After a tenure of eight years, a Tesla engineer has quit the company — and directed a seething missive toward his boss, Elon Musk.
"I do need to address the elephant in the room: the main reason I'm leaving is that I think Elon has dealt huge damage to Tesla's mission (and to the health of democratic institutions in several countries)," Giorgio Balestrieri, a staff algorithms engineer who joined in 2017, wrote in a scathing <a href="https://www.linkedin.com/feed/update/urn:li:activity:7371854385812824065/">LinkedIn post</a> last week. "Beyond that, Elon's leadership and decision making seem seriously compromised."
Balestrieri worked on Autobidder, Tesla's energy trading platform that allows individuals and businesses to monetize the energy they store in battery systems, often generated by renewables like wind and solar. 
According to Balestrieri, the company's battery storage initiatives are "key to operating low-carbon grids," while "data and algorithms have taken center stage in the energy transition."
But he no longer feels that Musk is committed to championing that transition to a greener energy future. 
"This is not just about politics: it's about lying to the public, manipulating public discourse, targeting minorities and supporting climate change deniers and political forces aligned with the oil and gas industry," Balestrieri wrote. "I think it's fairly indisputable that the current US administration is slowing down the energy transition."
Musk built Tesla and his own brand on a green promise: that solar panels and electric cars would help us get out of climate doom. 
But Musk then became a close ally of president Donald Trump — at least until a <a href="https://futurism.com/trump-shreds-musk">spectacular falling out</a> earlier this year — whose administration has targeted climate science and made overtures to the fossil fuels industry. In his first days in office, Trump <a href="https://www.nytimes.com/2025/01/20/climate/trump-emergency-oil-gas.html">signed executive orders</a> that promoted oil, coal, and gas companies, eliminated pollution protections, and declared a <a href="https://www.npr.org/2025/01/20/nx-s1-5268653/energy-emergency-trump-oil-evs">national "energy" emergency</a> to ramp up production of fossil fuels. He also formally <a href="https://www.nytimes.com/2017/06/01/climate/trump-paris-climate-agreement.html">withdrew the US from the Paris climate accord</a>, an international agreement to limit global warming to 2.7 degrees Fahrenheit, for the second time.
Trump has since tried to freeze billions of dollars in climate spending earmarked by Congress and his predecessor Joe Biden, though a judge has <a href="https://www.reuters.com/sustainability/climate-energy/us-judge-orders-immediate-thaw-climate-infrastructure-funds-politico-reports-2025-04-15/">temporarily blocked the order</a>. Meanwhile, federal research grants are <a href="https://futurism.com/trump-pulls-funds-research-climate">being canceled</a> for merely containing the word "climate," government websites are being <a href="https://futurism.com/trump-climate-change-government-websites">scrubbed of any information on climate change</a>, and NASA's crucial climate satellite missions <a href="https://futurism.com/white-house-orders-nasa-destroy-important-satellite">are set to be destroyed</a>.
Where has Musk stood on all of this? Who can say, since he's been too busy <a href="https://futurism.com/government-undo-elon-musk-doge-damage">eviscerating the federal government</a>, raging <a href="https://futurism.com/elon-musk-trial-trump-gold-card">against immigrants</a> and <a href="https://futurism.com/neoscope/elon-musk-traveling-box-drugs">allegedly getting high out of his mind</a>. But not too long ago, he did take a stand: during Trump's first term, Musk famously stormed out of his advisory council roles after the president pulled the US out of the Paris accords, and was hailed as a hero.
"Climate change is real," Musk <a href="https://x.com/elonmusk/status/870369915894546432">tweeted</a> in June 2017. "Leaving Paris is not good for America or the world."
That, incidentally, was the same year Balestrieri joined Tesla energy. Whatever warm feelings the stunt may have instilled have long gone cold, and it's safe to say that Musk's image as a eco-conscious, <a href="https://futurism.com/the-byte/elon-musk-donates-million-trees">tree-planting champion</a> has gone up in greenhouse-gas laden smoke.
"Given his huge (and growing, inexplicably) stake in Tesla," Balestrieri soberly reflected in the year of 2025, amid Tesla's plan to <a href="https://futurism.com/tesla-give-elon-musk-1-trillion">give Musk a trillion dollars in shares</a>, "I can't convince myself anymore that this is the right place to be."
"I wish the best to everyone who will stick around and keep working to accelerate the transition to sustainable energy," he added, "and I hope the best for Tesla and Tesla Energy."
More on Elon Musk: <a href="https://futurism.com/spacex-launch-fans-grief">People Who Loved Watching SpaceX Launches Can’t Stomach Them Anymore Since Elon Musk Started Being So Horrible</a>

Burning Bridges

"Elon's leadership and decision making seem seriously compromised."

