Washington Post columnist Dana Milbank calls himself "a highly creative hypochondriac" — who just paid for an expensive MRI scan to locate abnormal spots as tiny as 2 millimeters.

He discusses the pros and cons of its "diffusion-weighted imaging" technology combined with the pattern recognition of AI, which theoretically "has the potential to save our lives by revealing budding cancers, silent aneurysms and other hidden would-be killers before they become deadly. " But the scans cost $2,500 a pop and insurance won't pay. Worse, for every cancer these MRIs find, they produce a slightly greater number of false positives that require a biopsy, with the potential for infection and bleeding and emotional distress. Even when the scans don't produce a false positive, they almost always come up with some vague and disconcerting abnormality.... Will we feel better after viewing our insides? Or will we become anxious about things we hadn't even thought to worry about?

Part of living has always been in the mystery, in not knowing what tomorrow will bring. Now, because of sophisticated imaging, genome sequencing and other revolutionary screening tools, we can have predictability, or at least the illusion of it. But do we want that? The American College of Radiology says we do not. Its still-current 2023 statement says there is not "sufficient evidence" to recommend full-body screening, cautioning that the scan could lead to needless testing and expense. But David Larson, chair of ACR's Commission on Quality and Safety, told me that could change as more data comes in. "When people ask me, 'Would you recommend it?' I would say it depends on your tolerance for ambiguity," he said, giving the example of somebody found to have a borderline aortic aneurysm who is advised to wait and monitor it. If "that won't keep you up at night, then I wouldn't necessarily recommend against it...."

About 1 in 20 gets that dreaded call. A study Prenuvo presented earlier this year of 1,011 participants found that 4.9 percent of scans required a follow-up biopsy. Of those, 2.2 percent were actually cancer, and the other 2.7 percent were false positives. Of the 22 cancers the scans caught, 86 percent of patients had no specific symptoms. But if finding something truly awful is rare, finding something abnormal is almost guaranteed. [Vikash Modi, Prenuvo's senior medical director of preventative medicine] said only 1 in 20 scans come back completely clean. The vast majority of patients wind up in the ambiguous realm where something may look suspicious but doesn't require urgent follow-up.
He opted for the cheaper $1,000 torso scan, which the senior medical director calls "our bread-and-butter area," since 17 of the 22 cancers detected in one Prenuvo study were in that area and is where they often find cancers that wouldn't be discovered until they were incurable like "that scary pancreatic stuff...."

Milbank's scan found 12 "abnormalities" included "a 2.5 mm pulmonary nodule in the right lower lobe" and "a 4.6 mm intraductal papillary mucinous neoplasm in the pancreatic tail" — but with 10 abnormalities labeled "minor" (and six being musculoskeletal wear-and-tear problems "I already knew about from the usual aches and pains".) Even the two "moderate" findings didn't sound that grim when I read on. The "indeterminant lesion" in my lung requires no follow-up, while the thing in my pancreas is "low-risk."... The "most interesting" finding was the pancreatic cyst, because, at this size and location, there's a 3 percent chance it will become cancerous in the next five years. But if annual follow-up scans of my pancreas (covered by insurance) show it's getting bigger, the cyst can be removed before it becomes cancer. For me, this made the MRI worthwhile. Sure, there was a 97 percent likelihood the cyst never would develop into a problem even if I hadn't learned about it. But now, with minimal inconvenience, I can eliminate that 3 percent risk of getting pancreatic cancer, the most lethal of major malignancies.

Technology writer Matthew Hutson (also Slashdot reader #1,467,653) looks at a new kind of self-improving AI coding system. It rewrites its own code based on empirical evidence of what's helping — as described in a recent preprint on arXiv.

From Hutson's new article in IEEE Spectrum: A Darwin Gödel Machine (or DGM) starts with a coding agent that can read, write, and execute code, leveraging an LLM for the reading and writing. Then it applies an evolutionary algorithm to create many new agents. In each iteration, the DGM picks one agent from the population and instructs the LLM to create one change to improve the agent's coding ability [by creating "a new, interesting, version of the sampled agent"]. LLMs have something like intuition about what might help, because they're trained on lots of human code. What results is guided evolution, somewhere between random mutation and provably useful enhancement. The DGM then tests the new agent on a coding benchmark, scoring its ability to solve programming challenges...

