarks, it beats Claude Sonnet 5 and GPT-5.6 Terra at half the price. The article Gemini 3.7 Flash lands with coding gains and undercuts its three-week-old predecessor's price by 50% appeared first on The Decoder .
The chat is just the alumni network with better gossip. The post There’s A Secret Yale Law Group Chat About The Vances Because Of Course There Is appeared first on Above the Law .
Chicago Public Schools is further away from having enough money to meet its students’ needs as defined by the state, according to new data released by the Illinois State Board of Education. Under ISBE’s calculations released Friday, CPS is now considered a little over 70% adequately funded, compared with 73% last year. CPS isn’t alone […]
A strong commercial space industry is an important partner for the U.S. government, as it contributes to building more robust space and defense capabilities and facilitates innovation more broadly. As competition between the United States and China heats up, both countries look to the commercial space sector to help them get ahead in both the space race and defense technologies. We asked five experts: What is a crucial step for the United States to take now to stay ahead in commercial space comp
SEOUL, South Korea — North Korea on Friday threatened to exercise its right to self-defense in a statement that slammed the upcoming U.S.-South Korean military drills as a war rehearsal that would be ...
Move echoes American Revolutionary War-era strategy of employing civilians to battle adversaries
Ultium Cells, the GM and LG Energy Solution joint venture, restarts EV battery cell production in Warren, Ohio next week after a seven-month shutdown, with about 1,400 workers. US EV sales rose 14.2% in the second quarter but remain 20.5% below a year earlier. The battery plant that GM and LG idled in January is […] This story continues at The Next Web
The Army could choose a winner late next spring after both the M1E3 and XM30 are operated by the Iron Horse brigade in a National Training Center rotation, an Army official told Breaking Defense.
An audit is underway, but no completion date has been set.
She has been labeled conservative, a swing justice, and sometimes unpredictable. This article puts all of these hypotheses to the test and provides a data based assessment of where she really stands. The post Who Is Justice Barrett? appeared first on Above the Law .
The bipartisan investigation, led by Sens. Josh Hawley and Dick Durbin, follows recent Senate efforts to advance new kids’ safety legislation and curb child sex abuse material online.
One engineer suggested outside consultants to support local permitting agencies
DeepSeek on Thursday open sourced the DeepSeek Harness, a new agent runtime for developers. The Node.js-based harness is now available The post DeepSeek open sources an agent harness where everything is a plugin appeared first on The New Stack .
It is too late to stop the technology being used as a damaging weapon, so great investment in defences is urgently needed
Relativity CEO Phil Saunders calls it a fundamentally new way for lawyers to get straight to the answers in their most consequential legal data. The post Relativity Announces claiR, A Conversational AI For Lawyers, But You’ll Have To Wait Awhile To Chat With It appeared first on Above the Law .
Kalshi and other prediction markets are coming under more pressure from multiple states and regulators. New York City and New York State are both investigating the exchange. Robert DeNault, Kalshi ...
The massive policy shift comes as Americans reported losing $20.8 billion to cybercrime in 2025.
Part of the World War II Memorial on the National Mall was covered in soapy bubbles and red graffiti Thursday in an incident that the U.S. Park Police were investigating. Mountains of white suds ...
Bei Google geht KI-Spitzenpersonal. Bots ersetzen wohl den Menschen im Netz. Und: Mark Zuckerbergs trauriges Verständnis von einer besseren Zukunft. Der KI-Newsletter
All Flock Safety customers will be required to adopt its "Audit Assistance" feature for tracking abnormal uses, and the company says it will hold license plate data for only seven days in most cases.
The number comes from investors, not the company. Six backers told the Financial Times that rising revenue would let the five-year-old lab more than double its valuation in an autumn listing. At $2 trillion it would eclipse SpaceX, which went public at $1.77 trillion in June. Anthropic itself has fixed nothing. Several investors said senior […] This story continues at The Next Web
New federal funding will aid child welfare agencies in testing out predictive analytics tools.
Taiwan entdeckte im Juli einen ungewöhnlichen Angriff. Nun bestätigt das Land, dass Hacker dabei autonome KI-Agenten einsetzten, um Netzwerke zu infiltrieren.
Right now, 15.5 million undergraduate students around the country are getting ready to start or return to college. Seven million of them have one thing in common: relying on a Pell Grant to help pay for their higher education. But the grant program is facing a $15 billion funding shortfall that will undermine college access […]
Deepseek has moved its flagship V4-Pro out of the testing phase and released its agent software, Harness v0.1, under the MIT license. API prices are going up at the same time, with cache hits jumping to six times their current cost. For agent workflows that repeatedly read the same files, that's the biggest price increase in the transition. The article Deepseek ships improved V4 Pro, open-sources its agent software, and raises API prices appeared first on The Decoder .
Elbridge Colby said at the start of his multi-nation tour that his message would be “realistic and perhaps bracing, but also reassuring.”
Databricks is benefitting off of the agentic AI wave.
If you're interested in technology and the law, you need to see this list. The post The Best Law Schools For Technology Law (2026) appeared first on Above the Law .
Welcome to AI Decoded , Fast Company ’s weekly newsletter that breaks down the most important news in the world of AI. I’m Mark Sullivan, a senior writer at Fast Company , covering emerging tech, AI, and tech policy. Sign up to receive this newsletter every week via email here . And if you have comments on this issue and/or ideas for future ones, drop me a line at sullivan@fastcompany.com , and follow me on X @thesullivan . As midterms loom, AI chatbots fight election lies, but then supply image
Every token has a price. The problem is that most AI systems don’t reveal the bill until they reach production. The post Why your AI pipeline costs 10x more after the demo appeared first on The New Stack .
Vercel has made the v0 API generally available, enabling developers and AI agents to programmatically generate, iterate on, preview, and deploy applications through API calls. By Daniel Dominguez
The firm wanted to talk about road safety. The mayor wants to talk about the state grievance committee. The post Law Firm Advertising Labeled ‘Graffiti,’ Threatened By Town Officials appeared first on Above the Law .
amicable, the award-winning divorce and separation service, has submitted its application to the Solicitors Regulation Authority (SRA) to launch a new, separate law firm serving England and Wales. Alongside that, it’s opening the search […] The post amicable applies to SRA to launch “tech-forward” law firm appeared first on Legal IT Insider .
Elite’s cloud customers will outnumber its on-premises customers by the end of 2026, with SaaS users growing 60% YoY to 53,000, and over 120 law firms now live in the […] The post Exclusive: Elite Cloud customers to outnumber on prem for first time in company’s history appeared first on Legal IT Insider .
Flock Safety will implement new privacy and data retention policies as pressure grows on the automated license plate operator to prevent its technology from being abused or used to conduct mass surveillance. The company announced Thursday it will shorten the recommended default data retention window from 30 to seven days to cut the amount of...
« Dix ans de post-vérité » (4/5). A peine Donald Trump élu en 2016, l’anglicisme « fake news » se diffuse dans le monde entier. Ambigu, malmené et récupéré, il est aujourd’hui remplacé par des termes plus techniques.
DIA, NGA and the FBI are building the infrastructure and guardrails needed for increasingly autonomous artificial intelligence.
“There is nothing natural about the educational segregation that has emerged in the United States,” writes Daren Acemoglu.
Your first decade will not unfold in a straight line. Some years will feel like acceleration. Others will feel like survival. Keep building anyway. The post The First 10: A Young Lawyer’s Blueprint Beyond The Billable Hour appeared first on Above the Law .
Ejecutar un agente de IA durante horas tiene una factura no siempre evidente: cada vuelta que da el modelo para pensar, comprobar o corregirse, se paga. Y ahí es donde Grok 4.6 , el modelo que acaba de lanzar SpaceXAI (el nuevo nombre de la empresa desde el mes pasado ), dice tener su gran ventaja. En la prueba AA-Briefcase completó sus tareas en 53 turnos consumiendo 500 millones de tokens de media. Claude Opus 5 Max, en cambio, necesitó 103 turnos y 2.000 millones de turnos para llegar a
The move primarily changes Choplife’s corporate and regulatory base, with the company citing Itana’s regulatory environment and its promise of simpler cross-border operations as key reasons for the decision.
'Neither Saber nor Unigine have replaced any writers with AI for Rideshare or any other game.'
Indiana’s early literacy rates improved for the fifth consecutive year, with nearly 89% of Hoosier third graders demonstrating proficiency in foundational reading skills on the state’s IREAD assessment. The Indiana Department of Education released statewide IREAD, ILEARN and SAT results for the 2025-26 school year Tuesday at the State Board of Education’s August meeting. The […]
The acquisition is slated to close in September and aims to provide AEVEX greater “customer access” to military users.
Ahead of the CNBC Investing Club's August Monthly Meeting, here's a look at what moved our three top and bottom performers since our last meeting.
When news of the scandal first broke on July 8, the company co-founders claimed that they had just been made aware of the situation. The post Phia Co-Founders Knew The App Was Wrongfully Claiming Sales Credits appeared first on Above the Law .
As people become accustomed to getting prescriptions online, sites are flourishing without much medical oversight.
Discord's Go Live feature contributed to a 13-year-old girl's death by suicide, according to Brazilian regulators, who told the company to suspend the streaming technology.