Tesla Engineer Quits, Roasts Elon Musk in Spectacular Fashion

An increasing number of commercial airline passengers and flight attendants are being exposed to dangerous fumes while in the air, at times resulting in traumatic brain injuries that closely resemble the <a href="https://www.sciencedirect.com/science/article/abs/pii/S135917891930028X">deadly </a><a href="https://www.sciencedirect.com/science/article/abs/pii/S135917891930028X">concussions</a> experienced by professional football players.
As the <a href="https://www.wsj.com/business/airlines/air-travel-toxic-fumes-64839d6e">Wall Street Journal reports</a>, fume incidents have led to emergency landings and sickened passengers being taken to hospitals. Even airline pilots have reported affected reaction times and vision due to the hazardous gases.
Aircraft manufacturers, including Airbus and Boeing, have long known about the cause: malfunctions forcing oil and hydraulic fluid to leak into the engines, which are then vaporized at extreme heat. This mix can result in the production of neurotoxins, carbon monoxide, and other chemicals.
Even worse, according to the paper, manufacturers are cutting costs and downplaying health risks by lobbying against safety measures that could alleviate the problem.
It's a problem that can plague the vast majority of mass-produced aircraft. In commercial jets, fresh air being sucked into a jet engine "bleeds" into the cabin by being compressed and heated. But worn bearing seals can allow oil to mix with the air and be vaporized, leaking toxic compounds into the cabin.
During a February flight, smoke started coming through the vents. Attendants advised passengers to breathe through their clothing, and the pilots announced they would be returning to Atlanta. One of the airplane's engines' oil reservoirs, inspectors found, had vaporized into the cabin.
While manufacturers argue it's not nearly a big enough issue to warrant a comprehensive fix, according to the WSJ, industry insiders have raised concerns about the troubling trend.
The Federal Aviation Administration also notes that incidents are "rare" on its website, though they amounted to a staggering 330 fume events last year in the US alone.
However, the WSJ found in its own investigation that there were actually more than double that number of incidents. The real number could also be far higher, due to incidents that go unreported.
The worst offender, the WSJ found, was Airbus' A320 family of jets.
Experts are only beginning to understand the health implications. JetBlue flight attendant Florence Chesson, who inhaled fumes during a flight to Puerto Rico, told the newspaper that she had developed migraines, heart arrhythmias, and sensitivity to lights.
Occupational medicine specialist Robert Harrison, who has treated over 100 crew members following toxic exposure, has found that incidents can lead to damage to the central nervous system.
Despite the issue being formally recognized by the United Nations ten years ago, aircraft manufacturers have yet to make any major changes to mitigate risks, the WSJ reports.
A bipartisan bill <a href="https://www.congress.gov/bill/119th-congress/house-bill/4994/text">recently reintroduced</a> in Congress is calling for the installation of specialized filters and the phasing out of bleed air. A <a href="https://www.congress.gov/bill/118th-congress/senate-bill/615/text">previous attempt</a> was ultimately shut down after facing industry pushback last year.
That doesn't mean aircraft manufacturers couldn't fix the issue if they wanted to. Boeing's 787 Dreamliner features a new air circulation system that makes the problem impossible.
But when Boeing's leadership was determining how to market the new system when it was rolling out, according to emails obtained by the WSJ, one executive worried that "if we elaborate on the 787 air purification and say how great and important it is," it would draw attention to poor air quality on the company's existing planes.
More on airplanes: <a href="https://futurism.com/the-byte/united-airlines-starlink-wifi">United Airlines Shuts Down Starlink WiFi Service on Its Planes After the Antennae Caused Problems With Its Jets' Equipment</a>

In-Air Fumigation

Developments

The industry is attempt to downplay the health risks.

Commercial Airplanes Are Dosing Passengers With Neurotoxic Gases

In a <a href="https://arxiv.org/abs/2509.04664">paper</a> published earlier this month, OpenAI researchers <a href="https://futurism.com/openai-mistake-hallucinations">said they'd found the reason</a> why even the most powerful AI models still suffer from rampant "hallucinations," in which products like ChatGPT confidently make assertions that are factually false.
They found that the way we evaluate the output of large language models, like the ones driving ChatGPT, means they're "optimized to be good test-takers" and that "guessing when uncertain improves test performance."
In simple terms, the creators of AI incentivize them to guess rather than admit they don't know the answer — which might be a good strategy on an exam, but is outright dangerous when giving high-stakes advice about topics like <a href="https://futurism.com/neoscope/google-healthcare-ai-makes-up-body-part">medicine</a> or <a href="https://futurism.com/anthropic-chatbot-legal-music">law</a>.
While OpenAI claimed in an <a href="https://openai.com/index/why-language-models-hallucinate/">accompanying blog post</a> that "there is a straightforward fix" — tweaking evaluations to "penalize confident errors more than you penalize uncertainty and give partial credit for appropriate expressions of uncertainty" — one expert is warning that the strategy could pose devastating business realities.
In an <a href="https://theconversation.com/why-openais-solution-to-ai-hallucinations-would-kill-chatgpt-tomorrow-265107">essay for The Conversation</a>, University of Sheffield lecturer and AI optimization expert Wei Xing argued that the AI industry wouldn't be economically incentivized to make these changes, as doing so could dramatically increase costs.
Worse yet, having an AI repeatedly admit it can't answer a prompt with a sufficient degree of confidence could deter users, who love a confidently positioned answer, even if it's ultimately incorrect.
Even if ChatGPT admitted that it doesn't know the answer just 30 percent of the time, users could quickly become frustrated and move on, Xing argued.
"Users accustomed to receiving confident answers to virtually any question would likely abandon such systems rapidly," the researcher wrote.
While there are "established methods for quantifying uncertainty," AI models could end up requiring "significantly more computation than today’s approach," he argued, "as they must evaluate multiple possible responses and estimate confidence levels."
"For a system processing millions of queries daily, this translates to dramatically higher operational costs," Xing wrote.
Piling up the expenses at this juncture could prove disastrous. AI companies have bet big on scale, doubling down on expanding infrastructure to run increasingly power-hungry models. But try as they might, a return on investment appears to be <a href="https://www.axios.com/2025/08/21/ai-wall-street-big-tech">many years</a>, if not decades, out. So far, tens of billions of dollars worth of capital expenditures have <a href="https://www.wheresyoured.at/why-everybody-is-losing-money-on-ai/">eclipsed</a> relatively modest revenues.
In other words, increasing already sky-high operational costs — while alienating users — could be yet another major thorn in the side of firms like OpenAI as they race to reassure investors that there's a feasible business model in the long term.
Xing argued that the company's proposed fixes for hallucinations may work for "AI systems managing critical business operations or economic infrastructure" as "the cost of hallucinations far exceeds the expense of getting models to decide whether they’re too uncertain."
"However, consumer applications still dominate AI development priorities," he added. "Users want systems that provide confident answers to any question."
Arriving at a more uncertain answer faster is inherently cheaper for companies, which could disincentivize a more careful and confident approach involving fewer hallucinations.
How all of this will play out in the long term is anybody's guess, especially as market forces continue to shift and companies find more efficient ways to run their AI models.
But one thing is unlikely to change: guessing will always remain a far more economical and affordable option.
"In short, the OpenAI paper inadvertently highlights an uncomfortable truth," Xing concluded. "The business incentives driving consumer AI development remain fundamentally misaligned with reducing hallucinations."
"Until these incentives change, hallucinations will persist," he added.
More on hallucinations: <a href="https://futurism.com/openai-mistake-hallucinations">OpenAI Realizes It Made a Terrible Mistake</a>

