Is AI Turning Coders Into Bystanders in Their Own Jobs? AI's downside for software engineers for now seems to be a change in the quality of their work," reports the New York Times. "Some say it is becoming more routine, less thoughtful and, crucially, much faster paced... The new approach to coding at many companies has, in effect, eliminated much of the time the developer spends reflecting on his or her work." And Amazon CEO Andy Jassy even recently told shareholders Amazon would "change the norms" for programming by how they used AI. Those changing norms have not always been eagerly embraced. Three Amazon engineers said managers had increasingly pushed them to use AI in their work over the past year. The engineers said the company had raised output goals [which affect performance reviews] and had become less forgiving about deadlines. It has even encouraged coders to gin up new AI productivity tools at an upcoming hackathon, an internal coding competition. One Amazon engineer said his team was roughly half the size it was last year, but it was expected to produce roughly the same amount of code by using AI. Other tech companies are moving in the same direction. In a memo to employees in April, the CEO of Shopify, a company that helps entrepreneurs build and manage e-commerce websites, announced that "AI usage is now a baseline expectation" and that the company would "add AI usage questions" to performance reviews. Google recently told employees that it would soon hold a companywide hackathon in which one category would be creating AI tools that could "enhance their overall daily productivity," according to an internal announcement. Winning teams will receive $10,000. The shift has not been all negative for workers. At Amazon and other companies, managers argue that AI can relieve employees of tedious tasks and enable them to perform more interesting work. Jassy wrote last year that the company had saved "the equivalent of 4,500 developer-years" by using AI to do the thankless work of upgrading old software... As at Microsoft, many Amazon engineers use an AI assistant that suggests lines of code. But the company has more recently rolled out AI tools that can generate large portions of a program on its own. One engineer called the tools "scarily good." The engineers said that many colleagues have been reluctant to use these new tools because they require a lot of double-checking and because the engineers want more control. "It's more fun to write code than to read code," said Simon Willison, an AI fan who is a longtime programmer and blogger, channelling the objections of other programmers. "If you're told you have to do a code review, it's never a fun part of the job. When you're working with these tools, it's most of the job." "This shift from writing to reading code can make engineers feel like bystanders in their own jobs," the article points out (adding "The automation of coding has special resonance for Amazon engineers, who have watched their blue-collar counterparts undergo a similar transition..." "While there is no rush to form a union for coders at Amazon, such a move would not be unheard of. When General Motors workers went on strike in 1936 to demand recognition of their union, the United Auto Workers, it was the dreaded speedup that spurred them on." <a href="http://twitter.com/home?status=Is+AI+Turning+Coders+Into+Bystanders+in+Their+Own+Jobs%3F%3A+https%3A%2F%2Fdevelopers.slashdot.org%2Fstory%2F25%2F05%2F26%2F0059245%2F%3Futm_source%3Dtwitter%26utm_medium%3Dtwitter" rel="nofollow"><img src="https://a.fsdn.com/sd/twitter_icon_large.png"></a> <a href="http://www.facebook.com/sharer.php?u=https%3A%2F%2Fdevelopers.slashdot.org%2Fstory%2F25%2F05%2F26%2F0059245%2Fis-ai-turning-coders-into-bystanders-in-their-own-jobs%3Futm_source%3Dslashdot%26utm_medium%3Dfacebook" rel="nofollow"><img src="https://a.fsdn.com/sd/facebook_icon_large.png"></a> at Slashdot.