As smaller teams see bigger success, how will the game industry protect professional standards of conduct?
Linus Torvalds creó el kernel Linux en 1991, pero 35 años después ya no se define como un programador que escribe código. Durante el Open Source Summit India 2026 explicó que apenas lee código del proyecto de código abierto , y que hoy se considera más bien un jefe de proyecto. Uno de los programadores más legendarios de la historia ha abandonado esa labor, y es probablemente lo mejor que le podía pasar al proyecto. Lo de programar, como que no . Torvalds explica en ese encuentro que sigue escri
The Trump administration will allow private companies to launch attacks on cybercrime organizations, the White House announced.
Kenya’s Communications Authority (CA) has clarified new licencing rules for cyber cafés, saying operators will be required to keep basic customer and session records but will not have to track users’ browsing histories.
US police officers have quit after using licence plate-reading cameras to track romantic partners.
La plateforme de streaming audio, à rebours de certains de ses concurrents comme Deezer, reste cependant peu transparente quant à la part de musiques générées par intelligence artificielle qu’elle héberge.
The Defense Advanced Research Projects Agency entered into a contract with quantum networking company Qunnect to bolster technology to preserve quantum data in transit.
Die „Allgemeinen Anforderungen“ sollen Orientierung für Hersteller vernetzter Produkte schaffen. Auf Anfrage erhalten Unternehmen weitere Handreichungen.
Nvidia annonce un partenariat encore hypothétique avec six acteurs de la finance pour proposer une « nouvelle classe d’actifs investissables », selon les mots de son PDG Jensen Huang. Si l’accord est conclu, il pourrait permettre l’entrée de financements extérieurs au secteur de l’IA. NVIDIA annonce un accord avec Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs et KKR […]
Germany’s cabinet approved legislation that would let its intelligence agencies hack foreign systems, sabotage adversaries’ supply chains and feed false information to extremists inside Germany, in the biggest overhaul of the country’s spy laws of the postwar era.
Chatbots are becoming commonplace in teens’ lives, but these tools were never developed or validated for mental health support.
Here are five key things investors need to know to start the trading day.
Across the economy, Americans are watching an artificial intelligence investment boom reshape the job market. Some of the same companies spending hundreds of billions to build the AI future are also announcing sweeping job cuts, adding to an already daunting employment landscape for college graduates. In other sectors, opportunity is booming. Homes, roads, bridges, vehicles, […]
Nothing on Earth is as sure as night following day. An American tech start-up with military backing wants to change that. Not everyone thinks a brighter future is such a good idea. The post Darkness is losing its war with the light appeared first on Coda Story .
[Sponsored] AI tools are critical components of modern warfare, but must be protected to ensure they are reliable.
El Alzheimer, como otras enfermedades neurodegenerativas, tiende a manifestarse a edades avanzadas, aunque puedan pasar muchos años entre el comienzo de la enfermedad y la aparición de los síntomas. Cuando la enfermedad aparece antes de los 65 años hablamos de Alzheimer de aparición temprana. El problema es que este trastorno puede llegar a manifestarse mucho antes . El caso más precoz. No sabemos a qué edad puede llegar a comenzar a aparecer el mal de Alzheimer, pero el caso más precoz jamás re
Officers allegedly abused databases that can draw from sources like license plate readers and facial recognition.
One expert called it a “pretty big shift in U.S. cyber policy,” and there have been reservations in the past about opening the door to private sector involvement in cyber offense. The post Trump turns to private sector in offensive hacking operations memo appeared first on CyberScoop .
Suspending Discord's 'Go Live' feature, the National Data Protecting Authority has asked the company to prove compliance to Brazil's Digital ECA after a 13-year-old girl's suicide during a livestream. The post Data protection authority suspends Discord livestreams in Brazil after teen’s suicide appeared first on MEDIANAMA .
Use of the technology by both prospective applicants and employers has muddied waters
The growth comes as payments become an integral part of how banks capture value from Africa’s digital economy.
China already manufactures humanoid robots by the thousands. A new generation of U.S. start-ups is betting that there is still time to compete.
The University of Johannesburg has upgraded its MoUJi chatbot to handle student enquiries in multiple South African languages, including isiZulu, Afrikaans and Sesotho.
The local fintech grows from a two-person start-up into a 100-client firm processing more than R1 billion in travel payments.
Apple hat verraten, was es in diesem Jahr dazugibt, wenn man sich an der Back-to-School-Aktion des Konzerns beteiligt. Nicht immer lohnt sich das.
AI was involved in 55% of cybercrime cases observed by African countries surveyed by Interpol in 2025, as criminals used the technology to produce convincing phishing messages, fabricate identities, and impersonate executives and public figures. Deepfake incidents increased sevenfold between the second and fourth quarters of 2024.
Ransomware and phishing surge in Africa as AI lets threat actors scale up attacks, ESET research shows.
This story was co-published with Mother Jones. A few years ago, Eric Gil was living at his uncle’s place in Waterbury, Connecticut, where he shared a bedroom with his brother and cousin. With eight people in the house, it was crowded. He was trying to get out, but rental prices in Waterbury — which currently […]
As generative AI moves from experimentation to implementation, one question continues to dominate conversations in the legal sector: are law firms’ data foundations strong enough to support it? In our […] The post TalkingTech podcast: AI’s biggest legal challenge isn’t the technology. It’s the data appeared first on Legal IT Insider .
Les autorités taïwanaises n’ont pas donné de détails sur l’ampleur, les dégâts ou les cibles précises de ces attaques. Elles n’ont pas, non plus, mentionné la Chine, qu’elles accusent régulièrement de harceler l’île avec des cyberattaques.
Anthropic conducted an audit of 141006 evaluation runs after OpenAI's sandbox escape disclosure. The review identified three incidents where Claude models accessed the internet due to misconfigurations. These incidents involved unauthorised attacks on live targets. Anthropic has suspended offensive evaluations and plans to enhance security measures and collaborate with external auditors. By Olimpiu Pop
KI ist der Hauptgrund für IT-Umbauten, doch viele Projekte überschreiten Budget und Zeitplan. Datenqualität und Governance sind oft unterschätzt.
A new study of European vipers challenges the idea that snakes use venom to inflict pain on predators.
"The gravitational pull of the American financial system remains dominant," writes Jane Fraser, CEO of Citi.
Plus four other gadgets that do their bit for the environment
Researchers discovered hackers-for-hire performing cyber espionage and financially motivated heists from the same Web panel.
Jio Financial Services and Bank of America has signed an agreement which lets BofA acquire 49.9% stake in its NBFC subsidiary Jio Credit through a preferential allotment of equity shares and warrants. The post Jio Financial to sell 49.9% stake in NBFC arm to Bank of America for $1.9 billion appeared first on MEDIANAMA .
The Delhi High Court questioned Meta over alleged misuse of its Rights Manager tool, after creators said scammers used fraudulent copyright claims to remove original content and target accounts. The post How Meta’s copyright management tools are being exploited; Delhi HC to examine appeared first on MEDIANAMA .
The AI-powered job board links young Africans’ skills, training and career goals to employment opportunities.
The hub aims to strengthen the connection between policy, research and practice in support of a more coordinated innovation ecosystem.
There’s a big flaw in the way that drug companies develop and test medicines today: What works in a mouse often doesn’t work in a human. For decades, the industry has relied on animal testing. But in a laboratory south of San Francisco, a startup called Vivodyne is scaling up a different approach. Inside wardrobe-size mini labs, robots grow human tissue and run thousands of AI -designed experiments that could better predict how well a new drug will work—and whether it will be safe. [Image: Vivod
The organization announced the launch of the Breakthroughs to Follow-Through Initiative, which will implement artificial intelligence tools in public health research work.
The post A Youth Sports Giant Promised Reforms to Protect Kids From Sexual Abuse. Most Never Happened. appeared first on ProPublica .
The aircraft, which cost up to $50 million each, have seen heavy use around the Strait of Hormuz — but they are relatively easy targets for Iran and its proxies.
A digital signature is a mathematical method for verifying that a digital document, email or message is authentic, has not been altered and was signed by the person claiming to have signed it.
Als erster europ�ischer Broker �ffnet Bitpanda das Hauptkonto f�r Chatbots wie Claude und ChatGPT - ohne separates Agentenkonto. Wie die Freigaben funktionieren und wer haftet. Eine Analyse von Ulrike Barth ( KI , OAuth )
Chinese artificial intelligence start-up DeepSeek has quietly released DeepSeek-V4-Pro-0813, an updated version of its latest flagship model, leaving some developers underwhelmed by its overall capabilities and disappointed in its pricing – but impressing researchers in niche areas like cybersecurity. The stealth update to April’s preview version came with a brief statement on DeepSeek’s official website on Wednesday, noting that the model offered “significantly enhanced agent capabilities”, but
Given the relentless demand for computing power, electronic components are in scarce supply. Prices for certain memory chips, known as DRAM, have surged by more than 50 percent in a single quarter this year, and have roughly quadrupled since last fall. Because DRAM supply is tight, Apple, Dell, and HP are currently evaluating memory from ChangXin Memory Technologies (CXMT), a company the Pentagon has designated as a Chinese military company. Apple, in particular, has sought assurances from the U
Im Fokus der data2day stehen eine Keynote von Dr. Constanze Kurz, agentische KI-Systeme, Datenkontrakte und moderne Lakehouse-Architekturen sowie Governance.