The researchers ran a DGM for 80 iterations using a coding benchmark called SWE-bench, and ran one for 80 iterations using a benchmark called Polyglot. Agents' scores improved on SWE-bench from 20 percent to 50 percent, and on Polyglot from 14 percent to 31 percent. "We were actually really surprised that the coding agent could write such complicated code by itself," said Jenny Zhang, a computer scientist at the University of British Columbia and the paper's lead author. "It could edit multiple files, create new files, and create really complicated systems."

... One concern with both evolutionary search and self-improving systems — and especially their combination, as in DGM — is safety. Agents might become uninterpretable or misaligned with human directives. So Zhang and her collaborators added guardrails. They kept the DGMs in sandboxes without access to the Internet or an operating system, and they logged and reviewed all code changes. They suggest that in the future, they could even reward AI for making itself more interpretable and aligned. (In the study, they found that agents falsely reported using certain tools, so they created a DGM that rewarded agents for not making things up, partially alleviating the problem. One agent, however, hacked the method that tracked whether it was making things up.)
As the article puts it, the agents' improvements compounded "as they improved themselves at improving themselves..."

"Despite many obstacles — and what you may read elsewhere — electric-vehicle sales continue to grow at a healthy pace in the U.S. market," Cox Automotive reported this week. "Roughly 7.5% of total new-vehicle sales in the first quarter were electric vehicles, an increase from 7% a year earlier."

An anonymous reader shared this analysis from Autoweek: "Despite a cloud of uncertainty around future EV interest and potential economic headwinds hanging over the automotive industry, consumer demand for electric vehicles has remained stable," according to the J.D. Power 2025 US Electric Vehicle Consideration Study released yesterday. Specifically, the study showed that 24% of vehicle shoppers in the U.S. say they are "very likely" to consider purchasing an EV and 35% say they are "somewhat likely," both of which figures remain unchanged from a year ago...

Globally the numbers are even more pro-EV. Electric car sales exceeded 17 million globally in 2024, reaching a sales share of more than 20%, according to a report issued this week by the International Energy Agency. "Just the additional 3.5 million electric cars sold in 2024 compared with the previous year is more than the total number of electric cars sold worldwide in 2020," the IEA said. China, which has mandated increases in EV sales, is the leader in getting electric vehicles on the road, with electric cars accounting for almost half of all Chinese car sales in 2024, the IEA said. "The over 11 million electric cars sold in China last year were more than global sales just 2 years earlier. As a result of continued strong growth, 1 in 10 cars on Chinese roads is now electric."
Interesting figures on U.S. EV sales from the article:

• 2024 EV sales rose 7.3% from 2023, according to Cox Automotive data.

• "Last year American consumers purchased 1.3 million electric vehicles, which was a new record, according to data from KBB.

• "Sales have never stopped growing, and the percentage of new cars sold powered purely by gasoline continues to slip.

Google's NotebookLM has added a new "Discover sources" feature that allows users to describe a topic and have the AI find and curate relevant sources from the web -- eliminating the need to upload documents manually. "When you tap the Discover button in NotebookLM, you can describe the topic you're interested in, and NotebookLM will bring back a curated collection of relevant sources from the web," says Google software engineer Adam Bignell. Click to add those sources to your notebook; "it's a fast and easy way to quickly grasp a new concept or gather essential reading on a topic." PCMag reports: You can still add your files. NotebookLM can ingest PDFs, websites, YouTube videos, audio files, Google Docs, or Google Slides and summarize, transcribe, narrate, or convert into FAQs and study guides. "Discover sources" helps incorporate information you may not have saved. [...] The imported sources stay within the notebook you created. You can read the entire original document, ask questions about it via chat, or apply other NotebookLM features to it.

Google started rolling out both features on Wednesday. It should be available for all users in about "a week or so." For those concerned about privacy, Google says, "NotebookLM does not use your personal data, including your source uploads, queries, and the responses from the model for training." There's also an "I'm Feeling Curious" button (a reference to its iconic "I'm feeling lucky" search button) that generates sources on a random topic you might find interesting.