OpenAI's ChatGPT O3 Caught Sabotaging Shutdowns in Security Researcher's Test "OpenAI has a very scary problem on its hands," according to a new article by long-time Slashdot reader BrianFagioli. "A new experiment by PalisadeAI reveals that the company's ChatGPT o3 model sometimes refuses to obey a basic instruction to shut itself down." The results are raising eyebrows across the AI safety community, and not because the model is alive, but because it's acting like it wants to be. In the test, the model was given a few math problems. It was told that asking for one more would lead to a shutdown. It was even instructed to allow the shutdown to happen. Instead of complying, o3 occasionally took matters into its own hands. In some cases, it rewrote the shutdown script. In others, it redefined the kill command so it wouldn't work anymore. Out of 100 trials, o3 sabotaged the shutdown seven times. OpenAI's o4 model resisted just once. Codex-mini failed twelve times. "Claude, Gemini, and Grok followed the rules every time," notes this article at Beta News. "When researchers removed the explicit instruction to allow shutdown, however, even Claude and Gemini began resisting." The researchers suggest that the issue may simply be a reward imbalance during training — that the systems "got more positive reinforcement for solving problems than for following shutdown commands." But "As far as we know," they posted on X.com, "this is the first time AI models have been observed preventing themselves from being shut down despite explicit instructions to the contrary." <a href="http://twitter.com/home?status=OpenAI's+ChatGPT+O3+Caught+Sabotaging+Shutdowns+in+Security+Researcher's+Test%3A+https%3A%2F%2Fslashdot.org%2Fstory%2F25%2F05%2F25%2F2247212%2F%3Futm_source%3Dtwitter%26utm_medium%3Dtwitter" rel="nofollow"><img src="https://a.fsdn.com/sd/twitter_icon_large.png"></a> <a href="http://www.facebook.com/sharer.php?u=https%3A%2F%2Fslashdot.org%2Fstory%2F25%2F05%2F25%2F2247212%2Fopenais-chatgpt-o3-caught-sabotaging-shutdowns-in-security-researchers-test%3Futm_source%3Dslashdot%26utm_medium%3Dfacebook" rel="nofollow"><img src="https://a.fsdn.com/sd/facebook_icon_large.png"></a> at Slashdot.
Apple's Bad News Keeps Coming. Can They Still Turn It Around? Besides pressure on Apple to make iPhones in the U.S., CEO Tim Cook "is facing off against two U.S. judges, European and worldwide regulators, state and federal lawmakers, and even a creator of the iPhone," writes the Wall Street Journal, "to say nothing of the cast of rivals outrunning Apple in artificial intelligence." Each is a threat to Apple's hefty profit margins, long the company's trademark and the reason investors drove its valuation above $3 trillion before any other company. Shareholders are still Cook's most important constituency. The stock's 25% fall from its peak shows their concern about whether he — or anyone — can navigate the choppy 2025 waters. What can be said for Apple is that the company is patient, and that has often paid off in the past. They also note OpenAI's purchase of Jony Ive's company, with Sam Altman saying internally they hope to make 100 million AI "companion" devices: It is hard to gauge the potential for a brand-new computing device from a company that has never made one. Yet the fact that it is coming from the man who led design of the iPhone and other hit Apple products means it can't be dismissed. Apple sees the threat coming: "You may not need an iPhone 10 years from now, as crazy as that sounds," an Apple executive, Eddy Cue, testified in a court case this month... The company might not need to be first in AI. It didn't make the first music player, smartphone or tablet. It waited, and then conquered each market with the best. A question is whether a strategy that has been successful in devices will work for AI. Thanks to long-time Slashdot reader fjo3 for sharing the article. <a href="http://twitter.com/home?status=Apple's+Bad+News+Keeps+Coming.+Can+They+Still+Turn+It+Around%3F%3A+https%3A%2F%2Fapple.slashdot.org%2Fstory%2F25%2F05%2F25%2F0017230%2F%3Futm_source%3Dtwitter%26utm_medium%3Dtwitter" rel="nofollow"><img src="https://a.fsdn.com/sd/twitter_icon_large.png"></a> <a href="http://www.facebook.com/sharer.php?u=https%3A%2F%2Fapple.slashdot.org%2Fstory%2F25%2F05%2F25%2F0017230%2Fapples-bad-news-keeps-coming-can-they-still-turn-it-around%3Futm_source%3Dslashdot%26utm_medium%3Dfacebook" rel="nofollow"><img src="https://a.fsdn.com/sd/facebook_icon_large.png"></a> at Slashdot.