Cure Worse Than the Disease

It's a lose-lose situation.

Fixing Hallucinations Would Destroy ChatGPT, Expert Finds

If you're a coward and don't want to munch on awful-tasting shrooms to go on a psychedelic odyssey, you can buy delicious edibles infused with their active ingredient instead: psilocybin.
But maybe you shouldn't, because new research suggests that you might be getting ripped off taking these — or even putting your health at risk.
After testing 12 edible "magic mushroom" products sold in Portland, Oregon, Richard van Breemen and his colleagues found absolutely zero traces of psilocybin in any of them, they reported in a <a href="https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2838765">new study</a> published in the journal JAMA Network Open. And four appeared to contain no active, trip-inducing ingredients at all.
"We found no evidence of mushroom compounds of any kind, coming from any species," van Breemen, a pharmaceutical sciences professor at Oregon State University, <a href="https://www.scientificamerican.com/article/magic-mushroom-edibles-found-to-contain-no-psilocybin/">told Scientific American</a>.
Moreover, the researchers found ingredients that weren't written on the label, including — alarmingly — synthetic psychedelics whose effects on human health aren't well studied.
"We don't know what harm they might cause," van Breemen warned.
Psilocybin is touted for its trippy, recreational effects as much as it is for its purported <a href="https://futurism.com/neoscope/psilocybin-treatment-resistant-depression">mind-opening</a> <a href="https://futurism.com/neoscope/psilocibin-mushrooms-aging-study">health benefits</a>. Long illicit, some states like Oregon and New Mexico have in recent years moved to legalize or at least decriminalize the substance, but sometimes with the catch that they need to be administered in a supervised medical setting. And so outside those pricey official channels, loopholes allow shops to sell the psychedelic substance at cheap prices. This weird legal limbo has birthed a mini — and technically above-board — industry that peddles the drugs, but with little oversight.
"A lot of people are very curious about these substances," Mason Marks, a law professor at Florida State University, told SciAm. "And if you're in a state, like Oregon, that does not decriminalize them, people might go to these shops and buy these products that are either blatantly illegal or kind of in this gray area."
That's in a nutshell what van Breemen's team did for the study: they went around Portland buying a host of magic mushroom edibles — mostly gummies but one chocolate — from gas stations and convenience stores.
Initial tests quickly established that none of these actually contained psilocybin. Then, by using mass spectrometry and a technique called liquid chromatography to separate the psychedelic snacks into their individual components, the team found that seven edibles had undisclosed ingredients like caffeine and THC, the drug that produces cannabis' effects.
That's noteworthy on its own, since it's appalling to dose someone with something other than what they intended to take. The most disconcerting find, however, were two different synthetic psychedelic substances whose effects on human health have never been studied, mipracetin and 4-hydroxy-diethyltryptamine. That means oblivious customers, under the impression that they're simply enjoying a more convenient form of well-studied psilocybin, could be wreaking havoc on their brains without even knowing it.
Other finds suggest an industry in desperate need of stronger oversight. Two of the edible products contained psilocin, the substance that psilocybin is broken down into once it's ingested by your body. But the lack of other mushroom compounds suggest that the psilocin was synthetic, they concluded — so customers aren't getting the natural "magic mushrooms" that're being advertised, while also raising questions about how safely the drug was synthesized.
"Over the years, we've tested many botanical dietary supplements," van Breemen told SciAm. "Rarely do you find 12 products, none of which contain what they're supposed to contain. So the sample number is small, but the number of mislabeled or adulterated products was significant."
We see a similar Wild West scenario in the weed industry. Without widespread legalization but a huge appetite for THC products, shady weed vapes and cartridges have flooded the market. Some have been found to <a href="https://www.latimes.com/california/story/2024-12-19/we-tested-cannabis-products-for-pesticides-how-dirty-is-your-weed">contain pesticides</a> and <a href="https://www.healthline.com/health-news/toxic-heavy-metals-cannabis-vaping-liquids">toxic metals</a>. (So have nicotine vapes, <a href="https://futurism.com/neoscope/vape-clouds-horrifying-chemicals">for that matter</a>.)
While it's a good thing that fewer people are going to jail for getting high, a lagging regulatory framework has left an opening for questionable companies to peddle untested goods on eager — but naïve — clientele. Last year, the Centers for Disease Control and Prevention <a href="https://www.cdc.gov/environmental-health-studies/outbreak-investigation-diamond-shruumz-products/index.html">began investigating reports</a> of severe illness after consuming a popular brand of psilocybin edibles, SciAm noted. As of October 2024, 73 hospitalizations and potentially three deaths were tallied across 34 states.
"The mislabeling highlighted in this [study] causes harms," Lori Bruce, a Yale University bioethicist who studies psychedelics, told the magazine. "And as usage rates increase across the US, harms from retail products are also likely to increase."
More on drugs: <a href="https://futurism.com/ai-therapy-psychedelic-trip-sitter">People Are Taking Massive Doses of Psychedelic Drugs and Using AI as a Tripsitter</a>

Magic Psych

Mental Health

"We don't know what harm they might cause."

