Welcome to the March 17, 2025 edition of ACM TechNews, providing timely information for computer professionals three times a week.
|
|
More than 25% of all computer programming jobs in the U.S. have disappeared in the past two years, with the occupation now ranking among the 10 hardest-hit of 420-plus jobs for which the U.S. Bureau of Labor Statistics provides data. There are fewer programmers today in the U.S. than at any point since 1980. At the same time, jobs for software developers have fallen just 0.3%, similar to the broader industry.
[
» Read full article
*May Require Paid Registration
]
The Washington Post; Andrew Van Dam (March 14, 2025)
|
AI is fueling the most fundamental change to computing since the early days of the Internet. Just as companies completely rebuilt their computer systems to accommodate the new commercial Internet in the 1990s, they are now rebuilding from the bottom up, wiring together up to 100,000 chips to create powerful AI systems. The industry is also looking at new ways to house, power, and cool these systems to keep them from over-heating.
[
» Read full article
*May Require Paid Registration
]
The New York Times; Cade Metz; Karen Weise; Marco Hernandez (March 16, 2025); et al.
|
In an advisory posted last week, U.S. cybersecurity officials warned of a ransomware-as-a-service software called Medusa, which uses phishing campaigns as its main method for stealing victims’ credentials. Active since 2021, Medusa actors use a double extortion model, where they “encrypt victim data and threaten to publicly release exfiltrated data if a ransom is not paid,” according to the advisory. Since February, Medusa actors have hit more than 300 victims across various industries.
[
» Read full article
]
Associated Press; Sarah Parvini (March 15, 2025)
|
At the South by Southwest conference, World Wide Web inventor Tim Berners-Lee, an ACM A.M. Turing Award laureate, raised the question of who AI works for. Even if AI models are reliable, accurate, and unbiased, there will be concerns about whether company or user interests are paramount. Said Berners-Lee, "I want AIs to work for me to make the choices that I want to make. I don't want an AI that's trying to sell me something."
[
» Read full article
]
CNet; Jon Reed (March 12, 2025)
|
The new Exo 2 programming language, developed by researchers at the Massachusetts Institute of Technology, is a user-schedulable language in which programmers write "schedules" that control how a compiler generates code. The researchers created a scheduling library with about 2,000 lines of code in Exo 2, consolidating scheduling efforts across more than 80 high-performance kernels, each with up to 12 lines of code. Its performance was on par with or exceeded that of MKL, OpenBLAS, BLIS, and Halide.
[
» Read full article
]
MIT News; Adam Conner-Simons (March 13, 2025)
|
University of Science and Technology of China's Yin Juan announced the first quantum satellite communication link between Beijing and South Africa at the National People's Congress recent annual plenary meeting. The demonstration, which spanned 12,800 kilometers (7,954 miles), marked the southern hemisphere's first secure quantum key distribution experiment. Yin attributed the success of the demonstration to "achieving real-time secure communication between low-cost quantum micro-nano satellites and mobile ground stations."
[
» Read full article
]
South China Morning Post; Victoria Bela (March 13, 2025)
|
At around the same time researchers at D-Wave publicly announced they had achieved "quantum supremacy" in simulating magnetic materials, researchers at the Flatiron Institute released a paper explaining how they repurposed the 40-year-old belief propagation algorithm to solve a subset of the same problem using a classical computer. “For the … spin glass problem at hand, our classical approach demonstrably outperforms other reported methods,” the group wrote. “In [two cases] we are also able to reach errors noticeably lower than the quantum annealing approach employed by the D-Wave Advantage2 system.”
[
» Read full article
]
ScienceNews; Mara Johnson-Groh (March 12, 2025)
|
Later this month, Google will roll out a Pixel Watch 3 feature that can detect when users no longer have a pulse. Detection of a possible loss of pulse activates an infrared light, which penetrates deeper into the skin than the standard pulse-detection green light. Another algorithm determines whether any pulse detected matches the normal heart rhythm. An irregular haptic buzz also is triggered, and 911 is called if the user does not move within 35 seconds.
[
» Read full article
]
IEEE Spectrum; Elissa Welle (March 11, 2025)
|
San Francisco's Municipal Transportation Agency said robotaxis operated by Alphabet's Waymo racked up 589 tickets and $65,065 in fines for parking violations last year. The company's autonomous vehicles were cited for such violations as obstructing traffic, disobeying street cleaning restrictions, and parking in prohibited areas.
[
» Read full article
*May Require Paid Registration
]
The Washington Post; Lisa Bonos (March 13, 2025)
|
Of the U.S. tech jobs posted since January, almost 25% seek workers with AI skills, according to the University of Maryland's (UMD's) AI job tracker. AI-related listings accounted for 1.3% of all job postings in January, compared with tech job listings at 5.4%. According to UMD's Anil K. Gupta, the turning point for the AI job market was the launch of OpenAI's ChatGPT, which bumped AI-related job postings 68% from its fourth-quarter 2022 launch through the end of 2024. Over the same period, tech job postings declined 27%.
[
» Read full article
*May Require Paid Registration
]
The Wall Street Journal; Nate Rattner (March 10, 2025)
|
Last year, 660 security researchers who reported security bugs through Google's Vulnerability Reward Program (VRP) received a total of nearly $12 million in bug bounty rewards. The company says it has awarded $65 million in bug bounties since its first vulnerability reward program went live in 2010.
[
» Read full article
]
BleepingComputer; Sergiu Gatlan (March 10, 2025)
|
Researchers at startup Lila Sciences developed a generative AI program trained on published and experimental data, the scientific process, and reasoning, in the quest for "scientific superintelligence." The AI is tasked with generating new ideas and testing them in automated labs with a handful of human assistants. Said Lila cofounder Molly Gibson, “Our goal is really to give AI access to run the scientific method—to come up with new ideas and actually go into the lab and test those ideas.”
[
» Read full article
*May Require Paid Registration
]
The New York Times; Steve Lohr (March 10, 2025)
|
|