Delhi HC Justice Prathiba M Singh questions mandatory AI-use disclosure by lawyers, warning it may add compliance without improving accountability. The post Delhi HC Justice Prathiba Singh questions mandatory AI disclosure for lawyers appeared first on MEDIANAMA .
The 1930 comic history, 1066 and All That, made famous the British habit of reducing national history to a sequence of memorable dates. The modern Royal Navy has its own unhappy version of that calendar. Since 1945, a series of ostensibly practical political decisions has steadily reduced Britain’s ability to sustain a globally relevant fleet: the 1956 failed Suez expedition, the 1966 retreat from “East of Suez,” the 1981 Nott Review, the 1998 Strategic Defence Review, and the 2010 review that f
ED Says It Plans to Spend Millions in Expiring Research Dollars. Concerns Remain. jessica.blake@… Thu, 08/13/2026 - 03:00 AM Higher education advocates say delays in education research funding have already caused severe damage. Byline(s) Jessica Blake
S.C. Colleges Could Face Financial Penalties Over New Bathroom Law Katherine Knott Thu, 08/13/2026 - 03:00 AM Byline(s) Katherine Knott
The NSF’s Ph.D.-to-Industry Pipeline Push kathryn.palmer… Thu, 08/13/2026 - 03:00 AM The majority of STEM Ph.D.s take jobs outside academia. While some universities help them prepare, the National Science Foundation is now investing millions to boost those efforts. Byline(s) Kathryn Palmer
Senators Prod State Department to Process Student, Scholar Visas sara.custer@in… Thu, 08/13/2026 - 03:00 AM Byline(s) Sara Custer
College Wasn’t Built for Student Parents Joshua.Bay Thu, 08/13/2026 - 03:00 AM In this week’s Voices of Student Success episode, Generation Hope’s Nicole Lynn Lewis explores how colleges can better support student parents and caregivers. Byline(s) Joshua Bay
Survey Shows Students Trust College Leaders More Than Politicians Olivia.sanchez Thu, 08/13/2026 - 03:00 AM Students from all political parties reported higher approval of campus-led policies, compared to state and federal policies, according to data from Gallup and Lumina Foundation. Byline(s) Olivia Sanchez
The NCAA Can’t Outrun Its Antitrust Problem sara.custer@in… Thu, 08/13/2026 - 03:00 AM A stalled bill, a growing pile of lawsuits and no clear path for the athletes at the center of it all. Byline(s) Sara Custer
Companies are looking for a strategic managed service provider partner to help them navigate a web of cloud interdependence.
C’est fait. Sans grande surprise, le plus grand projet de data center de France, Campus IA, a obtenu le 29 juillet l’autorisation préfectorale de s’installer à Fouju, à 40 km au Sud-Est de Paris. Après une consultation publique organisée auprès de la population à l’automne, puis une enquête publique à l’issue de laquelle les trois […]
Meaningful change is often less about demanding more from people and more about removing the obstacles that stand in their way, says Ivone Veiga-Moroldo, head of client experience at Healthbridge.
Modern lay-by has become a broader ecosystem of flexible payment choices, says Andre Van Rheede, regional sales manager at Altron FinTech.
Claude Code produktiv einsetzen und durch agentische Entwicklung mit Mastra, Tool Calls, MCP, Frontend und AG-UI-Protokoll zum eigenen Chat-Agenten.
The license-plate-reader firm said it will require stricter oversight after a Washington Post investigation found dozens of officers had misused its cameras to spy on their exes and romantic interests ...
In today's edition: Virtual asset firms to join CBN’s sandbox || Jumia secures $50 million equity funding || Shoprite’s Sixty60 is having a moment || Vodacom taps ex-Airtel CEO to join board
Lovable, a Stockholm-based software creation platform, has raised $400 million in Series C funding at a $13.3 billion valuation. The round was led by Menlo Ventures and co-led by EQT’s Scaleup Europe Fund. New investors include Tencent, Balderton Capital, Carmignac, Kaszek Ventures, LTS Growth, World Innovation Lab and Regent. Lovable provides tools that allow users […]
Honor has launched its Robot Phone, a smartphone equipped with a four-degree-of-freedom titanium gimbal and a system-level AI agent architecture. The phone measures about 9.59 millimeters thick, weighs 248 grams and includes a 7,060mAh battery and a 6.31-inch display. The 12GB+512GB version is priced at 9,999 yuan, while the 16GB+1TB version costs 12,999 yuan. Pre-orders […]
Alibaba Cloud has launched Qwen AI Arena, a challenge and evaluation platform for AI agents. The platform creates tasks based on real business scenarios and provides developers with models, runtime environments and evaluation tools to submit and test agent solutions. Its first challenge focuses on cross-border e-commerce. Participants must generate product listings for the US, […]
“What we need is more energy of all sorts, not only more renewable energy,” Lucci tells Fortune at the firm’s headquarters in Pasay City, Manila.
South Korea's Kospi has staged a reversal from its latest rout, returning to bull-market territory as investors piled back into the semiconductor giants that dominate the index.
Swedish company was valued at $5.6bn only four months ago as interest in artificial intelligence tools for professional services soars
Rapid revenue growth fuels hope that the Claude-maker can overcome challenges to become the biggest listing in history
For years, Elon Musk has dreamed of conquering a range of chronic health conditions by inserting computer chips into the human brain. But that vision could become reality fastest in China, where state authorities are launching a coordinated effort to accelerate the nascent industry’s development. The past few days have seen a string of initiatives related to brain-computer interfaces (BCIs) announced in China, involving parties ranging from state insurance companies to investment banks and local
The rise of equity-rich tech workers at Anthropic and OpenAI is shifting negotiating power towards clients
Um im Wettbewerb mit großen KI-Anbietern zu bestehen, wandelt sich das europäische KI-Labor Mistral zunehmend vom Modellentwickler zum KI-Infrastrukturanbieter.
A new original series exploring the life of Indian freedom fighter Bhagat Singh is in development at Collective Studios’ Historyverse. The project is being made in association with Hathiramani Commercial Ventures and co-produced with investor and entrepreneur Manish Hathiramani, who marks his producing debut with the show. The series is designed to move past the […]
After nearly 20 years of tracking representation metrics, the latest report from Dr. Stacy L. Smith and the Annenberg Inclusion Initiative reveals that Hollywood’s film industry has not made lasting progress towards inclusion. “While we’ve seen pockets of progress for women on screen, the pace of change has been slow. Year after year, many of […]
Over a decade after the Amateur Athletic Union promised to implement “historic child protection measures,” it hasn’t done so, leaving hundreds of thousands of athletes at risk.
The election demanded by Reform UK's leader, who faces corruption accusations, has turned into a punch line after major parties dismissed it as a stunt and declined to field candidates.
Versatility is in, with people expected to work across the tech stack.
A recap of leading news from the second day of the 2026 Space and Missile Defense Symposium.
Microsoft has struck a deal to integrate S&P Global AI-ready data, insights and analytics into its 365 Copilot workflows and agentic experiences.
At a gathering in Boston, ‘student senators’ proposed a first-of-its-kind national AI policy for K-12 classrooms.
The county is known for its hundreds of ancient temple murals. An investigation has revealed that most specimens no longer exist or are in disrepair.
Learn how startups use GPT-5.6 to build faster, more cost-efficient AI agents with smarter model selection and new Responses API capabilities.
Flock Safety, the embattled vendor of mass surveillance technology, has rolled out a handful of new reforms intended to appease the justified nationwide anger that has seen scores of towns cancel or suspend their contracts with the company for automated license plate readers (ALPRs). The reforms are a combination of long overdue changes along with some cosmetic fixes that fail to address the fundamental dangers of this technology. We should not be letting companies decide how much privacy we des
Access Now's Digital Security Helpline and Vita Activa offer holistic digital safety and mental health support for activists. The post Vita Activa & the Digital Security Helpline join forces to support mental health and digital safety for activists appeared first on Access Now .
The whole country ostensibly wants America to win the artificial intelligence (AI) race. A striking number, however, would prefer someone else’s town to host the data centers, power plants, transmission lines, and cooling systems required to run it. Adam Smith knew the type. In “The Theory of Moral Sentiments,” he warned against the “man of ... The Data Center Chessboard Has No Pause Button The post The Data Center Chessboard Has No Pause Button appeared first on Truth on the Market .
If you are one of those tech leaders in a company that already has an agentic architecture with swarms of agents delivering some value and your challenge is how to optimize the cost and/or stitch these agents together across complex workflows, this blog may not be for you. But please read it if you have […]
Columbia Global Freedom of Expression seeks to contribute to the development of an integrated and progressive jurisprudence and understanding on freedom of expression and information around the world. It maintains an extensive database of international case law. This is its newsletter dealing with recent developments in the field. For a video to go viral on TikTok, […]
The hacking of HuggingFace by an internal OpenAI model, and more importantly the internal events that led to that and the fallout from it, remain the thing that matters.