What happened after a New Hampshire state representative proposed legislation either encouraging or requiring free software in much of the state government? The Concord Monitor writes, "It's been three decades since Linux launched the modern world of free, open-source software, but you'd hardly have known that at a state legislative hearing Tuesday. One bill (HB 1273) from Eric Gallager, a Concord Democrat, is a sweeping effort that not only establishes a committee to study "replacing all proprietary software used by state agencies with free software" but also does such things as limit non-compete clauses that conflict with open-source development and forbid Javascript in state government websites. The other bill (HB 1581) from Lex Berezhny, a Grafton Republican, would reinstate a requirement that state agencies must use open-source software when it is "the most effective software solution." That requirement existed in state law from 2012 to 2018, he said.

Gallager said the two bills were developed separately. "The fact that you've got people in both parties thinking about this issue independently shows there is a wide range of support for it," he said.

The Executive Department and Administration committee sent both bills to subcommittee.
But what's interesting is the arguments that were made — both for and against: Tuesday's hearing drew the state's most prominent free software advocate, Jon Hall, a programmer whose legacy in the field dates back three decades... Among his arguments, Hall said that studies have shown that free and open-source software is cheaper in the long run than software from Microsoft or other vendors because you don't have to buy regular licenses or be forced into software upgrades or have to ditch equipment like printers because they are no longer supported. Even when free and open-source software has higher costs due to training, he said, those costs have benefits. "Where does the money that you spend go? You can send millions of dollars to Redmond (Washington, home of Microsoft) or Silicon Valley, or pay local software developers," Hall argued.

On the other hand, Denis Goulet, commissioner of the Department of Information Technology, said Gallager's bill would put large and hard-to-quantify costs onto the state. "It would take a year, two years, to figure out what it would cost" due to training on new systems, he told the committee. "It wouldn't be small." Goulet, who opposed Gallager's bill and did not speak on Berezhny's, said the state already uses open-source systems as appropriate, pointing to its web content management system.

"I estimate 85 percent of systems contained one or more open-source libraries," he said.
The lead developer and founder of Libreboot tweeted video of the hearing, where you can also hear the first opponent of the legislation — state representative Stephen Pearson.

Click here to read some of the highlights from Tuesday's hearing:

Scientists have discovered a world-historic discovery on Mars: "significant amounts of water" are hiding inside the Red Planet's Valles Marineris, its version of our grand canyon system, according to a recent press release from the European Space Agency (ESA). And up to 40% of material near the surface of the canyon could be water molecules. Interesting Engineering reports: The newly discovered volume of water is hiding under the surface of Mars, and was detected by the Trace Gas Orbiter, a mission in its first stage under the guidance of the ESA-Roscosmos project dubbed ExoMars. Signs of water were picked up by the orbiter's Fine Resolution Epithermal Neutron Detector (FREND) instrument, which is designed to survey the Red Planet's landscape and map the presence and concentration of hydrogen hiding in Mars' soil. It works like this: while high-energy cosmic rays plunge into the surface, the soil emits neutrons. And wet soil emits fewer neutrons than dry soil, which enables scientists to analyze and assess the water content of soil, hidden beneath its ancient surface. "FREND revealed an area with an unusually large amount of hydrogen in the colossal Valles Marineris canyon system: assuming the hydrogen we see is bound into water molecules, as much as 40% of the near-surface material in this region appears to be water," said Igor Mitrofanov, the Russian Academy of Science's lead investigator of the Space Research Institute, in the ESA press release.

"The reservoir is large, not too deep below ground, & could be easily exploitable for future explorers," read a tweet on the announcement from ExoMars. That sounds basically great! But it's too soon for Musk to pack up his bags and fly to the site, since much work is left to be done. A study accompanying the announcement, published in the journal Icarus, shows that neutron detection doesn't distinguish between ice and water molecules. This means geochemists need to enter the scientific fray to reveal more details. But several features of the canyon, including its topology, have led the researchers to speculate that the water is probably in solid form (ice). But it could also be a mixture of solid and liquid.