Scientists Tested What's Really in "Psilocybin" Chocolate and Found Something Alarming

We tend to get fixated on realizing certain contraptions from science fiction, no matter how impractical. <a href="https://futurism.com/the-byte/electric-car-giant-drone-propellers">Flying cars</a>. <a href="https://futurism.com/i-flew-a-jetpack">Jet packs</a>. And now, humanoid robots.
It's only the latter, however, that's anticipated to one day be a <a href="https://www.morganstanley.com/insights/articles/humanoid-robot-market-5-trillion-by-2050">multitrillion dollar market</a>, with big firms like Tesla leading the charge. In March, Tesla CEO Elon Musk <a href="https://www.inc.com/kit-eaton/was-tesla-robots-capitol-hill-debut-a-preview-of-factory-job-losses/91167677">boasted</a> that his automaker would build 5,000 Optimus robots by the end of this year. Responding to Musk's claim, <a href="https://www.scmp.com/tech/big-tech/article/3303666/chinese-humanoid-robot-maker-agibot-plans-match-musks-optimus-output-year">so did the Shanghai-based startup Agibot</a>. Norwegian robot maker 1X announced it would be <a href="https://techcrunch.com/2025/03/21/1x-will-test-humanoid-robots-in-a-few-hundred-homes-in-2025/">testing its robot in thousands of homes</a> before 2026, too. And the AI robotics firm Figure <a href="https://www.forbes.com/sites/johnkoetsier/2025/01/30/figure-plans-to-ship-100000-humanoid-robots-over-next-4-years/">plans to ship</a> 100,000 humanoid robots in the next four years. 
Yet few of these promised robots have actually been built. And there're still huge existential questions looming over the industry, as outlined in a new <a href="https://spectrum.ieee.org/international-conferences-sydney">writeup by </a><a href="https://spectrum.ieee.org/humanoid-robot-scaling">IEEE Spectrum</a> — and it's not clear if some of these obstacles will be overcome in the near future.
Exemplifying the pie-in-the-sky thinking that's dominating the space is that many are betting on advances in AI — another tech awash in outrageous promises and clouded by a murky future — magically resulting in all-purpose robots, according to Melonee Wise, former chief product officer at Agility Robotics.
"I think what a lot of people are hoping for is they're going to AI their way out of this," Wise told IEEE. "But the reality of the situation is that currently AI is not robust enough to meet the requirements of the market."
That's painfully apparent. In a <a href="https://futurism.com/elon-musk-new-robot-demo-bad">video demo shared last week</a>, Salesforce CEO Mark Benioff, accompanied by Musk, asks a Tesla Optimus bot for a Coke. But the bot takes an awkward number of seconds before responding, cuts out mid-sentence, and then stands unresponsively while Benioff repeatedly tells the robot to get going. When it does, it moves agonizingly slowly, loudly clunking along like an oversized RC toy.
Reminder: there's supposed to be 5,000 of these ready to be shipped by the end of the year, with scant months to go. (Perhaps this is why Tesla has mostly relied on <a href="https://futurism.com/the-byte/tesla-robots-remotely-controlled-analyst">humans teleoperating its automatons</a> whenever it's tried to show them off.)
Another surprisingly pedestrian technological hurdle is battery life. Agility's "Digit" robot, for example, can run for 90 minutes tops while taking 9 minutes to fully recharge. But only 30 minutes of that time is usable under normal circumstances, since the extra hour of charge is held in reserve for when work needs to be put on pause, which happens a lot on the factory floor — otherwise it might run out of battery mid-task, according to IEEE. Now imagine a factory floor with hundreds of these robots with heavy battery packs that need to be topped off every half-hour.
"No one wants to deal with that," Wise said.
But maybe we're putting the cart before of the horse here. Will factories even think swapping out their reliable and highly specialized machines — or human workers — with experimental automatons is worth the investment?
"The bigger problem is demand," Wise told IEEE. "I don't think anyone has found an application for humanoids that would require several thousand robots per facility," she added.
Amazon is one of the biggest example of an industrial powerhouse embracing robot automation, with <a href="https://futurism.com/robots-outnumber-humans-amazon-warehouses">robots on track to outnumber humans at its vast warehouses</a> — but its most successful ones are highly specialized units like robotic arms designed for loading and unloading boxes. Though the e-commerce giant is <a href="https://futurism.com/amazon-delivery-robots">experimenting with humanoid machines</a>, like <a href="https://www.cnbc.com/2025/05/07/meet-amazons-robot-vulcan-the-first-with-a-sense-of-touch.html">Agility's "Digit" bot</a>, it's nowhere near to deploying them at a large scale.
Even if manufacturers are willing to gamble, industrial applications for humanoid robots would have trouble getting off the ground, according to Wise, because they would fall under the already heavily regulated field of industrial machinery. That means the industry wouldn't enjoy the same leeway that emerging tech like autonomous cars enjoyed, for example.
All told, while the field of humanoid robotics is generating loads of hype — and even glimpses here and there of sci-fi potential — it hasn't shown that it's anywhere near ready to becoming the juggernaut its investors are banking on.
More on robots: <a href="https://futurism.com/video-humanoid-robot-laundry">Wild Video Shows Humanoid Robot Effortlessly Folding Laundry</a>

Robots In Demise

An all-capable robo-butler, coming to households soon? We wouldn't hold our breath.