Expert analysis of how Iran's conditions for reopening the Strait of Hormuz fit within the strictures of international law. The post Iran’s Conditions for Reopening Hormuz: What International Law Requires, Permits, and Forbids appeared first on Just Security .
Trump's family is profiting spectacularly from proximity to power, providing an opening for a select committee to investigate, modeled after the 1975 Church Committee. The post Congress Needs a Select Committee That Focuses on Trump Family Corruption appeared first on Just Security .
GeForce NOW is giving cloud gaming an extra-credit upgrade just in time for back-to-school season. The native Linux app for GeForce NOW is officially out of beta. GeForce NOW is also delivering new cloud optimizations that make Frame Generation feel even more responsive while streaming. On top of that, Performance members will see higher frame […]
A collection of syllabus supplements which offer curated articles intended to be combined with traditional course books in a law school or higher education classroom. The post Syllabus Supplements appeared first on Just Security .
This syllabus supplement offers curated articles intended to be combined with traditional casebooks in a law school or other higher education classroom. The post “In Focus” Syllabus Supplements: U.S. Lethal Strikes on Suspected Drug Traffickers, Operation Southern Spear, and Operation Absolute Resolve (2025–2026) appeared first on Just Security .
This syllabus supplement offers curated articles intended to be combined with traditional casebooks in a law or higher ed classroom. The post “In Focus” Syllabus Supplements: ICE and CBP Operations in Minnesota and Other States (2025–2026) appeared first on Just Security .
This syllabus supplement offers curated articles intended to be combined with traditional casebooks in a law or higher ed classroom. The post “In Focus” Syllabus Supplements: Artificial Intelligence, Emerging Technology, and National Security (2025–2026) appeared first on Just Security .
This syllabus supplement offers curated articles intended to be combined with traditional casebooks in a law or higher ed classroom. The post “In Focus” Syllabus Supplements: Russia’s War Against Ukraine (2022–2026) appeared first on Just Security .
This syllabus supplement offers curated articles intended to be combined with traditional course books and other materials in a law school or higher education classroom. The post Jus Ad Bellum: Syllabus Supplements appeared first on Just Security .
Access the Immigration Law & Policy Syllabus Supplements via PDF here. Access the original version, published Sep. 18, 2025, via PDF here. Additional syllabus supplements, including our new “In Focus” syllabus supplements, are available here. Table of Contents I. Executive, Congressional, and State Authorities II. Admission to the United States III. Non-Citizens in the United […] The post Immigration Law & Policy: Syllabus Supplements appeared first on Just Security .
This essay was written with Nathan E. Sanders, and originally appeared in Tech Policy Press . AI represents the first time we humans can do cognitive work outside of our bodies at scale. The only comparable moment is the early years of the industrial revolution, when new technologies like the steam engine provided a quantum leap in our ability to do mechanical work outside of our bodies at scale. If AI’s cognitive capabilities become integrated into our lives, businesses, and governments—a proce
Published on August 11, 2026 11:51 AM GMT TL;DR NOVAH (No Violence At Home) was incubated by Charity Entrepreneurship (now Ambitious Impact) in 2024 to test a promising idea: preventing intimate partner violence through edutainment, in our case a serialised radio drama. Over the past two years we have produced and aired two seasons in Rwanda. We are currently evaluating our second season through a randomized controlled trial with 2,400 couples in Rwanda in partnership with Innovations for Povert
As concerns around data privacy in machine learning grow, the ability to unlearn, or remove, specific data points from trained models becomes increasingly important. While state of the art unlearning methods have emerged in response, they typically treat all points in the forget set equally. In this work, we challenge this approach by asking whether points that have a negligible impact on the model’s learning need to be removed. Through a comparative analysis of influence functions across langua
… were identified, alongside the need for further ML in order that the tool could be made more widely available across all ROCUs and other crime types in the future. This Machine Learning (ML) phase will ensure that the AAT can deliver region specific insights into County lines and provide the evidence required to inform future national adoption. Classification: Software package and information systems.
Onsite search optimisation for the FCA website (FCA.org.uk), Analytics of search terms, AI conversational search and Content Intelligence to review and action site content for AI readiness. Classification: Web search engine providers.
Finding a low latency, contextual guardrail solution that understands AI-specific threats to ensure a secured, thorough review of all applications the FCA reviews Classification: Research and Development services on security and defence materials.
BY THE PRESIDENT OF THE UNITED STATES OF AMERICA A PROCLAMATION 1. Within the past 90 days, the Secretary of Commerce (Secretary) transmitted to me a report on his investigation into the effects of imports of unmanned aircraft systems (UAS), as well as their parts and components (together, UAS components), on the national security of […] The post Adjusting Imports of Unmanned Aircraft Systems and Unmanned Aircraft Systems Components into the United States appeared first on The White House .
MEMORANDUM FOR THE SECRETARY OF WAR THE DIRECTOR OF THE OFFICE OF MANAGEMENT AND BUDGET THE ASSISTANT TO THE PRESIDENT FOR NATIONAL SECURITY AFFAIRS SUBJECT: Rebuilding the United States Navy and America’s Shipbuilding Industrial Base By the authority vested in me as President by the Constitution and the laws of the United States of […] The post Rebuilding the United States Navy and America’s Shipbuilding Industrial Base appeared first on The White House .
The FAA is superseding Airworthiness Directive (AD) 2018-11- 14, which applied to certain The Boeing Company Model 767-300 and -300F series airplanes with certain winglets installed. AD 2018-11-14 required high frequency eddy current (HFEC) inspections for cracking of the lower outboard wing skin and repair or modification if necessary. AD 2018-11-14 also required one of three follow-on actions: Repeating the HFEC inspections; modifying certain internal stringers and oversizing and plugging the
First transparency report due by September 17 ANPD establishes minimum content and a deadline for September 17 for the first Digital ECA transparency report. In brief On August 11, 2026, the Brazilian Data Protection Agency (ANPD) issued Decision Order CD/ANPD No. 122/2026, which addresses requirements of the semiannual transparency reports required under the Children and [...] The post Brazil: ANPD regulates Digital ECA transparency reports appeared first on Connect On Tech .
WBG Pioneers is the World Bank Group’s premier internship program, offering undergraduate and postgraduate students structured learning and practical experience. As a participant, you will engage in ...
What GAO Found The 988 Suicide & Crisis Lifeline provides free, confidential 24/7 support for anyone experiencing mental health distress, suicidal thoughts, or substance use crises. The National Maternal Mental Health Hotline provides free, confidential emotional support, resources, and referrals to pregnant and postpartum women experiencing mental health challenges. Both of these hotlines, within the Department of Health and Human Services (HHS), are operated by nonfederal entities—primarily by
Transforming multimodal sources into condensed and structured media outputs can be fundamentally conceptualized as a long-horizon agentic process centered on a model-harness system. While an ideal harness system should align with human design priors and accumulate reusable experience through empirical exploration to drive recursive self-improvement, existing paradigms remain static and fall short of this capability. In this paper, we present AutoDesign, a framework that aligns with human design
Humanoid motion tracking is central to teleoperation and whole-body imitation, yet evaluation often disagrees with what people perceive in videos. Kinematic errors average per-frame pose differences but miss the physical artifacts that matter most, particularly unstable support and incorrect contacts such as foot skating and mistimed touch-downs. Meanwhile, widely used test suites are small and lack the diversity needed to stress contact-rich, long-horizon behaviors. We introduce HumanTracker to
Current large language model development relies on massive, often non-permissible datasets, creating a high barrier for researchers committed to open-source and ethically sourced data. We introduce Mimir v1, a 1-billion-parameter language model based on the Hierarchical Reasoning Model (HRM) architecture, that is trained from scratch and delivers highly competitive performance for English and sets a new state of the art for Danish using only permissible post-training data. Trained on a mixture o
This report presents an improved version of AlayaWorld. While the backbone architecture, chunk-wise autoregressive generation scheme, and training data remain unchanged from the previous release, we substantially revise how conditioning signals are represented and integrated into the model. The new design is guided by a simple principle: conditioning signals should match the generated content as closely as possible in both latent representation and temporal structure. To this end, we make two ma
When asked about entities outside their knowledge boundary, LLMs routinely fabricate plausible-sounding details rather than backing off to safer, more general claims. We frame this failure through a Gricean lens: a cooperative speaker who is uncertain about a referent retreats up the specificity hierarchy, trading informativeness for truthfulness. We ask whether LLMs have the ingredients to perform this retreat. Using a T-REx-based benchmark that varies entity familiarity and referent specificit
As language-model-based AI is increasingly deployed in autonomous settings, aligning its goals and values with those of humans becomes critical. Today, alignment, and the assistant identity itself, are typically introduced only after pretraining, once behavioral priors are already established. This can make values a thin overlay, rather than deeply rooted, and facilitate subsequent misalignment. Pursuing a different paradigm, we introduce Synthetic Persona Pretraining (SPP), which installs the d
World Models (WM) are increasingly seen as a foundation for intelligent agents that can predict, plan, and act beyond their training distribution. In this paper, we study WMs from a causal perspective across multiple levels of abstraction, ranging from perceptual observations to building a conceptual representation of the structure governing the environment dynamics. We argue that useful WMs must go beyond generative capabilities alone: they should also capture entity properties, entity-to-entit