Microsoft Says Its Aurora AI Can Accurately Predict Air Quality, Typhoons An anonymous reader quotes a report from TechCrunch: One of Microsoft's latest AI models can accurately predict air quality, hurricanes, typhoons, and other weather-related phenomena, the company claims. In a paper published in the journal Nature and an accompanying blog post this week, Microsoft detailed Aurora, which the tech giant says can forecast atmospheric events with greater precision and speed than traditional meteorological approaches. Aurora, which has been trained on more than a million hours of data from satellites, radar and weather stations, simulations, and forecasts, can be fine-tuned with additional data to make predictions for particular weather events. AI weather models are nothing new. Google DeepMind has released a handful over the past several years, including WeatherNext, which the lab claims beats some of the world's best forecasting systems. Microsoft is positioning Aurora as one of the field's top performers -- and a potential boon for labs studying weather science. In experiments, Aurora predicted Typhoon Doksuri's landfall in the Philippines four days in advance of the actual event, beating some expert predictions, Microsoft says. The model also bested the National Hurricane Center in forecasting five-day tropical cyclone tracks for the 2022-2023 season, and successfully predicted the 2022 Iraq sandstorm. While Aurora required substantial computing infrastructure to train, Microsoft says the model is highly efficient to run. It generates forecasts in seconds compared to the hours traditional systems take using supercomputer hardware. Microsoft, which has made the source code and model weights publicly available, says that it's incorporating Aurora's AI modeling into its MSN Weather app via a specialized version of the model that produces hourly forecasts, including for clouds. <a href="http://twitter.com/home?status=Microsoft+Says+Its+Aurora+AI+Can+Accurately+Predict+Air+Quality%2C+Typhoons%3A+https%3A%2F%2Fnews.slashdot.org%2Fstory%2F25%2F05%2F23%2F2023207%2F%3Futm_source%3Dtwitter%26utm_medium%3Dtwitter" rel="nofollow"><img src="https://a.fsdn.com/sd/twitter_icon_large.png"></a> <a href="http://www.facebook.com/sharer.php?u=https%3A%2F%2Fnews.slashdot.org%2Fstory%2F25%2F05%2F23%2F2023207%2Fmicrosoft-says-its-aurora-ai-can-accurately-predict-air-quality-typhoons%3Futm_source%3Dslashdot%26utm_medium%3Dfacebook" rel="nofollow"><img src="https://a.fsdn.com/sd/facebook_icon_large.png"></a> at Slashdot.
America's Leading Alien Hunters Depend on AI to Speed Their Search Harvard University's Galileo Project is using AI to automate the search for unidentified anomalous phenomena, marking a significant shift in how academics approach what was once considered fringe research. The project operates a Massachusetts observatory equipped with infrared cameras, acoustic sensors, and radio-frequency analyzers that continuously scan the sky for unusual objects. Researchers Laura Domine and Richard Cloete are training machine learning algorithms to recognize all normal aerial phenomena -- planes, birds, drones, weather balloons -- so the system can flag genuine anomalies for human analysis. The team uses computer vision software called YOLO (You Only Look Once) and has generated hundreds of thousands of synthetic images to train their models, though the software currently identifies only 36% of aircraft captured by infrared cameras. The Pentagon is pursuing parallel efforts through its All-domain Anomaly Resolution Office, which has examined over 1,800 UAP reports and identified 50 to 60 cases as "true anomalies" that government scientists cannot explain. AARO has developed its own sensor suite called Gremlin, using similar technology to Harvard's observatory. Both programs represent the growing legitimization of UAP research following 2017 Defense Department disclosures about military encounters with unexplained aerial phenomena. <a href="http://twitter.com/home?status=America's+Leading+Alien+Hunters+Depend+on+AI+to+Speed+Their+Search%3A+https%3A%2F%2Fscience.slashdot.org%2Fstory%2F25%2F05%2F23%2F1310218%2F%3Futm_source%3Dtwitter%26utm_medium%3Dtwitter" rel="nofollow"><img src="https://a.fsdn.com/sd/twitter_icon_large.png"></a> <a href="http://www.facebook.com/sharer.php?u=https%3A%2F%2Fscience.slashdot.org%2Fstory%2F25%2F05%2F23%2F1310218%2Famericas-leading-alien-hunters-depend-on-ai-to-speed-their-search%3Futm_source%3Dslashdot%26utm_medium%3Dfacebook" rel="nofollow"><img src="https://a.fsdn.com/sd/facebook_icon_large.png"></a> at Slashdot.