When You Read the Fine Print, Humanoid Robots Are Going Absolutely Nowhere

In his 1963 scifi story "The Invincible," the Polish writer Stanisław Lem imagined an artificial species of free-floating nanobots which roamed the atmosphere of a far-off planet. Like tiny bugs, the microscopic beings were powerless alone, but together they could form cooperative swarms to gather energy, reproduce, and ultimately defend their territory from predators with deadly force.
Unlike the story’s human protagonists, the "black cloud" of bots was incapable of reasoning beyond the simple logic of animal instincts. But when the two life forms inevitably come into conflict, literary critic <a href="https://solaris.lem.pl/ksiazki/beletrystyka/niezwyciezony/96-poslowie-niezwyciezony">Jerzy Jarzębski writes</a>, human evolution proves its mettle over the mindless automaton — not by eradicating the deadly species, but by making a conscious decision to let it live.
Lem probably never imagined his evolutionary parable of living dust was just a few decades from becoming a reality — or that it would become the inspiration for the development of a real-life military technology known as "smart dust."
Starting out as a <a href="https://people.eecs.berkeley.edu/~pister/SmartDust/SmartDustBAA97-43-Abstract.pdf">theoretical research proposal</a> to the Defense Advanced Research Projects Agency (DARPA) — the Cold War-era military tech bureau behind everything from <a href="https://www.af.mil/News/Article-Display/Article/703894/evolution-of-gps-from-desert-storm-to-todays-users/">GPS navigation</a> to the <a href="https://www.intereconomics.eu/contents/year/2025/number/2/article/big-tech-and-the-us-digital-military-industrial-complex.html">modern internet</a> — smart dust is now being developed for use in a wide variety of industries, from <a href="https://www.sciencedirect.com/science/article/abs/pii/S0010854523004769?via%3Dihub">environmental studies</a> to <a href="https://www.nature.com/articles/s41598-024-58021-x">commercial mining</a>.
That’s <a href="https://interestingengineering.com/innovation/smart-dust-scientists-world-sensor">according to Interesting Engineering</a>, which recently published a rundown of the state of present-day smart dust after decades of development. Though "dust" remains a bit of a misnomer — it's more like a bunch of tiny sensors capable of delaying data to a central device — there's a large body of theoretical and simulated work laying a path for practical microengineering that's steadily coming into its own.
Case in point, while nanotech began as an effort to build relatively simple <a href="https://ieeexplore.ieee.org/document/1037346">wireless receivers</a> around the size of a grain of rice, thanks to decades of R&amp;D, some motes being developed are now <a href="https://www.google.com/amp/s/theintelligentminer.com/2019/04/09/the-intelligent-guide-to-smart-dust-in-mining/amp/">nearly invisible</a> to the naked eye, measuring in at anywhere from 1 cubic millimeter to .02 cubic millimeters.
As early as 2003, micro-sensor platforms like <a href="https://people.eecs.berkeley.edu/~culler/papers/mica-sensors.pdf">Crossbow Technology, Inc's</a> "MICA" and <a href="https://www.computer.org/csdl/magazine/co/2001/01/r1044/13rRUxjQy7J">UC Berkeley's</a> "Spec" have successfully detected all kinds of variables while measuring in at mere millimeters, recording changes in humidity, light and temperature.
Recent developments within the past 10 years have expanded these sensor's abilities to record<a href="https://convergenceinstruments.com/pdf/Dust_Issues.pdf?srsltid=AfmBOorfzg44d-GjMjUDQSS4aiovu-JCbIpg5dSCLBH-AZhojNCwqQNY"> various levels of sound</a>, and work is underway to develop motes capable of detecting the <a href="https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202419052">chemical composition of the air</a>. They can be used individually to record changes in the <a href="https://engineering.berkeley.edu/news/2016/11/neural-dust/">human body</a>, or <a href="https://www.pnas.org/doi/10.1073/pnas.1233824100">deployed in swarms</a> to identify biological compounds.
In the future, the mites are hoped to be able to report a <a href="https://scienceofsingularity.com/2020/07/22/smart-dust-utility-fog-virtual-people-the-future-of-programmable-matter/">near-infinite amount of data</a> in suspended, 3D environments — like a microscopic version of Bill Paxton’s "Dorothy" sensors in the 1996 weather thriller "Twister."
Per IE, the current "smart dust industry," made up of tech companies like <a href="https://www.processindustryinformer.com/dust-networks-and-emerson-process-management-partner-to-deliver-fips-197-advanced-encryption-standard-certified-solutions/">Emerson Process Management</a> and <a href="https://www.cs.wustl.edu/~lu/wsn_articles/smart_dust_cnn.pdf">Hewlett-Packard</a>, was valued at around $115 million in 2022. By 2032, it’s expected to reach nearly $400 million.
While various militaries are keen on developing smart dust for intelligence reasons, much of the present research is carried in university and corporate labs. An Israeli firm called Stardust Solutions, for example, <a href="https://thebulletin.org/2025/03/reach-for-the-stardust-former-israel-atomic-energy-commission-official-leads-controversial-geoe">drew concerns</a> from the Bulletin of the Atomic Scientists when it announced its intentions of releasing a variation on smart dust to block out the Sun — involving inert particular matter in conjunction with an atmospheric monitoring system — in violation of <a href="https://www.ciel.org/news/us-israeli-start-up-unveils-reckless-geoengineering-gamble/">international geoengineering laws</a>.
While the tech is pretty dystopian as it is, there’s a lot of room for improvement. The need to interface with a centralized data-processing unit, for example, means the tiny units can’t travel too far from their human controller. Their usable lifespan is likewise pretty short, though that’s changing with innovations in <a href="https://interestingengineering.com/innovation/rice-grain-size-microbatteries-power-smart-dust">energy-harvesting capabilities</a> via light, vibrations, and electromagnetic fields.
One thing’s abundantly clear: now might be a good time to invest in an air purifier.
More on tiny tech: <a href="https://futurism.com/china-nano-drone">Chinese Military Shows Off Fly-Sized Drones for Covert Ops</a>

Little Miracles

Is anyone else itchy?

