STEM News | Issue 101, 1-8-2026

A fundamental flaw leaves LLMs strikingly vulnerable to attack

Published in MIT Technology Review — 30 Jul 2026 — by Will Douglas Heaven

OpenAI called the Hugging Face attack unprecedented. But we’ve been here before.

Published in MIT Technology Review — 27 Jul 2026 — by Will Douglas Heaven

A Hacker’s Arrest Reveals Microsoft Can Track Users Via a Windows Device ID

Published in PC Magazine — 6 July 2026 — Michael Kan

The Power of 10: New Rules for the Digital World

Published in ACM Communications — 24 Jun 2026 — by Sarah Spiekermann Hoff, Marc Langheinrich, Johannes Hoff, et.al.

Artificial Intelligence for Software Engineering: From Probable to Provable

Published in ACM Communications — 26 May 2026 — by Bertrand Meyer

AI powers citizen-led disaster relief from afar for Venezuela

Published in Rest Of World — 15 Jul 2026 — by Daniella Fernández, Andrea Paola Hernández

China Just Dropped Another Bomb on America’s Frontier AI Companies

Published in Gizmodo — 16 Jul 2026 — by Ece Yildirim

AI can predict how you’ll respond to a survey. But that’s not the same as understanding you

Published in The Conversation — 9 Jul 2026 — by Steve Bickley

Landmark German ruling declares Google’s AI Overviews are Google’s own words and makes it liable for false answers

Published in The Decoder — 11 Jun 2026 — by Matthias Bastian

Emily Bender Sets the Record Straight on “Stochastic Parrots”: Five years later, its lead author revisits the paper in the age of ChatGPT

Published in IEEE Spectrum — 30 Jun 2026 — by Gwendolyn Rak

10 things that matter in AI right now

Published in MIT Technology Review — 21 Apr 2026 — Editorial

Open-Source LLMs in 2026: The Free AI Models Everyone Will Be Using While You’re Still Overpaying

Published in Medium — 6 Jul 2026 — by Anurag Goel

Standalone Magnetometry Is the New GPS: Companies are turning the tiny compasses in phones into full navigation alternatives

Published in IEEE Spectrum — 2 Jun 2026 — by Lucas Laursen

Google loses fight over record $4.7 billion EU antitrust fine

Published in CNBC — 2 Jul 2026 — by Arjun Kharpal

World Cup technology: from ref cams to AI analysts, cutting‑edge research is changing the game

Published in The Conversation — 22 Jun 2026 — by Thomas Allen

How should European scientists respond to the US AI kill switch?

Published in Science|Business — 25 Jun 2026 — by David Matthews

Anthropic to disable its most advanced AI models after US order limiting foreign access

Published in The Guardian — 13 Jun 2026 — by Reuters

Anthropic suspends new AI tools over US government security concerns

Published in BBC — 13 Jun 2026 — by Harry Sekulichand, Shiona McCallum

The AI layoff wave is becoming a powder keg

Published in TechChurch — 15 Jun 2026 — by Connie Loizos

How JPL Keeps the 13-Year-Old Curiosity Rover Doing Science

Published in IEEE Spectrum — 9 Jun 2026 — by Evan Ackerman

Version of AI tool ‘too powerful for public’ released to public

Published in BBC — 9 Jun 2026 — by Kali Hays

UN report warns AI could soon use 3% of world’s electricity and more water than we need to drink

Published in The Conversation — 4 Jun 2026 — by Amanda Turnbull-McRae

Inside the Software Making Electric Heavy Trucks Practical

Published in IEEE Spectrum — 27 May 2026 — by Benjamin Skuse

Democrats Want a Military AI Restriction Law Following Anthropic’s Pentagon Fallout

Published in Gizmodo — 9 Jun 2026 — by Ece Yildirim

Why Orbital Data Centers Are Harder Than Silicon Valley Thinks

Published in IEEE Spectrum — 11 Jun 2026 — by Andrew Cavalier

Trump signs AI safety order seeking voluntary review of new models

Published in NPR — 2 Jun2026 — by Shannon Bond, Deepa Shivaram

OpenAI just released its answer to Claude Mythos

Published in The Verge — 12 May 2026 — by Stevie Bonifield

AI Models Trained on Physics Are Changing Engineering Large physics models are increasingly used to bypass simulation

Published in IEEE Spectrum — 9 Apr 2026 — by Dina Genkina

Server Failures Turn Safe Cars Into Costly Dead Weight Cybersecurity incident leaves hundreds of U.S. drivers stranded

Published in IEEE Spectrum — 15 Apr 2026 — by Willie D. Jones

America’s coming war over AI regulation

Published in MIT Technology Review — 23 Jan 2026 — by Michelle Kim

What’s next for Chinese open-source AI

Published in MIT Technology Review — 12 Feb 2026 — by Caiwei Chen

Millions of AI agents imperiled by critical vulnerability in open source package

Published in Ars Technica — 26 May 2026 — by Dan Goodin

Researchers let AI models run a simulated society. Claude was the safest—and Grok committed 180 crimes and went extinct within 4 days

Published in Fortune — 28 May 2026 — by Jake Angelo

The Price of Intelligence: Can AI Afford to Be Sustainable?