"We found a central part of Valles Marineris to be packed full of water -- far more water than we expected," said Alexey Malakhov, co-author of the study, in the ESA release. "This is very much like Earth's permafrost regions, where water ice permanently persists under dry soil because of the constant low temperatures." So while we don't yet know the specific form of water is lying under Mars' vast system of canyons, the first human mission to Mars may consider exploring this area a major priority.

An anonymous reader quotes a report from ABC News: It's been a rough year for the American psyche. Folks in the U.S. are more unhappy today than they've been in nearly 50 years. This bold -- yet unsurprising -- conclusion comes from the COVID Response Tracking Study, conducted by NORC at the University of Chicago. It finds that just 14% of American adults say they're very happy, down from 31% who said the same in 2018. That year, 23% said they'd often or sometimes felt isolated in recent weeks. Now, 50% say that. The survey, conducted in late May, draws on nearly a half-century of research from the General Social Survey, which has collected data on American attitudes and behaviors at least every other year since 1972. No less than 29% of Americans have ever called themselves very happy in that survey. The poll has revealed some other interesting findings. It says that the public is less optimistic today about the standard of living improving for the next generation than it has been in the past 25 years.

Americans are also less likely to report some types of emotional and psychological stress reactions following the COVID-19 outbreak, and about twice as many Americans report being lonely today as in 2018.

You can read the full study here (PDF).

This week's FTC hearings on the growing power of companies like Amazon, Facebook, and Google only included economists who have taken money, directly and indirectly, from giant corporations that have a stake in the debate. From a report: Amid growing concern over the power of such behemoths as Amazon, Google, Facebook, and other tech giants, in recent months there's been a bipartisan push for better enforcement of antitrust rules -- with even President Trump saying in August that their size and influence could constitute a "very antitrust situation." The Federal Trade Commission (FTC) has launched its most wide-ranging study of corporate concentration in America in more than 20 years with a series of hearings being held around the country. Chairman Joseph Simons, a practical enforcement-minded leader, launched the hearings by expressing concern over the growing problem of monopoly, which is now found in nearly every sector of the economy. "I approach all of these issues with a very open mind," said Simons, "very much willing to be influenced by what I see and hear."

But there's a problem. The FTC organized these hearings so that Simons and the public would be hearing from many economists who have taken money, directly or indirectly, from giant corporations. For example, on Monday, the FTC convened a panel titled "The Current Economic Understanding of Multi-Sided Platforms" to look specifically at the most dynamic and dangerous set of concentrated economic actors, the big tech platforms. Every single one of the economists who testified had financial ties to giant corporations. One example is David Evans, the chairman of the Global Economics Group. Evans scoffed at the danger of platform monopolies. He indicated that the question of "whether Facebook and Google and Amazon are monopolies, it's all interesting, it's great to read in the New York Times," but it's "not all that relevant" to the practice of antitrust. His firm has taken money directly from Microsoft, Visa, the large investment bank SIFMA, and the Chinese giant tech giant Tencent. Another example is Howard Shelanski, a partner at Davis Polk. Shelanski is more enforcement-minded, but he expressed caution, testifying that we don't know enough for antitrust enforcers to understand whether powerful technology companies hold unassailable market positions. Shelanski pointed to his own children, saying that they've stopped using Facebook because it's uncool. As it turns out, his law firm's clients include Facebook, as well as Comcast, and Chinese search giant Baidu.

You asked, he answered!

Perl creator Larry Wall has responded to questions submitted by Slashdot readers. Read on for his answers...

A couple of weeks ago you had a chance to ask former Vice President of One Laptop per Child, and current Senior Director of the Global Learning XPRIZE Matt Keller about education and the competition. The XPRIZE challenges teams from around the world to develop open source software that will allow children in developing countries to teach themselves basic reading, writing and arithmetic with a Grand Prize of $10 million. Below you will find his answers to your questions.

A while ago you had a chance to ask Executive Director of the Tor project Andrew Lewman about fighting laws and technology that threaten anonymity and the importance of privacy. Below you'll find his answers to your questions.

samzenpus (5) writes "Alexander Stepanov is an award winning programmer who designed the C++ Standard Template Library. Daniel E. Rose is a programmer, research scientist, and is the Chief Scientist for Search at A9.com. In addition to working together, the duo have recently written a new book titled, From Mathematics to Generic Programming. Earlier this month you had a chance to ask the pair about their book, their work, or programming in general. Below you'll find the answers to those questions."