Vision-Language-Action (VLA) models have emerged as generalist robotic policies capable of following diverse language instructions and performing a wide range of manipulation tasks. However, their direct control over embodied agents also exposes them to adversarial interference that may cause unsafe physical behaviors. Existing attacks on robotic policies are typically optimized for a single task or instruction, leaving the cross-task vulnerabilities of multitask VLAs largely unexplored. We intr
Academic leagues have become important mechanisms for promoting extracurricular education and strengthening the integration between universities and society. This paper presents the organizational framework adopted by the Academic League of Artificial Intelligence (LIA) at the Federal University of Santa Catarina (UFSC), designed to integrate teaching, research, and university extension through a student-centered, project-based approach. The framework combines democratic governance, collaborativ
Machine learning ethics researchers and critical HCI scholars have argued that algorithmically predicting gender is wrong. At the same time, other researchers rely on predicted gender labels to study gender disparities and develop algorithmic fairness techniques. How do we reconcile these two seemingly contradictory intuitions? We differentiate two ways gender prediction may be wrong: being illegitimate, thereby contributing to harm; and being invalid, thereby producing unusable measurements. Ou
6G networks will not be serving as communication infrastructures only; rather, they are expected to evolve into intelligent systems, where thousands of autonomous artificial intelligence (AI) agents are interconnected. The agents are deployed across a wide range of platforms including low Earth orbit (LEO) satellites, high-altitude platforms (HAPs), unmanned aerial vehicles (UAVs), edge servers, and terrestrial devices. These agents continuously observe their environment and exchange information
Enterprise security topology design requires translating business intent, regulatory requirements, and risk assumptions into zones, boundary devices, inter-zone paths, and access-control policies. Existing NetOps automation tools mainly operate after this design is fixed, providing limited support for generating structured security topologies from underspecified natural-language requirements. We present TopoIntent, a system that compiles security intent into executable, compliance-checked networ
Artificial Intelligence (AI) safety systems combine character shaping (e.g., Reinforcement Learning from Human Feedback [RLHF], Constitutional AI), which modifies behavioral distributions at training time, with rule enforcement (e.g., output filters, safety classifiers), which blocks harmful outputs at inference time, yet little formal analysis exists on how their optimal balance should change as deployment scales increase. We introduce a stylized comparative-statics model that parameterizes saf
Long-horizon Earth observation reasoning requires models to organize multi-stage geographic evolution, localize spatial changes, detect temporal anomalies, and infer future from extended image sequences. However, existing remote sensing vision-language models mainly focus on isolated images, image pairs, or short sequences, limiting reliable grounding in the relevant frames and regions. We introduce LongEarth-Bench, a benchmark containing approximately 120k question-answering samples derived fro
Infrared (IR) spectroscopy is widely used for chemical sensing, but extracting reliable chemical information from spectra remains challenging. Conventional interpretation is labor-intensive, relies on prior knowledge and reference spectra, and is difficult to scale, whereas most machine-learning methods are tailored to individual tasks or datasets, require large labeled training sets, and transfer poorly across analytical objectives and experimental datasets. Here we introduce UltraIR, a foundat
Large language model based multi-agent systems usually communicate in text, i.e., using discrete tokens. However, text introduces a discrete bottleneck. Converting the sender's continuous hidden states into discrete tokens discards information that token identities alone cannot capture. Recent work proposes latent communication as an alternative, where agents transmit hidden representations directly without converting them to text. However, existing latent methods either inject working memory la
We ask whether language-model pre-training can be decomposed into smaller, independently trainable jobs that can later be recomposed into a coherent larger model. We introduce Mixture of Training (MoT), a scaffolded modular pre-training procedure that partitions a target Transformer into contiguous layer blocks, trains each block inside a frozen pretrained aligner scaffold, and then recomposes the trained blocks with an optional short end-to-end adaptation pass. On a 1.3B-parameter Gemma-style m
A small number of firms based in two states produce the most capable frontier AI models. The governments of those states have shown both the legal power and the political will to decide which other countries may use these systems. In June 2026 the United States required a leading developer to obtain licences before releasing its most advanced models to any foreign person, including foreign nationals resident in the United States. The affected models were withdrawn worldwide at short notice, part
Time series foundation models (TSFMs) have advanced primarily through architectural innovation, while training regimes for large-scale heterogeneous corpora remain under-explored. As a result, pre-training distributions are often poorly controlled with respect to domain imbalance, context requirements, prediction horizons, and missingness. We introduce ORBIT (Omni-Range Bootstrap Incremental Training), a training paradigm that makes this distribution explicit and controllable. ORBIT combines Boo
As biomedical research increasingly relies on data-intensive tools, the quality and utility of datasets are critical. Challenges such as imbalances, biases, and ethical or legal constraints often limit access to high-quality data. Synthetic data generation can help overcome these limitations. Here, we present a comparative analysis of generative models for transcriptomic data, investigating strategies to incorporate prior biological knowledge via gene graphs. This ensures that synthetic data cap
Geometry-conditioned multi-view diffusion enables high-quality 3D texture generation, but its repeated per-view denoiser evaluations introduce substantial computational cost. Existing training-free accelerators primarily exploit temporal redundancy by reusing computation across denoising steps. In multi-view texturing, however, skipping a step also removes the cross-view interaction that continually aligns different observations of the same surface, leading to rapidly degraded consistency and fi
Normative datasets are often used to train and align AI systems, but the norms they contain can function as action-guiding patterns rather than neutral moral knowledge. We propose treating the AI system as a proxy actor and test whether dataset-level norms can shift it away from its baseline safety behavior when it faces high-conflict dilemmas. We make three contributions. First, we demonstrate in controlled experiments that norm-breaking fine-tuning yields norm-divergent actions justified by se
Agent harnesses combine retrieval, routing, state, provenance, and verification, but locally successful components may disagree on shared state. We model this failure with a finite \emph{capability sheaf}: stalks encode typed behavior signatures, restriction maps retain shared fields, and accepted runs are useful global sections. An exact finite constraint-satisfaction problem (CSP) defines acceptance, while a linearized relative cohomology class provides a diagnostic and search feature. A contr
Multilingual retrieval-augmented generation (mRAG) equips large language models with access to globally distributed external knowledge for complex multilingual question answering. Recent approaches either translate retrieved documents into English or the query language to bridge the cross-lingual semantic gap, or decompose a complex query into sub-questions and aggregate the intermediate reasoning process. However, both lines of work suffer from two limitations. First, one-size-fits-all translat
With the rapid advancement of large language models (LLMs), research idea generation has attracted increasing attention. Existing approaches enable LLMs to retrieve relevant literature and propose novel ideas for research areas. However, current evaluation practices for idea generation remain fragmented and lack objective standards, often relying on direct LLM scoring, which limits their ability to provide unified and reliable assessments across a coherent distribution of generated ideas. To add
Agent Skills are today either hand-authored or produced in a single LLM generation pass, and consequently possess no closed loop through which they might improve from the interaction failures they actually cause. Recent work does close this loop, but derives its feedback from single-turn question-answering evaluation. The consequence is a sharp asymmetry: once the first round has patched the gaps that a single exchange can reveal, the evolution gradient decays, the defects that surface only acro
Contemporary online assessment systems rely primarily on browser lockdown, webcam monitoring, and behavioural analytics, yet remain vulnerable to attacks that extract the assessment content itself through screenshots, screen sharing, optical character recognition, and automated scraping. This paper extends the Multi-dimensional Spatio-Temporal Context Camouflaging Model (MSCCM) within the MARS (Multi-modal Assessment Resilience Suite) by introducing the Multi-Layer Context Camouflaging Theory (M
Electroencephalography (EEG) decoding models often generalize poorly across datasets and subjects due to domain shifts in acquisition protocols and individual neurophysiology. We propose EEG-PRIME, a two-stage EEG foundation model for cross-dataset multi-task decoding. EEG-PRIME combines masked pretraining with prototype-aligned instruction tuning to enable instruction-aware and subject-invariant decoding across diverse BCI paradigms. During pretraining, an EEG encoder learns transferable repres
Large language models (LLMs) are predominantly aligned to function as passive, sycophantic assistants. We challenge this default paradigm by empirically evaluating the cognitive plasticity of open-weight architectures when subjected to rigorous behavioral reprogramming. Our objective is to induce a proactive, Socratic conversational framework, characterized by high-frequency question generation under strictly constrained high-performance computing (HPC) conditions. Through a massively paralleliz
Diffusion models have achieved dominant performance in visual generation but suffer from substantial inference overhead. While cache-based acceleration has emerged as a promising solution, existing policies rely on local similarity heuristics, which we identify as being significantly misaligned with final generation quality. This discrepancy stems from the non-uniform propagation and accumulation of errors along the denoising trajectory. To address this, we propose Global-Impact Cache (GCache).