Scientists Working on "Smart Dust" That Can Spy on a Room While Drifting Throught the Air

SpaceX has <a href="https://futurism.com/the-byte/former-nasa-official-spacex">lots</a> <a href="https://futurism.com/the-byte/elon-musk-feud-jeff-bezos-spacex-starship">of</a> <a href="https://futurism.com/elon-musk-spacex-company-town-rights">enemies</a> — but this may be the most unexpected yet.
As <a href="https://www.space.com/space-exploration/human-spaceflight/the-turtles-and-the-nudists-will-have-to-migrate-spacex-plan-for-starship-launches-from-florida-sparks-debate-among-space-coast-residents">Space.com reported</a>, members of the American Association for Nude Recreation (AANR) — yes, that's a real group — were among those protesting the Elon Musk-owned company's proposed use of Kennedy Space Center on the Sunshine State's so-called Space Coast.
Among the many potential impacts associated with bringing SpaceX to Florida — including hearing damage to local residents and flight delays across the state, per various opponents — is the repeat closure of Playalinda Beach, a remote location on the Canaveral National Seashore that's one of Florida's four main public beaches <a href="https://www.visitflorida.com/beaches/faq/nude-beaches/">where nudity is legal</a>.
Popular among the clothing-optional set, Playalinda could be closed up to 60 times per year due to Starship operations — and Deborah-Sue Stevens, a <a href="https://aanrwest.org/media/attachments/2022/09/07/2022-regionalassemblyminutes.pdf">former regional director</a> for the nudist organization's Western branch, says those closures will harm many members of her community.
"[There are] probably a quarter million people that travel and think like I do," Stevens said, calling into the Zoom meeting from California, "who look for destinations that are beautiful and surrounded by like-minded people."
Erich Schuttauf, the executive director of AANR — which bills itself as the largest and longest-established nudist organization in the US with foundations dating back to 1931 — added that Playalinda closures might affect the non-nude community too.
As Schuttauf explained, nudists may end up traveling to nearby Volusia County, whose <a href="https://www.news-journalonline.com/story/news/local/volusia/2024/03/27/apollo-beach-what-to-know-about-volusia-countys-nude-beach/73110804007/">clothing-optional Apollo Beach</a>, located further along the Canaveral shore, is also popular among those who like to sunbathe in their birthday suits.
Should Playalinda be removed as an option, Schuttauf warned, nudists may overtake Apollo and end up going over the clothing optional line, causing some decidedly Floridian conflicts between them and their clothed counterparts.
Not everyone was interested in preserving the nude beach's current status, however. Max West, a random man quoted by Space.com who claimed during the Zoom call that he plans to move to the area to photograph Starship launches, essentially suggested that the nudists suck it up.
"The turtles and the nudists will have to migrate," West said, presumably alluding to <a href="https://www.cnn.com/2025/07/01/americas/space-x-debris-mexico-sea-turtles-intl-latam">disputes over whether</a> SpaceX's rocket debris pose a hazard to endangered sea turtles . "That's the cost that you have to pay for this incredible stuff that's happening."
Unfortunately, it looks like that wannabe SpaceX photographer may end up getting his way.
Although the FAA hasn't yet completed its environmental review for SpaceX's request to move to Florida, and has yet to determine whether Musk's company can meet both its financial and safety requirements to operate at the Kennedy Space Center, construction of a Starship tower has nevertheless been underway for more than a year — giving the impression that approval is a foregone conclusion.
More on SpaceX: <a href="https://futurism.com/spacex-starship-delays">SpaceX Suddenly Seems Pretty Terrified to Launch Starship After Long String of Super Expensive Explosions</a>

Naked Glum

"The turtles and the nudists will have to migrate."