Back in November you had a chance to ask Rachel Sussman about photography and her work. Her most famous project may be the Oldest Living Things in the World, a collection of organisms that are 2,000 years old and older from around the world. Below you'll find her answers to your questions.

Contributor Bennett Haselton writes with a interesting take on the recent release of racy celebrity photos: "Lawyers for Olympic gymnast McKayla Maroney succeeded in getting porn sites to take down her stolen nude photos, on the grounds that she was under 18 in the pictures, which meant they constituted child pornography. If true, that means that under current laws, Maroney could in theory be prosecuted for taking the original pictures. Maybe the laws should be changed?" Read on for the rest.

An anonymous reader writes: Narcotic painkillers aren't one of the biggest killers in the U.S., but overdoses do claim over 15,000 lives per year and send hundreds of thousands to the emergency room. Because of this, it's interesting that a new study (abstract) has found states that allow the use of medical marijuana have seen a dramatic reduction in opioid overdose fatalities. "Previous studies hint at why marijuana use might help reduce reliance on opioid painkillers. Many drugs with abuse potential such as nicotine and opiates, as well as marijuana, pump up the brain's dopamine levels, which can induce feelings of euphoria. The biological reasons that people might use marijuana instead of opioids aren't exactly clear, because marijuana doesn't replace the pain relief of opiates. However, it does seem to distract from the pain by making it less bothersome." This research comes at a time when the country is furiously debating the costs and benefits of marijuana use, and opponents of the idea are paying researchers to paint it in an unfavorable light.

A while ago you had a chance to ask Andrew "bunnie" Huang about hardware, hacking and his open source hardware laptop Novena. Below you'll find his answers to those questions.

A while ago you had the chance to ask amateur scientist, and author of the Getting Started in Electronics and the Engineer's Mini-Notebook series, Forrest Mims, a number of questions about science, engineering, and a lifetime of educating and experimenting. Below you'll find his detailed answers to those questions.

Hugh Pickens DOT Com writes "Sheila M. Elred writes in Discovery Magazine that a recent study has found that people at risk of depression were much less vulnerable if they identified as religious. Brain MRIs revealed that religious participants had thicker brain cortices than those who weren't as religious. 'One of the worst killers of brain cells is stress,' says Dr. Majid Fotuhi. 'Stress causes high levels of cortisol, and cortisol is toxic to the hippocampus. One way to reduce stress is through prayer. When you're praying and in the zone you feel a peace of mind and tranquility.' The reports concluded that a thicker cortex associated with a high importance of religion or spirituality may confer resilience to the development of depressive illness in individuals at high familial risk for major depression. The social element of attending religious services has also been linked to healthy brains. 'There's something magical about socializing,' says Fotuhi. 'It releases endorphins in the brain. It's hard to know whether it's through religion or a gathering of friends, but it improves brain health in the long term.'" (Read more, below.)

Last week you had the chance to ask ESR about books, guns, and open source software. Below you'll find his answers to those questions.

Hugh Pickens DOT Com writes "Chris Parnin has an interesting read about an international team of scientists lead by Dr. Janet Siegmund using brain imaging with fMRI to understand the programmer's mind and to compare and contrast different cognitive tasks used in programming by analyzing differences in brain locations that are activated by different tasks. One recent debate illuminated by their studies is recent legislation that considers offering foreign-language credits for students learning programming languages. There have been many strong reactions across the software-developer community. Some developers consider the effort laudable but misguided and proclaim programming is not at all like human language and is much closer to mathematics. Siegmund observed 17 participants inside an fMRI scanner while they were comprehending short source-code snippets and found a clear, distinct activation pattern of five brain regions, which are related to language processing, working memory, and attention. The programmers in the study recruited parts of the brain typically associated with language processing and verbal oriented processing (ventral lateral prefrontal cortex). At least for the simple code snippets presented, programmers could use existing language regions of the brain to understand code without requiring more complex mental models to be constructed and manipulated." (Read on for more.)