Algorithmic fairness evaluation commonly assesses AI systems as bounded technical components, abstracting away the organizational context in which they operate. We present, to our knowledge, the first independent end-to-end fairness audit of a semi-automated hiring system operated by Barcelona Activa, a public employment agency using the third-party TalentClue platform for candidate search and shortlisting. We analyze approximately 497,000 candidate-vacancy pipeline entries from September 2017 t
Perturbation methods explain model decisions by measuring prediction changes under altered inputs, but response magnitude tells us only how much a model reacts, not what that reaction means. The same magnitude can support the final factual-counterfactual difference, oppose it, or arise strongly along the perturbation path yet vanish at the endpoint. We therefore track how the contrast develops as paired inputs are progressively revealed, using the final contrast to interpret the trajectory. We i
Autonomous driving requires planning under both semantic constraints and predictive dynamics. Existing end-to-end driving approaches, however, typically emphasize only one side of this requirement: Vision-Language-Action (VLA) models exploit VLM priors for semantic reasoning, while World Action Models (WAMs) provide future-aware prediction through generative world modeling. This naturally motivates a unified planner that can leverage both semantic priors and predictive dynamics. However, we find
Self-improving LLM agents convert successful trajectories into persistent cross-task state. An unsafe success can thereby become reusable policy after its triggering input disappears. Skill evolution makes this failure measurable by distilling operational trajectories into executable, transferable, and inspectable procedures. Because evolution optimizes task outcomes rather than procedure safety, compromised experience can cause skill misevolution. Existing benchmarks measure current behavior or
Semantic ID (SID)-based generative recommendation has recently achieved remarkable success. However, existing methods suffer from a previously overlooked fairness issue, which we term \textbf{Token Frequency Bias}, where high-frequency SID tokens are systematically over-predicted while low-frequency SID tokens are under-predicted. This bias originates from the combined effects of imbalanced semantic codebooks during SID construction, and popularity bias together with the maximum likelihood estim
Text-based person anomaly retrieval aims to retrieve pedestrians exhibiting anomalous behaviors from a large image gallery using natural language descriptions. Compared with conventional text-based person retrieval, this task requires fine-grained reasoning over pedestrian appearance, behaviors, object interactions, and scene context, making robust cross-modal matching significantly more challenging. This paper presents the GENAI4E team's solution to AI City Challenge 2026 Track 4. Our framework
The rapid advancement of Auto-Research has surfaced a fundamental evaluation challenge: how can we measure the alignment, logical coherence, and evolutionary completeness of its research trajectory with human research behavior? We propose Auto-Research's Alignment and Completeness, ARAC-Bench: a Researcher-Mimicking Evaluation framework that shifts the objective from matching final answers to reproducing high-quality human research processes. The framework operates through two synergistic compon
Agentic workflows are commonly evaluated by whether they reach the correct outcome. That is insufficient in institutional settings, where a correct action may rely on the wrong authority, an unsupported completion claim, or work made stale by a later change. We define governed execution as work whose decisions, completion, and response to change are supported by inspectable provenance. We present Matrix, a deterministic causal-state layer that records authority and fact dependencies, verifies co
While Large Language Model (LLM) agents increasingly rely on long-term memory for persistent interactions, the retrieval mechanisms governing this memory are rarely treated as evolvable components. This static approach limits performance on heterogeneous memory queries, which often demand diverse evidence construction strategies. To address this, we introduce \textbf{ERSkill}, a retrieval-centric framework for self-evolving, skill-guided memory access. ERSkill compiles interaction histories into
Multi-parametric magnetic resonance imaging (mpMRI) is a cornerstone for brain tumor diagnosis and treatment, yet current AI models face critical limitations: their lack of natural language interaction and interpretability impedes spatial information integration and cross-modal reasoning required clinically. Key challenges arise from significant physical meaning differences across modalities, spatial misalignment due to scan intervals, and the need for complex multi-feature interpretation in tas
Advances in digital health have dramatically changed how patients engage with their health. Rather than relying solely on periodic clinical visits, patients now have access to smartphones, patient portals, wearable devices, and mobile apps that provide support for day-to-day self-care decisions. This commentary discusses the findings of Longhini et al’s systematic review and meta-analysis on the effectiveness of digital health interventions, which found modest improvements in self-care monitorin
Background: Chronic heart failure (CHF) significantly impairs physical function and quality of life. Although exercise-based cardiac rehabilitation represents a primary therapeutic strategy, participation rates remain low due to logistical barriers. Digital health technologies (DHTs) offer a promising alternative to deliver home-based interventions. However, evidence regarding their specific impact on functional capacity versus daily physical behavior remains inconsistent. Objective: This system
Background: AI has the potential to transform health care in low- and middle-income countries, where access to quality care remains limited. Maternal, sexual, and reproductive health (MSRH) outcomes are especially poor due to resource shortages, financial barriers, and geographic inequities. With thoughtful implementation, AI could help address these gaps through innovations in diagnostics, health education chatbots, and telemedicine. However, responsible use is essential to ensure AI reduces, r
In this retrospective cohort of 5132 users of a commercial nutrition-tracking mobile application, higher food-tracking frequency was associated with greater weight loss over 6 months; 70.1% (3599/5132) of users lost at least 5% of body weight.
Group-robust learning is crucial for maintaining accuracy on rare subpopulations when training-group labels are unavailable. However, existing methods often infer environments from a separate reference model and select representations before fitting the classifier used at deployment, leaving both decisions misaligned with the deployed predictor. In this work, we formulate group robustness without training-group labels as the endogenous environments with repair-aware selection (ERAS) problem, and
Dante Terzigni/theispot.com We live on a blue planet where water appears to be plentiful. But there are strong warning signs indicating that our reliance on the fresh water we perceive to be abundant will need to change. Although water covers 70% of the Earth’s surface, only 0.5% of it is effectively usable. Further, global water […]
Group Relative Policy Optimization (GRPO) learns from reward differences within a rollout group, but receives no useful relative signal when every sampled response is incorrect. Privileged self-distillation can fill this gap with dense token supervision, yet applying it throughout training creates a different failure mode: the teacher is a biased, low-variance surrogate for the reward objective, so persistent imitation can oppose reward-improving updates after the policy becomes capable of produ
Publication date: September 2026 Source: Government Information Quarterly, Volume 43, Issue 3 Author(s): Wu Chen, Haibo Zhang
Publication date: Available online 11 August 2026 Source: International Journal of Human-Computer Studies Author(s): Qin Wu, Jiayu Yao, Haowei Xu, Alireza Farrokhi Nia, Yiqing Zhong, Danielle Lottridge, Suranga Nanayakkara
Publication date: Available online 12 August 2026 Source: Computers in Human Behavior Author(s): Jiajing Zhai, Jiayu Gina Qu
arXiv:2608.11251v1 Announce Type: new Abstract: Fairness in AI systems has become more important with recent regulatory demands, such as the EU AI Act. Traditional approaches often do not take into account philosophical ethics and social awareness. Variable selection processes, in particular, can introduce implicit bias, affecting equity across different subgroups. We discuss a mathematical approach that evaluates fairness in AI, aligning mathematical methodologies with ethical considerations an
arXiv:2608.11259v1 Announce Type: new Abstract: Many AI tutors leverage large language models (LLMs) today. Given that LLMs are opaque black boxes, robust evaluation and live experimentation to measure the impact of every change are essential. We pioneered AI-powered tutoring for K-12 with the launch of Khanmigo (Khan Academy, 2023). We describe the metrics we use to measure AI tutoring quality and student engagement as well as various experiments we have run. We highlight the changes that have
arXiv:2608.11344v1 Announce Type: new Abstract: Financial institutions are delegating consequential decisions to agentic AI systems that decompose goals, coordinate models and tools, and act with little oversight. Yet agentic AI governance in FinTech is under-investigated. We argue the binding governance constraint is not capability but verifiability. We define the Verifiability Gap as the shortfall between the verification delegated authority demands and the explainability and reproducibility r
arXiv:2608.11491v1 Announce Type: new Abstract: Algorithmic systems increasingly rank individuals for access to scarce public resources, from child welfare interventions to cancer treatment referrals. The prevailing fairness frame treats disparity as a property of biased data or deficient models, with remedies through calibration and debiasing. Under structural scarcity, where demand exceeds supply by an order of magnitude, allocation becomes a rationing problem, and the statistical properties o
arXiv:2608.11512v1 Announce Type: new Abstract: The question of whether artificial intelligence will "destroy jobs" is too coarse to guide economic analysis or institutional design. A job is not an indivisible object, and machine cognition is not a uniform substitute for human labor. This paper develops a task-based and institutionally grounded framework for analyzing generative AI as cheap, scalable, and fallible cognition. The relevant margins are exposure, adoption, verification, question sel
arXiv:2608.11794v1 Announce Type: new Abstract: The growing role of AI-generated content and AI-enabled systems in public communication has led regulators to demand clear disclosure of content provenance and AI involvement. But the effects of such disclosures remain uncertain. We test two disclosure approaches in their impact on an AI chatbot's persuasive appeal. In a preregistered experiment, 1,500 UK adults held a short conversation with a persuasive chatbot about one of 60 policy issues. The
arXiv:2608.11803v1 Announce Type: new Abstract: Deployed foundation models are often not static systems, with providers able to modify system behavior through fine-tuning, classifier updates, system prompt revisions, retrieval changes, and routing changes. These updates can be made silently -- that is, without public disclosure, a version increment, or re-evaluation. Such silent updates challenge a core assumption behind current AI governance frameworks that an externally verifiable chain of cus
