2025-11-30 04:29:19
The AI Boom Is Based on a Fundamental MistakeBenjamin Riley | The Verge
“The problem is that according to current neuroscience, human thinking is largely independent of human language—and we have little reason to believe ever more sophisticated modeling of language will create a form of intelligence that meets or surpasses our own.”
Why Google’s Custom AI Chips Are Shaking Up the Tech IndustryAlex Wilkins | New Scientist ($)
“Nvidia’s position as the dominant supplier of AI chips may be under threat from a specialized chip pioneered by Google, with reports suggesting companies like Meta and Anthropic are looking to spend billions on Google’s tensor processing units.”
What’s Next for AlphaFold: A Conversation With a Google DeepMind Nobel LaureateWill Douglas Heaven | MIT Technology Review ($)
“It was five years ago this week that AlphaFold 2’s debut took scientists by surprise. Now that the hype has died down, what impact has AlphaFold really had? How are scientists using it? And what’s next? I talked to Jumper (as well as a few other scientists) to find out.”
A Humanoid Robot-Shaped Bubble Is Forming, China WarnsRobert Hart | The Verge
“Speaking at a press briefing, National Development and Reform Commission spokesperson Li Chao said China’s humanoid robotics industry needs to balance ‘the speed of growth against the risk of bubbles.’ Investment has been pouring into the sector despite there being few proven use cases for the bots, Li said, risking a flood of ‘highly similar’ models as funding for research and development shrinks.”
Supermassive Dark Matter Stars May Be Lurking in the Early UniverseLeah Crane | New Scientist ($)
“We may have seen the first hints of strange stars powered by dark matter. These so-called dark stars could explain several of the most mysterious objects in the universe, while also giving us hints about the true nature of dark matter itself.”
Crypto Winter Will Be Different This TimeKen Brown | The Information ($)
“[Stablecoins’] broader use also creates more ways for a stablecoin crisis to emerge and spread across the globe. It is here that the links to the traditional financial system matter. If investors dump their stablecoins, as they did with Circle, the companies sell the assets that back them, potentially causing turmoil in Treasurys, money markets, and the like.”
MIT Study Finds AI Is Already Capable of Replacing 11.7% of US WorkersGrace Snelling | Fast Company
“In an interview with CNBC, Prasanna Balaprakash, ORNL director and co-leader of the research, described [the Iceberg Index model] as a ‘digital twin for the US labor market.’ Using that base of data, the index analyzes to what extent digital AI tools can already perform certain technical and cognitive tasks, and then produces an estimate of what AI exposure in each area looks like.”
AI Isn’t Just Automating Jobs. It’s Creating New Layers of Human WorkEnrique Dans | Fast Company
“When an AI drafts a report, someone still has to verify its claims (please, do not forget this!), check for bias, and rewrite the parts that don’t sound right. When an agent summarizes a meeting, someone has to decide what actually matters. Automation doesn’t erase labor; it just moves it upstream, from execution to supervision.”
The First Large-Scale Cyberattack by AINury Turkel | The Wall Street Journal ($)
“A state-backed threat group, likely Chinese, crossed a threshold in September that cybersecurity experts have warned about for years. According to a report by Anthropic, attackers manipulated its AI system, Claude Code, to conduct what appears to be the first large-scale espionage operation executed primarily by artificial intelligence.”
The post This Week’s Awesome Tech Stories From Around the Web (Through November 29) appeared first on SingularityHub.
2025-11-28 23:00:00
Behavior can be deceptive. What matters for consciousness is not what you do, but how you do it.
You might think a honey bee foraging in your garden and a browser window running ChatGPT have nothing in common. But recent scientific research has been seriously considering the possibility that either, or both, might be conscious.
There are many different ways of studying consciousness. One of the most common is to measure how an animal—or artificial intelligence—acts.
But two new papers on the possibility of consciousness in animals and AI suggest new theories for how to test this—one that strikes a middle ground between sensationalism and knee-jerk skepticism about whether humans are the only conscious beings on Earth.
Questions around consciousness have long sparked fierce debate.
That’s in part because conscious beings might matter morally in a way that unconscious things don’t. Expanding the sphere of consciousness means expanding our ethical horizons. Even if we can’t be sure something is conscious, we might err on the side of caution by assuming it is—what philosopher Jonathan Birch calls the precautionary principle for sentience.
The recent trend has been one of expansion.
For example, in April 2024 a group of 40 scientists at a conference in New York proposed the New York Declaration on Animal Consciousness. Subsequently signed by over 500 scientists and philosophers, this declaration says consciousness is realistically possible in all vertebrates (including reptiles, amphibians and fishes) as well as many invertebrates, including cephalopods (octopus and squid), crustaceans (crabs and lobsters) and insects.
In parallel with this, the incredible rise of large language models, such as ChatGPT, has raised the serious possibility that machines may be conscious.
Five years ago, a seemingly ironclad test of whether something was conscious was to see if you could have a conversation with it. Philosopher Susan Schneider suggested if we had an AI that convincingly mused on the metaphysics of consciousness, it may well be conscious.
By those standards, today we would be surrounded by conscious machines. Many have gone so far as to apply the precautionary principle here too: the burgeoning field of AI welfare is devoted to figuring out if and when we must care about machines.
Yet all of these arguments depend, in large part, on surface-level behavior. But that behavior can be deceptive. What matters for consciousness is not what you do, but how you do it.
A new paper in Trends in Cognitive Sciences that one of us (Colin Klein) coauthored, drawing on previous work, looks to the machinery rather than the behavior of AI.
It also draws on the cognitive science tradition to identify a plausible list of indicators of consciousness based on the structure of information processing. This means one can draw up a useful list of indicators of consciousness without having to agree on which of the current cognitive theories of consciousness is correct.