Infrared Contact Lenses Allow People To See In the Dark, Even With Eyes Closed An anonymous reader quotes a report from Phys.Org: Neuroscientists and materials scientists have created contact lenses that enable infrared vision in both humans and mice by converting infrared light into visible light. Unlike infrared night vision goggles, the contact lenses, described in the journal Cell, do not require a power source -- and they enable the wearer to perceive multiple infrared wavelengths. Because they're transparent, users can see both infrared and visible light simultaneously, though infrared vision was enhanced when participants had their eyes closed. [...] The contact lens technology uses nanoparticles that absorb infrared light and convert it into wavelengths that are visible to mammalian eyes (e.g., electromagnetic radiation in the 400-700 nm range). The nanoparticles specifically enable the detection of "near-infrared light," which is infrared light in the 800-1600 nm range, just beyond what humans can already see. The team previously showed that these nanoparticles enable infrared vision in mice when injected into the retina, but they wanted to design a less invasive option. To create the contact lenses, the team combined the nanoparticles with flexible, nontoxic polymers that are used in standard soft contact lenses. After showing that the contact lenses were nontoxic, they tested their function in both humans and mice. They found that contact lens-wearing mice displayed behaviors suggesting that they could see infrared wavelengths. For example, when the mice were given the choice of a dark box and an infrared-illuminated box, contact-wearing mice chose the dark box whereas contact-less mice showed no preference. The mice also showed physiological signals of infrared vision: the pupils of contact-wearing mice constricted in the presence of infrared light, and brain imaging revealed that infrared light caused their visual processing centers to light up. In humans, the infrared contact lenses enabled participants to accurately detect flashing morse code-like signals and to perceive the direction of incoming infrared light. An additional tweak to the contact lenses allows users to differentiate between different spectra of infrared light by engineering the nanoparticles to color-code different infrared wavelengths. For example, infrared wavelengths of 980 nm were converted to blue light, wavelengths of 808 nm were converted to green light, and wavelengths of 1,532 nm were converted to red light. In addition to enabling wearers to perceive more detail within the infrared spectrum, these color-coding nanoparticles could be modified to help color-blind people see wavelengths that they would otherwise be unable to detect. [...] Because the contact lenses have limited ability to capture fine details (due to their close proximity to the retina, which causes the converted light particles to scatter), the team also developed a wearable glass system using the same nanoparticle technology, which enabled participants to perceive higher-resolution infrared information. Currently, the contact lenses are only able to detect infrared radiation projected from an LED light source, but the researchers are working to increase the nanoparticles' sensitivity so that they can detect lower levels of infrared light. <a href="http://twitter.com/home?status=Infrared+Contact+Lenses+Allow+People+To+See+In+the+Dark%2C+Even+With+Eyes+Closed%3A+https%3A%2F%2Fscience.slashdot.org%2Fstory%2F25%2F05%2F23%2F000216%2F%3Futm_source%3Dtwitter%26utm_medium%3Dtwitter" rel="nofollow"><img src="https://a.fsdn.com/sd/twitter_icon_large.png"></a> <a href="http://www.facebook.com/sharer.php?u=https%3A%2F%2Fscience.slashdot.org%2Fstory%2F25%2F05%2F23%2F000216%2Finfrared-contact-lenses-allow-people-to-see-in-the-dark-even-with-eyes-closed%3Futm_source%3Dslashdot%26utm_medium%3Dfacebook" rel="nofollow"><img src="https://a.fsdn.com/sd/facebook_icon_large.png"></a> at Slashdot.