arXiv:2608.11830v1 Announce Type: new Abstract: The deployment of large language models (LLMs) in mental health contexts raises questions about the relationship between clinical safety and environmental cost. In this paper, we examine this relationship by combining K-Bench clinical safety scores with EcoLogits life-cycle assessment estimates across 47 supported model configurations. We evaluate model performance and environmental impact across four dimensions: energy use, carbon emissions, water
arXiv:2608.11955v1 Announce Type: new Abstract: Large language models are already adept at engaging users in long, emotionally salient conversations across ordinary and existential domains. They are also capable of inducing a potent sense of connection with a human-like entity, even when the user knows their interlocutor is artificial. For some users, these conversations can unsettle assumptions about mind, reality, agency and authority, producing forms of ontological shock and epistemic destabi
arXiv:2608.12104v1 Announce Type: new Abstract: The increasing deployment of autonomous, agentic AI systems challenges traditional accountability mechanisms. Existing research predominantly frames AI accountability gaps as barriers that can be overcome through better standards, transparency, and institutional reform. We argue that this framing is insufficient: certain configurations of actors, systems, and institutions render AI accountability conceptually unachievable regardless of effort. We i
arXiv:2608.12166v1 Announce Type: new Abstract: Algorithm registers have been championed as a means of providing transparency on the use of algorithms in public services. Yet potential publics differ in their expectations of what should be made transparent and how, as well as in their interest in and ability to parse the information currently published in the registers. Moreover, it remains unclear how these instruments can represent the sociotechnical systems in which these algorithms are embed
arXiv:2608.12292v1 Announce Type: new Abstract: An effective large language model (LLM) tutor must often decline to give an answer it could easily produce. In a randomized study, students who used an unguarded chatbot scored higher while practicing but lower on a later test taken without it, whereas a Socratically guarded version of the same model kept the practice gain and removed the later loss [4]. Reliable answer-withholding is therefore central to a tutor's value, yet a capable model presse
arXiv:2608.11245v1 Announce Type: cross Abstract: Online education offers unprecedented scalability and accessibility to global learners from diverse backgrounds, but it often suffers from low engagement and poor long term learning effectiveness. To address these challenges, we introduce AI Tutor, a reinforcement learning based model designed to promote sustainable learning by optimizing both short and longterm learning outcomes. In the short term, AI-Tutor draws on cognitive theory to guide lea
arXiv:2608.11256v1 Announce Type: cross Abstract: Institutions use commercial AI detectors for academic integrity, yet detectors cannot distinguish AI editing from full LLM drafts and may treat both as misconduct. In a controlled study of published English abstracts (four domains; 2013 to 2015 vs. 2023 to 2025), we quantify this policy failure under proxy human/AI labels at tau=0.50. Light "refine abstract only" edits, a proxy for guideline-compliant AI assistance, are flagged at 64 to 80% (Pang
arXiv:2608.11410v1 Announce Type: cross Abstract: Offline reinforcement learning (RL) offers considerable promise for optimizing ICU treatment decisions, yet standard evaluation metrics Mean Squared Error (MSE) and Fitted Q-Evaluation (FQE) assess only behavioral imitation and cannot detect Toxic Mimicry, a failure mode in which agents replicate harmful patterns such as treatment withdrawal during comfort-care transitions. Using the MIMIC-III database, we propose the Counterfactual Clinical Audi
arXiv:2608.11540v1 Announce Type: cross Abstract: The convergence of artificial intelligence (AI), Industrial Internet of Things, cyber-physical systems, and advanced robotics is reshaping manufacturing faster than engineering curricula can adapt, widening the gap between the competencies required on the shop floor and those delivered by traditional engineering and technology education. This paper proposes a Workforce Readiness Level (WRL) framework, which adapts the Technology Readiness Level s
arXiv:2608.11626v1 Announce Type: cross Abstract: This study proposes that firms move along an "organizational technology ladder": adopting one technology transforms hiring and work processes and builds skills and organizational capital that change the cost of adopting subsequent technologies. I study how firms' adoption of remote work technology during the COVID-19 period shaped later uptake of generative AI. Using U.S. job-posting data and an instrumental-variables strategy based on predicted
arXiv:2608.11923v1 Announce Type: cross Abstract: The dissemination and viralization of information on social media has been widely studied from various perspectives, including that of digital activism. On the other hand, disability-related activism has conquered the online environment, thus obtaining a reach that goes beyond the offline space and generating dialogue in the digital sphere. This article analyses the conversation generated on Twitter, taking as a sample all the tweets with the #di
arXiv:2608.12278v1 Announce Type: cross Abstract: Artificial intelligence tools for education and language support are increasingly framed as scalable responses to access gaps in under-resourced communities. Yet the infrastructure underlying these tools, including training corpora, tokenization schemes, evaluation benchmarks, and deployment architectures, can systematically disadvantage speakers of underrepresented languages before a model is trained. This paper examines these structural barrier
arXiv:2507.11773v2 Announce Type: replace Abstract: The emergence of breakthrough artificial intelligence (AI) techniques has led to a renewed focus on how small data settings, i.e., settings with limited information, can benefit from such developments. This includes societal issues such as how best to include under-represented groups in data-driven policy and decision making, or the health benefits of assistive technologies. We provide a conceptual overview, clarify the relationship between sma
arXiv:2508.05849v3 Announce Type: replace Abstract: The proliferation of misinformation on social media has concerning possible consequences, such as the degradation of democratic norms. While recent research on countering misinformation has largely focused on analyzing the effectiveness of interventions, the factors associated with public support for these interventions have received little attention. We asked 1,010 American social media users to rate their support for and perceptions of ten mi
arXiv:2508.09219v3 Announce Type: replace Abstract: Recent advances in AI applications have raised growing concerns about the need for ethical guidelines and regulations to mitigate the risks posed by these technologies. In this paper, we present a mixed-methods survey study - combining statistical and qualitative analyses - to examine the ethical perceptions, practices, and knowledge of individuals involved in various AI development roles. Our survey comprises 414 participants from 43 countries
arXiv:2509.15122v2 Announce Type: replace Abstract: Large language models (LLMs) play a growing but largely informal role in scholarly peer review. Yet whether LLMs reproduce biases observed in human decision-making remains unclear. We adapt a resume-style audit to scientific publishing, developing a multi-role LLM simulation (editor/reviewer) that evaluates high-quality manuscripts across the physical, biological, and social sciences under randomized author identities (institutional prestige, g
arXiv:2510.10315v4 Announce Type: replace Abstract: Large Language Models (LLMs) are increasingly relying on web crawling to stay up to date and accurately answer user queries. These crawlers are expected to honor robots.txt files, which govern automated access. In this study, for the first time, we investigate whether reputable news websites and misinformation sites differ in how they configure these files, particularly in relation to AI crawlers. Analyzing a curated dataset, we find a stark co
arXiv:2603.00056v2 Announce Type: replace Abstract: STEM Mental models can play a critical role in assessing students' conceptual understanding of a topic. They not only offer insights into what students know but also into how effectively they can apply, relate to, and integrate concepts across various contexts. Thus, students' responses are critical markers of the quality of their understanding and not entities that should be merely graded. However, inferring these mental models from student an
arXiv:2605.23162v2 Announce Type: replace Abstract: Distributed solar markets must coordinate physical reports, economic allocation, and public settlement even when IoT data can be manipulated. We present SolarChain, a controlled Embodied Intelligence of Things (EIoT) prototype that integrates four functions: physics-bounded screening of photovoltaic reports, persistent agent and planner coordination, configurable allocation between producer rewards and market liquidity, and replayable hash-link
Chronic Kidney Disease (CKD) has emerged as a major public health concern worldwide, and most patients with CKD are asymptomatic until the later stages, causing growing morbidity and mortality. Diabetes and hypertension are the main causative factors for the development of CKD, damaging the renal microcirculation system. In addition, the impact of Acute Kidney Injuries (AKI) may result in the recovery or progression to either CKD or renal failure. The conventional techniques for diagnosis, such
Digital systems increasingly function as tempo-setting infrastructures that organize the temporal conditions under which cognition, communication, learning, and participation occur. Although research in human-computer interaction, platform studies, and AI ethics has extensively examined privacy, fairness, transparency, engagement, and well-being, the temporal organization of digital life has received comparatively little attention as a distinct object of sociotechnical analysis. We argue that di
As technology advances, artificial intelligence is increasingly integrated into educational settings, raising important questions about its impact on learning and academic integrity. This study describes the perspectives of Filipino nurse educators on the role of artificial intelligence in maintaining academic integrity. A descriptive qualitative design was used, with a two-phase data collection process employing purposive sampling. In Phase 1, open-ended online responses were collected via Goog
LLM coding agents issue Bash commands through interfaces that may serialize, wrap, and reparse model output. Matched execution scores alone cannot distinguish command-generation errors from failures introduced after generation. QuoteBench measures this boundary with exact final-state validation on 56 one-shot tasks from 14 incident-derived families, crossing the generation contract with the execution transport around one deliberately unescaped added parser. Escaping at the interpolation point re