Published in IEEE Computer — May 2026 — by Antonio Mastropaolo

Optical Fiber Networks Can Keep Rail Networks Safe Distributed acoustic sensing uses existing fiber to measure unexpected vibrations

Published in IEEE Spectrum — 16 Apr 2026 — by Michelle Hampson

AI models that lie, cheat and plot murder: how dangerous are LLMs really?

Published in Nature — 8 Oct 2025 — by Matthew Hutson

Google says it likely thwarted effort by hacker group to use AI for ‘mass exploitation event’

Published in CNBC — 11 May 2026 — by Samantha Subin

What Anthropic’s Mythos Means for the Future of Cybersecurity

Published in Schneier on Security — 28 April 2026 — by Barath Raghavan

Mythos and Cybersecurity

Published in Schneier on Security — 17 April 2026 — by David Lie

What Anthropic’s Mythos Means for the Future of Cybersecurity

Published in IEEE Spectrum — 23 Apr 2026 — by Bruce Schneier, Barath Raghavan

AI in the emergency department: promising, powerful but still unproven

Published in The Conversation — 7 May 2026 — by Ewen Harrison

600 Google employees are urging CEO Sundar Pichai to block classified Pentagon AI deals

Published in Quartz — 27 Apr 2026 — by Colleen Cabili

Study Finds A Third of New Websites are AI-Generated

Published in 404 Media — 27 Apr 2026 — by Matthew Gault

AI’s energy obsession just got a reality check

Published in MIT Technology Review — 28 Jan 2025 — by James O’Donnell

China’s AI upstart DeepSeek drops new model. Will it make waves like last year?

Published in CNN — 24 Apr 2026 — by John Liu

AI is making workers more productive. Their paychecks aren’t keeping up

Published in Quartz — 21 Apr 2026 — by Quartz Staff

No humans allowed: scientific AI agents get their own social network

Published in Nature — 20 Apr 2026 — by Jenna Ahart

How AI agents will change research: a scientist’s guide

Published in Nature — 3 Oct 2025 — by Elizabeth Gibney

AI economics for the common good

Published in Nature Machine Intelligence — 17 Apr 2026 — by Francesco Fuso Nerini

From embodied intelligence to physical AI

Published in Nature Machine Intelligence — 24 Apr 2026 — by Editorial

Claude Mythos and Project Glasswing: why an AI superhacker has the tech world on alert

Published in The Conversation — 13 Apr 2026 — by Stan Karanasios, Saeed Akhlaghpour

OpenAI rips Anthropic, distances itself from Microsoft

Published in Axios — 13 Apr 2026 — by Nathan Bomey

Google, Microsoft, Meta All Tracking You Even When You Opt Out, According to an Independent Audit

Published in 404Media — 14 Apr 2026 — by Matthew Gault

The AI Revolution in Math Has Arrived

Published in Quanta Magazine — 13 Apr 2026 — by Konstantin Kakaes

Why having “humans in the loop” in an AI war is an illusion

Published in MIT Technology Review — 16 Apr 2026 — by Uri Maoz

Horizon Europe applications are skyrocketing. Is AI to blame?

Published in Science|Business — 11 Nov 2025 — by Goda Naujokaitytė, David Matthews

AI Coding Assistants Are Getting Worse: Newer models are more prone to silent but deadly failure modes

Published in IEEE Spectrum — 8 Jan 2026 — by Jamie Twiss

Meet the early-adopter judges using AI: As the line between helping and judging blurs, the cost of errors is steep

Published in MIT Technology Review — 11 Aug 2026 — by James O’Donnell

Government by AI? Trump Administration Plans to Write Regulations Using Artificial Intelligence

Published in ProPublica — 26 Jan 2026 — by Jesse Coburn

AI Boosts Research Careers but Flattens Scientific Discovery: New analysis suggests AI tools narrow the span of ideas explored

Published in IEEE Spectrum — 19 Jan 2026 — by Elie Dolgin

Data Analysis: Why Is It So Complicated?

Published in ACM Queue — 13 Feb 2026 — by Alice Jackson

NASA Let AI Drive the Perseverance Rover

Published in IEEE Spectrum — 15 Feb 2026 — by Evan Gough

Exclusive: Pentagon threatens Anthropic punishment

Published in Axios — 16 Feb 2026 — by Dave Lawler, Maria Curi, Mike Allen