Welcome to the June 9, 2025 edition of ACM TechNews, providing timely information for computer professionals three times a week.
|
|
U.S. President Trump on Friday signed an executive order rolling back cybersecurity policies set in place by previous administrations. Trump's order removed requirements around testing the use of AI to defend energy infrastructure, funding federal research programs around AI security, and directing the Pentagon to “use AI models for cyber security.” Trump’s order also removed requirements that agencies start using quantum-resistant encryption “as soon as practicable” and that federal contractors attest to the security of their software.
[
» Read full article
]
TechCrunch; Anthony Ha (June 7, 2025)
|
The OpenWebSearch.eu consortium, made up of 14 leading research institutions from across Europe, has released a pilot of a new infrastructure that aims to make Web search fairer, more transparent, and commercially unbiased. The European Open Web Index (OWI) project aims to build a public Web index as an alternative to indexes held by companies like Google and Microsoft in the U.S., Baidu in China, and Yandex in Russia. Europe currently does not have a search index of its own, making it vulnerable to digital dependence.
[
» Read full article
]
CERN (Switzerland) (June 6, 2025)
|
Some of Alan Turing's most important works, including a personal copy of his Ph.D. dissertation, is going up for auction June 17. The papers are "offprints," or copies of papers distributed in small batches within academia, and also include "On Computable Numbers," considered the first computer programming manual, and "The Chemical Basis of Morphogenesis," Turing's last major published work.
[
» Read full article
]
LiveScience; Stephanie Pappas (June 6, 2025)
|
A letter to the U.S. Congress signed by 260 state lawmakers from both parties calls for the removal of a provision from President Trump's "big, beautiful bill" that would impose a decade-long moratorium on state-level AI regulations. The state lawmakers said in their letter, "Over the next decade, AI will raise some of the most important public policy questions of our time, and it is critical that state policymakers maintain the ability to respond."
[
» Read full article
*May Require Paid Registration
]
The Washington Post; Will Oremus; Andrea Jiménez (June 3, 2025)
|
Bill Atkinson, the Apple Computer designer who created the software that enabled the visual approach pioneered by the company’s Lisa and Macintosh computers, died on Thursday at 74. Atkinson programmed QuickDraw, a software layer used for both computers that made it possible to display shapes, text, and images on screens efficiently. Atkinson is credited with such key aspects of graphical computing as “pull down” menus and the “double-click” gesture of a mouse.
[
» Read full article
*May Require Paid Registration
]
The New York Times; John Markoff (June 9, 2025)
|
OpenAI's latest report on malicious uses of its AI models states that a “significant number” of recent violations came from China. The ChatGPT developer said it had disrupted several attempts to leverage its models for cyber threats and covert influence operations in the three months since Feb. 21.
[
» Read full article
*May Require Paid Registration
]
The Wall Street Journal; Mauro Orru (June 6, 2025)
|
A new simulation method developed by researchers at the Massachusetts Institute of Technology allows animators to simulate rubbery and elastic materials in a way that preserves the physical properties of the material. The researchers uncovered a hidden mathematical structure in equations that capture how elastic materials deform on a computer. Explained MIT's Leticia Mattos Da Silva, "Our method aims to stay true to physical laws while giving more control and stability to animation artists.”
[
» Read full article
]
MIT News; Adam Zewe (June 6, 2025)
|
An international team led by researchers at Estonia's University of Tartu Institute of Computer Science has come up with a method of repurposing old smartphones to create miniature datacenters. The method involves removing the phones’ batteries, using external power sources to prevent chemical leaks, and connecting four devices together with 3D-printed casings, resulting in a functional prototype datacenter costing only around 8 euros (around $9) per device. In an underwater test, the prototype helped monitor marine life by automatically counting sea species.
[
» Read full article
]
The University Network (June 6, 2025)
|
A report from the U.N. International Telecommunication Union revealed a 150% average increase in indirect carbon emissions from four leading AI-focused tech companies (Amazon, Microsoft, Alphabet, and Meta) between 2020 and 2023. The spike is largely attributed to the high energy demands of AI systems and datacenters. The report warned that unchecked AI expansion could result in annual emissions reaching 102.6 million tons of CO2 equivalent, comparable to the yearly emissions of some mid-sized industrialized nations.
[
» Read full article
]
Computing (U.K.); Dev Kundaliya (June 6, 2025)
|
Amazon is reportedly developing AI software for humanoid robots that could perform the role of delivery workers. Testing is taking place at an obstacle course at an Amazon facility in San Francisco, according to tech news site The Information, citing an insider. Amazon envisions the robots traveling in its Rivian vans, one of which is being used in the testing area.
[
» Read full article
]
The Guardian (U.K.); Dan Milmo (June 5, 2025)
|
A gesture-based AI system developed by researchers at Taiwan's National Penghu University of Science and Technology could help educators better manage large groups of students. Each student is given a device with a screen and webcam that uses the Mediapipe gesture-tracking library to track 21 joint points in the hand as they write or draw in midair using their fingers. The student's work appears on their individual screen, as well as on a larger screen at the front of the classroom.
[
» Read full article
]
IEEE Spectrum; Michelle Hampson (June 5, 2025)
|
A team of researchers from German, Polish, and Romanian universities and vendor Resemble AI found that re-recording deepfake audio with natural acoustics in the background allows the audio to bypass detection models. By generating synthetic speech, playing it, and rerecording it with background noise, researchers found "significant vulnerability" with equal error rates in the top-performing detection model jumping from 4.7% to 18.2%. The reason, according to the researchers, is that replaying audio "removes key artifacts relied upon by detection models."
[
» Read full article
]
Dark Reading; Alexander Culafi (June 4, 2025)
|
|