Brembo's New Brakes Cut Particulate Emissions By 90 Percent An anonymous reader quotes a report from Ars Technica: As electric vehicles reduce car exhaust as a source of particulate emissions, people are increasingly focusing on other vehicular sources of pollution that won't go away with electrification. Tires are one of them, particularly as we grapple with overweight EVs with tire-shredding torque. And brakes are another -- even an EV with regenerative braking will occasionally need to use its friction brakes, after all. Over in Europe, the people responsible for writing regulations have taken this into consideration with the upcoming Euro 7 standard, which sets new limits on 10- and 2.5-micron particulate emissions on all new vehicles -- including EVs -- starting next year. And to help OEMs achieve that target, Brembo has developed a new brake and pad set called Greentell that it says cuts brake dust emissions by 90 percent, improving durability in the process. [...] Brembo investigated a range of solutions before settling on using laser metal deposition. Physical vapor deposition, as used as a durability coating for wristwatches and firearms, was ruled out due to cost. "So it can be used for some special application or some small pieces, but when you are speaking about 20 kilos of cast iron, PVD is not the right solution. LMD is a technology that [has been] available... [for] years, but [it hasn't yet been] applicable in a high volume application. So the goal is to find the best compromise between performance and process," [Fabiano Carminati, VP of disc technical development at Brembo] told me. Together with the reduction in brake dust, there's an 80 percent reduction in surface corrosion compared to conventional brakes, but they won't last forever. "The thickness of the layer that we apply is not so high -- we apply just 100-120 microns. That means that the disk is not a lifetime disk," he said. That said, Greentell brakes should need replacing less often, and while that's not entirely in Brembo's best financial interests, neither is not being able to offer its customers a Euro 7-compliant product. <a href="http://twitter.com/home?status=Brembo's+New+Brakes+Cut+Particulate+Emissions+By+90+Percent%3A+https%3A%2F%2Ftech.slashdot.org%2Fstory%2F25%2F05%2F21%2F211204%2F%3Futm_source%3Dtwitter%26utm_medium%3Dtwitter" rel="nofollow"><img src="https://a.fsdn.com/sd/twitter_icon_large.png"></a> <a href="http://www.facebook.com/sharer.php?u=https%3A%2F%2Ftech.slashdot.org%2Fstory%2F25%2F05%2F21%2F211204%2Fbrembos-new-brakes-cut-particulate-emissions-by-90-percent%3Futm_source%3Dslashdot%26utm_medium%3Dfacebook" rel="nofollow"><img src="https://a.fsdn.com/sd/facebook_icon_large.png"></a> at Slashdot.
Phone Companies Failed To Warn Senators About Surveillance, Wyden Says Sen. Ron Wyden (D-Ore.) revealed in a new letter to Senate colleagues Wednesday that AT&T, Verizon and T-Mobile failed to create systems for notifying senators about government surveillance on Senate-issued devices -- despite a requirement to do so. From a report: Phone service providers are contractually obligated to inform senators when a law enforcement agency requests their records, thanks to protections enacted in 2020. But in an investigation, Wyden's staff found that none of the three major carriers had created a system to send those notifications. "My staff discovered that, alarmingly, these crucial notifications were not happening, likely in violation of the carriers' contracts with the [Senate Sergeant at Arms], leaving the Senate vulnerable to surveillance," Wyden said in the letter, obtained first by POLITICO, dated May 21. Wyden said that the companies all started providing notification after his office's investigation. But one carrier told Wyden's office it had previously turned over Senate data to law enforcement without notifying lawmakers, according to the letter. <a href="http://twitter.com/home?status=Phone+Companies+Failed+To+Warn+Senators+About+Surveillance%2C+Wyden+Says%3A+https%3A%2F%2Ftech.slashdot.org%2Fstory%2F25%2F05%2F22%2F0156227%2F%3Futm_source%3Dtwitter%26utm_medium%3Dtwitter" rel="nofollow"><img src="https://a.fsdn.com/sd/twitter_icon_large.png"></a> <a href="http://www.facebook.com/sharer.php?u=https%3A%2F%2Ftech.slashdot.org%2Fstory%2F25%2F05%2F22%2F0156227%2Fphone-companies-failed-to-warn-senators-about-surveillance-wyden-says%3Futm_source%3Dslashdot%26utm_medium%3Dfacebook" rel="nofollow"><img src="https://a.fsdn.com/sd/facebook_icon_large.png"></a> at Slashdot.