Nudists Declare War on SpaceX

OpenAI claims to have figured out what's driving "hallucinations," or AI models' strong tendency to make up answers that are factually incorrect.
It's a major problem plaguing the entire industry, greatly undercutting the usefulness of the tech. Worse yet, experts have found that the problem is <a href="https://futurism.com/ai-industry-problem-smarter-hallucinating">getting worse as AI models get more capable</a>.
As a result, despite incurring astronomical expenses in their deployment, frontier AI models are still prone to <a href="https://futurism.com/gpt-5-huge-factual-errors">making inaccurate claims</a> when <a href="https://futurism.com/ai-makes-up-answers">faced with a prompt they don't know the answer to</a>.
Whether there's a solution to the problem remains a hotly debated subject, with some experts arguing that hallucinations are <a href="https://futurism.com/the-byte/impossible-chatbots-stop-lying-experts">intrinsic to the tech itself</a>. In other words, large language models may be a <a href="https://www.livescience.com/technology/artificial-intelligence/current-ai-models-a-dead-end-for-human-level-intelligence-expert-survey-claims">dead end</a> in our quest to develop AIs with a reliable grasp on factual claims.
In a <a href="https://arxiv.org/abs/2509.04664">paper</a> published last week, a team of OpenAI researchers attempted to come up with an explanation. They suggest that large language models hallucinate because when they're being created, they're incentivized to guess rather than admit they simply don't know the answer.
Hallucinations "persist due to the way most evaluations are graded — language models are optimized to be good test-takers, and guessing when uncertain improves test performance," the paper reads.
Conventionally, the output of an AI is graded in a binary way, rewarding it when it gives a correct response and penalizing it when it gives an incorrect one.
In simple terms, in other words, guessing is rewarded — because it might be right — over an AI admitting it doesn't know the answer, which will be graded as incorrect no matter what.
As a result, through "natural statistical pressures," LLMs are far more prone to hallucinate an answer instead of "acknowledging uncertainty."
"Most scoreboards prioritize and rank models based on accuracy, but errors are worse than abstentions," OpenAI wrote in an <a href="https://openai.com/index/why-language-models-hallucinate/">accompanying blog post</a>.
In other words, OpenAI says that it — and all its imitators across the industry — have made a grave structural error in how they've been training AI.
There'll be a lot riding on whether the issue is correctable going forward. OpenAI claims that "there is a straightforward fix" to the problem: "Penalize confident errors more than you penalize uncertainty, and give partial credit for appropriate expressions of uncertainty."
Going forward, evaluations need to ensure that "their scoring discourages guessing," the blog post reads. "If the main scoreboards keep rewarding lucky guesses, models will keep learning to guess."
"Simple modifications of mainstream evaluations can realign incentives, rewarding appropriate expressions of uncertainty rather than penalizing them," the company's researchers concluded in the paper. "This can remove barriers to the suppression of hallucinations, and open the door to future work on nuanced language models, e.g., with richer pragmatic competence."
How these adjustments to evaluations will play out in the real world remains to be seen. While the company claimed its latest GPT-5 model hallucinates less, users were <a href="https://futurism.com/gpt-5-disaster">left largely unimpressed</a>.
For now, the AI industry will have to continue reckoning with the problem as it justifies tens of billions of dollars in capital expenditures and <a href="https://www.npr.org/2024/07/12/g-s1-9545/ai-brings-soaring-emissions-for-google-and-microsoft-a-major-contributor-to-climate-change">soaring emissions</a>.
"Hallucinations remain a fundamental challenge for all large language models, but we are working hard to further reduce them," OpenAI promised in its blog post.
More on hallucinations: <a href="https://futurism.com/gpt-5-huge-factual-errors">GPT-5 Is Making Huge Factual Errors, Users Say</a>

Lies Generator

It's "working hard" to correct it.

OpenAI Realizes It Made a Terrible Mistake

For many rocketry enthusiasts, watching SpaceX launches was an exciting hobby — one that they'd <a href="https://texashighways.com/culture/people/elon-musk-space-x-site-turned-boca-chica-into-a-tourist-attraction/">travel down</a> to the very bottom of Texas to witness in person at the Starbase launch facility off the Gulf of Mexico.
But with CEO Elon Musk now fully entrenched in <a href="https://futurism.com/the-byte/tesla-drivers-sue-elon-musk-right-symbols">bigoted right-wing politics</a>, that joy has turned to grief for former fans, who now lament the loss of their once-joyful pastime.
"Most of the time, watching a rocket launch as a grown-up fills me with the same feelings I had when I saw my first as a kid — fear, glee, wonder and excitement — as a spaceship makes its way into a largely foreign environment," wrote Emily Carney, a space historian and author, <a href="https://spacenews.com/the-grief-of-a-fandom-on-starship-musk-and-losing-the-spark/">for SpaceNews</a>. "It really never gets old."
However, at the 10th Starship launch from at the end of August — SpaceX's "<a href="https://futurism.com/part-of-starship-explodes">most successful</a>" test flight of the heavy-lift vehicle yet, which should have been an exciting feat to witness — Carney recounted feeling dejected instead of elated.
"I should be happy, right? Yeah, um, no," she wrote. "I have been a space fan for pretty much all of my life, yet I couldn’t find the enthusiasm even to watch the launch and mission unfold."
Turning to Threads, Instagram's Twitter clone, the historian found she wasn't alone in that feeling.
"It’s very sad how SpaceX probably has had its biggest success tonight with Starship and I, a space freak for nearly 44 years, couldn’t be bothered to watch it," <a href="https://www.threads.com/@emilycarneyspace/post/DN1rkg1Wsqs?xmt=AQF0YNfQMnY1sWn7GcCiFArMrmKDAl0zH0pP1iI8_nndLw">Carney posted</a>.
"This is how I feel," <a href="https://www.threads.com/@mike.maddox.5243/post/DN48vfTioaD?xmt=AQF0YNfQMnY1sWn7GcCiFArMrmKDAl0zH0pP1iI8_nndLw">remarked another user</a>. "Space obsessed since I watched Apollo as a five year old. Space mad. Read every book I could, built my own rockets, named a child after a cosmonaut, cried the whole way around [Kennedy Space Center] — and I just feel nothing."
"I’m trying to take comfort in the fact that a lot of good, heartfelt effort and dedication have gone into the Starship program long before Musk became whatever the pure awfulness he is today," <a href="https://www.threads.com/@photo_ryan/post/DN1ufnw2BFy?xmt=AQF0YNfQMnY1sWn7GcCiFArMrmKDAl0zH0pP1iI8_nndLw">wrote yet another Threads user</a>. "Science, NASA, and the space program will succeed and continue in spite of him. He’s definitely ruined much of the joy and tainted this for so many though."
It's not hard to see why those onetime SpaceX-watchers feel so crestfallen, given that Musk's horrendous politics quite literally <a href="https://futurism.com/neoscope/mothers-children-dying-musk-cuts">have a death toll</a>. Thanks to the <a href="https://futurism.com/elon-musk-usaid-constructing-reality">billionaire's draconian cuts</a> to grants for the United States Agency for International Development (USAID) during his time in Donald Trump's White House, an estimated 360,000 people <a href="https://www.schatz.senate.gov/news/press-releases/schatz-details-trump-administrations-destruction-of-usaid-deadly-consequences-that-followed-as-senate-considers-codifying-doge-cuts">have already died</a> from starvation and lack of medication, and an additional 14 million lives could be lost should those cuts be codified and extended through 2030, per a <a href="https://newsroom.ucla.edu/stories/USAID-cuts-global-impact-14-million-deaths">new study</a> in The Lancet, an esteemed medical journal.
With the USAID cuts alone being enough for any avowed SpaceX-watcher to turn their back on the Musk-owned company, there are plenty of other reasons for people like Carney — whose book "<a href="https://www.nebraskapress.unl.edu/nebraska/9781496241399/star-bound/">Star Bound</a>," co-written with space scion Bruce McCandless III, details the history of NASA's spaceflight program — to grieve.
"When asked by some why I felt a lack of interest in [the latest launch of] Starship, I pointed out Musk’s January gesture, which to me could not be mistaken as anything other than what it appeared to be: a Nazi salute," she wrote of the CEO's infamous Sieg Heil at Trump's inauguration. "I saw this happen with my eyes."
Add in Musk's <a href="https://futurism.com/civil-rights-groups-horrified-elon-musk-racist">racism</a>, <a href="https://futurism.com/jewish-leaders-boycott-elon-musk-nazi-salute">anti-Semitism</a>, and "<a href="https://www.nbcnews.com/news/world/south-africa-racist-white-farmers-trump-musk-genocide-ramaphosa-rcna190749">white genocide</a>" outbursts, and you have a wannabe spaceman who "never had to enter himself into US politics and space policy," Carney wrote, instead choosing to enter that fray and <a href="https://futurism.com/mars-society-expert-elon-musk-ruining">irreparably harm</a> NASA and the US space program along the way.
As for erstwhile SpaceX fans, that hurts.
"A considerable part of the space fandom is experiencing a specific kind of grief," she wrote, "when the thing you most enjoyed is now associated with other things that are irretrievably dark."
More on Musk's public: <a href="https://futurism.com/elon-musk-most-hated-person-america-gallup">Elon Musk Appears to Now Be the Most Hated Person in America, According to New Research</a>