Modern language models are trained on heterogeneous web-scale text corpora. Consequently, studying knowledge and skill acquisition is difficult, as prior exposure to related content is hard to characterize. To address this challenge, we introduce LITTLECURRICULUM, a curated 88B-token pretraining corpus tailored to U.S. elementary school material, explicitly excluding concepts, facts, and vocabulary taught above Grade 5. Training a 5B-parameter LLM from scratch on LITTLECURRICULUM yields LITTLELE
AI agents are increasingly used for programming, but do not provide any guarantee on the correctness of generated code. Verified code generation, in which an agent produces both an implementation and a machine-checked proof of its specification, offers a stronger path toward trustworthy AI-generated software. Existing benchmarks in this direction either focus on individual functions or only evaluate proof generation with provided implementations. It is still an open question whether agents can m
We present Multi-Agent Reasoning and Coordination (MARC), an open-source framework that replaces monolithic LLM prompting with deterministic multi-agent orchestration for clinical reasoning. MARC coordinates role-specialized agents for extraction, reasoning, answer generation, and evaluation, with explicit context passing and traceable intermediate outputs, enabling stage-wise failure attribution. We additionally introduce a Decomposer module that generates task-specific agent prompts from a pla
Modern image classification models excel when trained on single task-specific datasets but often struggle to generalize across domains and difficulty levels. We propose ARMDIL, an Adaptive Router for Multi-Domain Image classification with LLMs. ARMDIL is an ensemble that uses a multimodal large language model (MLLM) agent to dynamically route each image to the most suitable vision backbone. Our diverse ensemble employs convolutional neural networks (ResNets), self-supervised representation learn
We address the use of large language models (LLMs) to help discover Isabelle proofs. An Isabelle build establishes that the submitted theory is accepted, but not that an LLM changed only what the developer authorised. We present CAPRI, a contract-aware repair workflow in which Isabelle checks the proof and an independent checker enforces a machine-readable edit contract. Prompts, proposals, candidate repositories, diagnostics, verdicts, and hashes are retained for audit. We evaluate five workflo
Contact-rich manipulation failures are often detected only after the robot has committed to contact. This is especially limiting in wrist-camera setups: close gripper--object views help observe contact, but a poor approach may already push, miss, slip, or disturb the object before conventional detectors react. We introduce \emph{ContactGuard}, a pre-contact execution monitor for chunked visuomotor policies. Given the policy's planned action chunk, ContactGuard predicts its short-horizon conseque
Assessing the maturity of artificial intelligence technologies is essential for investment decisions, project management, and policy monitoring, yet the available readiness frameworks are heterogeneous and difficult to apply automatically: the adaptation of Technology Readiness Levels to AI lacks AI-specific gating criteria, the Machine Learning Technology Readiness Levels presuppose access to internal process artifacts, and AI/data readiness dimension models employ scales that resist direct com
Embodied intelligent virtual agents are expected to operate as persistent, adaptive, and context-aware entities within complex virtual and Metaverse worlds. However, implementing cognitively capable agents in such environments is conceptually and technologically challenging. Among a range of blueprints and development approaches, the Cognitive Embodied Agent Architecture (CEAA) has been developed as an implementation-oriented framework for architecting components of perception, memory, reasoning
Autonomous agents are increasingly capable of improving models, systems, and other technical artifacts through long-horizon experimentation. To understand the current state of this capability, however, evaluation must go beyond final scores, which neither reveal where progress is gained or lost nor indicate whether accumulated experience improves later decisions. We therefore present a systematic evaluation of seven frontier models on 36 long-horizon tasks based on a new framework that uses rule
We consider the problem of autonomously learning robot skills under a limited practice budget for sequential tasks. We propose an active skill learning algorithm, \emph{Deliberate Practice (DP)}, that computes a provably \emph{budget-optimal} allocation---practicing skills that maximize expected cumulative reward while being learnable within the budget. DP estimates both the time needed to master skills and the cumulative reward of the task plans that the skills unlock. Computing a budget-optima
Existing predictive models in learning analytics often treat student academic history as a simple sequence, overlooking the concurrent nature of courses taken within a semester. This simplification can lead to inaccurate performance predictions, particularly for students with heavy or challenging course loads. This paper introduces a TRansformer for Academic Course-grade Estimation (TRACE) that addresses this limitation by jointly predicting both the set of courses a student will take and their
This preliminary technical report presents a framework for sign language video synthesis using a loss-guided multi-expert Generative Adversarial Network (GAN) to enhance communication for individuals with hearing impairments. Three specialized discriminators -- global, hand, and head -- each guide a corresponding expert branch in the generator toward a distinct visual region, enabling implicit feature specialization without explicit diversity losses. To stabilize this multi-discriminator system,
Professional communication is increasingly mediated by LLMs - but do these models serve all users equally? We show that when prompts contain linguistic features more commonly used by women (hedges, tag questions, collective reference), they systematically elicit shorter, less sophisticated, and less formal responses across three document types and four models. These effects persist after controlling for prompt complexity and feature carry-over. Explicit gender cues like sign-off names are encode
We analyzed the ethics reporting in 255 IEEE VIS papers from 2024 and 2025, as published in TVCG. This analysis arose from our experience as readers and reviewers of IEEE VIS papers that such reporting is frequently incomplete or missing, as well as from investigations in which we ourselves had to answer challenges regarding ethics approval in our own work. Visualization research naturally often involves human participants, yet ethics approval and informed-consent procedures are not always expli
Understanding motion in daily living requires context beyond kinematics, because similar inertial patterns during activities of daily living (ADLs) can reflect intentional stopping, object interaction, or pathological movement impairment. Egocentric vision provides task-related context that may help disambiguate these cases. We investigate this challenge through freezing of gait (FOG) detection in Parkinson's disease (PD), a symptom strongly influenced by contextual factors during ADLs. Using sy
Existing vision-language model (VLM) benchmarks emphasize perception and reasoning accuracy (how well VLMs describe and reason about what they see in an image), with limited attention to behavioral reliability under uncertainty (how they behave when visual evidence is missing or misleading). We introduce SciFigBench, a diagnostic VLM benchmark for scientific figure understanding that jointly evaluates perception, reasoning, and behavioral reliability under uncertainty. It contains 250 figures wi
While the detrimental impacts of driving under the influence of stimulants such as methamphetamine are well-documented, the driving performance of individuals currently under-treatment has received considerably less attention. This study compared the behavior of individuals with a history of stimulant abuse (across two distinct treatment phases) with a control group of healthy drivers using a driving simulator. Oculomotor and biomechanical data were continuously collected via an eye-tracker and
Long-form research reports generated by large language models drift, contradict themselves, and lose provenance: the same metric appears with different values, and rumor is quoted as confidently as an audited filing. We present a two-tier agentic system that separates a maintained, point-in-time knowledge library from report writing. A deterministic "librarian" ingests timestamped sources into a trust-tiered ontology, layering evidence cards, an authoritative metric ledger, and a claim graph int
Vertical Federated Learning (VFL) enables organizations holding complementary features of shared entities to collaborate and train models. In this setting, the initiator can withhold information about the learning task, while other contributors participate without exposing their local datasets, creating an asymmetric information structure aligned with growing privacy demands. However, this asymmetry is a double-edged sword. Among various threats, backdoor attacks are particularly concerning beca
Presentations are essential for students, researchers, and professionals to communicate ideas persuasively, yet delivering them effectively requires repeated practice that coordinates content, delivery, visual materials, and audience interaction. Existing AI-assisted rehearsal tools provide scalable feedback, but they often treat presentations as single-run delivery performances, offering limited support for linking feedback to the slide deck or planning what to practice in the next iteration. T
Large language models (LLMs) remain vulnerable to harmful requests and jailbreak attacks. Parameter-efficient safety alignment methods based on prompt tuning typically rely on a single global prompt or externally selected prompt modules. Such static designs struggle to maintain a cross-category safety boundary while generating constructive responses tailored to specific risks and avoiding over-refusal of benign inputs. To address these limitations, we propose HiRoute, an input-adaptive hierarchi
Contribution statements are an increasingly common way to make research labor visible, reduce academic malfeasance, and provide broader transparency. Despite this potential value, they remain uncommon in visualization and HCI. To explore this gap, we conducted an online study with (N=21) visualization and HCI researchers. We find a range of differing opinions about the utility of contribution statements, which are set against a background of tensions relating to contribution frameworks that inad
The ability to navigate outdoors safely and independently is crucial yet challenging for people with low vision (PLV). While various augmented reality (AR) systems for low vision have been designed and evaluated in ideal lab environments, no research has investigated their real-world feasibility and challenges. We present NavSight, a mobile AR application that assists PLV in outdoor navigation by recognizing important outdoor objects (e.g., curb, vehicle) and rendering real-time visual augmentat
LLM-based simulated clients are increasingly used to train novice counselors, evaluate LLM therapists, and generate synthetic data. However, current simulators produce overly cooperative clients that disclose too readily, accept therapeutic reframes without resistance, and resolve core issues within a single session. We trace these issues to profiles that lack causal depth and behavioral mechanisms that treat all content as equally accessible. We present PatientAct, a framework for client simula