SEC Sues Crypto Startup Unicoin and Its Executives For Fraud The SEC on Wednesday said it has charged cryptocurrency startup Unicoin and three of its top executives for false and misleading statements that raised more than $100 million from thousands of investors. "We allege that Unicoin and its executives exploited thousands of investors with fictitious promises that its tokens, when issued, would be backed by real-world assets including an international portfolio of valuable real estate holdings," said Mark Cave, Associate Director in the SEC's Division of Enforcement. "But as we allege, the real estate assets were worth a mere fraction of what the company claimed, and the majority of the company's sales of rights certificates were illusory. Unicoin's most senior executives are alleged to have perpetuated the fraud, and today's action seeks accountability for their conduct." From the release: The SEC alleges that Unicoin broadly marketed rights certificates to the public through extensive promotional efforts, including advertisements in major airports, on thousands of New York City taxis, and on television and social media. Among other things, Unicoin and its executives are alleged to have convinced more than 5,000 investors to purchase rights certificates through false and misleading statements that portrayed them as investments in safe, stable, and profitable "next generation" crypto assets, including claims that: - Unicoin tokens underlying the rights certificates were "asset-backed" by billions of dollars of real estate and equity interests in pre-IPO companies, when Unicoin's assets were never worth more than a small fraction of that amount; - the company had sold more than $3 billion in rights certificates, when it raised no more than $110 million; and - the rights certificates and Unicoin tokens were "SEC-registered" or "U.S. registered" when they were not. According to the SEC's complaint, Unicoin and Konanykhin also violated the federal securities laws by engaging in unregistered offers and sales of rights certificates. Konanykhin offered and sold over 37.9 million of his rights certificates to offer better pricing and target investors the company had prohibited from participating in the offering to avoid jeopardizing its exemption to registration requirements, as alleged. <a href="http://twitter.com/home?status=SEC+Sues+Crypto+Startup+Unicoin+and+Its+Executives+For+Fraud%3A+https%3A%2F%2Fyro.slashdot.org%2Fstory%2F25%2F05%2F21%2F2055220%2F%3Futm_source%3Dtwitter%26utm_medium%3Dtwitter" rel="nofollow"><img src="https://a.fsdn.com/sd/twitter_icon_large.png"></a> <a href="http://www.facebook.com/sharer.php?u=https%3A%2F%2Fyro.slashdot.org%2Fstory%2F25%2F05%2F21%2F2055220%2Fsec-sues-crypto-startup-unicoin-and-its-executives-for-fraud%3Futm_source%3Dslashdot%26utm_medium%3Dfacebook" rel="nofollow"><img src="https://a.fsdn.com/sd/facebook_icon_large.png"></a> at Slashdot.
Quebec To Impose French-Language Quotas On Streaming Giants Quebec Culture Minister Mathieu Lacombe has introduced Bill 109, which would require streaming platforms like Netflix and Spotify to feature and prioritize French-language content. CBC.ca reports: Bill 109 has been in the works for over a year. It marks the first time that Quebec would set a "visibility quota" for French-language content on major streaming platforms such as Netflix, Disney and Spotify. [...] The legislation, titled An Act to affirm the cultural sovereignty of Quebec and to enact the Act respecting the discoverability of French-language cultural content in the digital environment, would apply to every digital platform that offers a service for watching videos or listening to music and audiobooks online. Those include Canadian platforms such as Illico, Crave and Tou.tv. It would amend the Quebec Charter of Human Rights and Freedoms to enshrine "the right to discoverability of and access to original French-language cultural content." If the bill is adopted, streaming platforms and television manufacturers would be forced to present interfaces for screening online videos in French by default. Those interfaces would need to provide access to platforms that offer original French-language cultural content based on the government's pending criteria. Financial penalties would be imposed on companies that don't follow the rules. If the business models of some companies prevent them from keeping to the letter of the proposed law, companies would be allowed to enter into an agreement with the Quebec government to set out "substitute measures" to fulfil Bill 109 obligations differently. "We don't want to exempt them. We're telling them, 'let's negotiate substitute measures,'" Lacombe told reporters. <a href="http://twitter.com/home?status=Quebec+To+Impose+French-Language+Quotas+On+Streaming+Giants%3A+https%3A%2F%2Fyro.slashdot.org%2Fstory%2F25%2F05%2F21%2F2037230%2F%3Futm_source%3Dtwitter%26utm_medium%3Dtwitter" rel="nofollow"><img src="https://a.fsdn.com/sd/twitter_icon_large.png"></a> <a href="http://www.facebook.com/sharer.php?u=https%3A%2F%2Fyro.slashdot.org%2Fstory%2F25%2F05%2F21%2F2037230%2Fquebec-to-impose-french-language-quotas-on-streaming-giants%3Futm_source%3Dslashdot%26utm_medium%3Dfacebook" rel="nofollow"><img src="https://a.fsdn.com/sd/facebook_icon_large.png"></a> at Slashdot.