Fan Fatality

"A considerable part of the space fandom is experiencing a specific kind of grief."

People Who Loved Watching SpaceX Launches Can’t Stomach Them Anymore Since Elon Musk Started Being So Horrible

Nepal's government collapsed this week following an ill-advised attempt to ban social media outright.
The lifting of the social media ban on Monday was unable to quell a surge in anger. Youth-led groups of disaffected "Gen Z" activists clashed with police in the streets of the capital, Kathmandu, protesting growing wealth inequality and government corruption.
Over <a href="https://www.bbc.com/news/articles/c179qne0zw0o">50 people were killed</a> in the clashes as angry crowds set government buildings ablaze, forcing prime minister Sharma Oli to resign on Tuesday.
Ironically, social media — one of the core aspects of this week's clashes — has turned into an important forum as protesters looked to elect a new leader. As the <a href="https://www.nytimes.com/2025/09/11/world/asia/nepal-protest-genz-discord.html">New York Times reports</a>, young people are using Discord, a group chat platform conventionally used for playing video games, to debate Nepal's fate — and future leader.
"The Parliament of Nepal right now is Discord," a 23-year-old content creator told the newspaper.
It's not just a backroom conversation — the combination of voice, video, and text chats was reportedly being discussed on national TV, yet another wild demonstration of the power of social media during periods of civil unrest.
According to the NYT, the Discord group's widespread support of Nepal’s former chief justice, Sushila Karki, had grown palpable by Wednesday, the day after Oli was ousted.
One server, dubbed "<a href="https://www.thestreet.com/crypto/policy/nepals-gen-z-picks-new-leader-on-discord-crypto-community-in-shock">Youth Against Corruption</a>," accumulated <a href="https://www.france24.com/en/live-news/20250912-from-discord-to-bitchat-tech-at-the-heart-of-nepal-protests">more than 145,000 members</a>.
Discord polls put Karki at the top of the list.
"The point was to simulate a kind of mini-election," Shaswot Lamichhane, a channel moderator and recent high school graduate, told the NYT.
Despite plenty of chaos and organizers struggling to tamp down calls for violence, the young protesters' arguments appeared to have been convincing. Karki <a href="https://www.bbc.com/news/articles/c179qne0zw0o">became the nation's first female prime minister</a> earlier today following dialogue between protest and government leaders.
Crypto markets prediction platform Polymarket <a href="https://x.com/Polymarket/status/1966546206777090541">called</a> Karki the "first world leader elected via Discord."
"I think I just witnessed the first internet-native revolution," one user on the Discord server wrote, as <a href="https://www.thestreet.com/crypto/policy/nepals-gen-z-picks-new-leader-on-discord-crypto-community-in-shock">quoted by The Street</a>.
While it's a particularly powerful example of the influence of social media, Carnegie Endowment for International Peace social movements expert Steven Feldstein told the NYT that the tech may be less successful in creating a "stable political structure in the long term."
Analysts are now calling on the international community to back an interim government.
"Nepal’s closest neighbours, China and India, should support the restoration of democracy and civilian order, and the dialogue among all political forces in Nepal," Nepal's Centre for Social Inclusion and Federalism executive chairman Vijay Kant Karna <a href="https://www.scmp.com/week-asia/politics/article/3325265/nepals-gen-z-protesters-back-ex-chief-justice-sushila-karki-interim-leadership">told the South China Morning Post</a>.
"Meanwhile, all democratic talks should be held at the president’s residence, not at the army barracks," he added.
More on Discord: <a href="https://futurism.com/the-byte/students-hospitalized-hackers-horrifying-pictures">Students Hospitalized After Hackers Send Horrifying Pictures</a>