Some indicators (such as the need to resolve trade-offs between competing goals in contextually appropriate ways) are shared by many theories. Most other indicators (such as the presence of informational feedback) are only required by one theory but indicative in others.
Importantly, the useful indicators are all structural. They all have to do with how brains and computers process and combine information.
The verdict? No existing AI system (including ChatGPT) is conscious. The appearance of consciousness in large language models is not achieved in a way that is sufficiently similar to us to warrant attribution of conscious states.
Yet at the same time, there is no bar to AI systems—perhaps ones with a very different architecture to today’s systems—becoming conscious.
The lesson? It’s possible for AI to behave as if conscious without being conscious.
Biologists are also turning to mechanisms—how brains work—to recognize consciousness in non-human animals.
In a new paper in Philosophical Transactions B, we propose a neural model for minimal consciousness in insects. This is a model that abstracts away from anatomical detail to focus on the core computations done by simple brains.
Our key insight is to identify the kind of computation our brains perform that gives rise to experience.
This computation solves ancient problems from our evolutionary history that arise from having a mobile, complex body with many senses and conflicting needs.
Importantly, we don’t identify the computation itself—there is science yet to be done. But we show that if you could identify it, you’d have a level playing field to compare humans, invertebrates, and computers.
The problem of consciousness in animals and in computers appear to pull in different directions.
For animals, the question is often how to interpret whether ambiguous behavior (like a crab tending its wounds) indicates consciousness.
For computers, we have to decide whether apparently unambiguous behavior (a chatbot musing with you on the purpose of existence) is a true indicator of consciousness or mere roleplay.
Yet as the fields of neuroscience and AI progress, both are converging on the same lesson: when making judgement about whether something is consciousness, how it works is proving more informative than what it does.
This article is republished from The Conversation under a Creative Commons license. Read the original article.
The post Are Animals and AI Conscious? Scientists Devise New Theories for How to Test This appeared first on SingularityHub.
2025-11-27 23:00:00
A range of CRISPR gene therapies are taking aim at chronically high cholesterol, reducing the risk of heart disease.
The gene editor CRISPR is tackling fatty molecules in the body that contribute to one of the world’s top killers: cardiovascular disease.
At the American Heart Association Scientific Sessions 2025 (AHA 2025) this month, Scribe Therapeutics, a startup based in Alameda, California, presented three CRISPR formulations that slashed dangerously high lipid levels in lab-grown cells, mice, and monkeys.
With a single injection, their flagship formulation lowered “bad cholesterol” levels in primates for over 515 days. The treatment used a type of genetic manipulation called epigenome editing that doesn’t directly change the genetic code, potentially reducing side effects.
Two other CRISPR formulations targeted lipoprotein(a) and triglycerides, both fatty substances that form clumps inside blood vessels when at high levels. An injection in mice slashed the molecules by over 95 percent in early trials.
The therapies join other emerging efforts using CRISPR to tackle cardiovascular disease. If the results translate to humans, a daily pill—often taken for decades—may become a thing of the past.
“These results demonstrate that comprehensive engineering of CRISPR technologies can produce medicines with markedly improved safety and performance, surpassing the limitations of early Cas9-based systems,” Benjamin Oakes, cofounder and CEO of Scribe, said in a press release.
High cholesterol haunts millions of Americans. A silent killer, the fatty molecules clog up blood vessels and raise the risk of heart attack, vascular disease, and stroke. Physicians recommend daily statins and dietary changes to manage cholesterol levels, but the regime is hard to follow—especially for years or decades.
Cholesterol comes in multiple forms. Some of these protect the heart and blood vessels. Others lead to clogged arteries. LDL, or low-density lipoprotein, normally transports molecules from the liver to the body’s cells to maintain essential functions, such as building membranes, producing hormones, and creating vitamin D. Too much LDL, however, leads to a buildup of plaques that harden blood vessels and narrow their diameter. This means the heart must work harder to pump blood through the body.
After years of research, scientists identified a gene called PCSK9 that, if overactive, increases the levels of LDL circulating in the blood. FDA-approved drugs that inhibit the PCSK9 protein show promise for lowering cholesterol. But inhibiting the gene itself could offer a longer-term solution.
There have been early successes. In 2023, a small clinical trial in people genetically prone to dangerously high levels of cholesterol found a single infusion of a precise gene editor decreased artery-clogging fat by almost half. Participants had a single mutated DNA letter in the PCSK9 gene that caused their LDL levels to skyrocket. Using base editing—a version of CRISPR—the team engineered a therapy to correct the genetic typo.
A similar strategy could also benefit other populations with high cholesterol. However, base editing permanently alters the genome and could trigger unexpected DNA changes.
Enter epigenetic editors. Rather than directly altering DNA letters, this technology targets the molecular machinery that switches genes on or off. Because epigenetic editors don’t directly change the genetic code, the approach could potentially be safer than gene editing.
Last year, one team employed designer molecules called zinc-finger proteins, a favorite gene-editing tool predating CRISPR, to shut down PCSK9 without changing the gene itself. A single injection slashed cholesterol levels in mice and kept them low for nearly a year—roughly half the mice’s lifespan.
AHA 2025 built on those results.
Scribe developed an epigenetic silencer to suppress PCSK9 using CRISPR-CasX. Like the original version, CRISPR-Cas9, CRISPR-CasX has a guide RNA that tethers CasX—a tiny scissor enzyme—to genes involved in regulating PCSK9 activity and shuts them down.
In monkeys, a single infusion of the treatment slashed LDL levels up to 68 percent. Unlike DNA edits, epigenetic modifications are often lost when cells divide, meaning the drug could lose efficacy over time, especially in rapidly regenerating organs like the liver. But the monkey’s LDL levels remained low for over 515 days without otherwise stressing their livers. Also, the drug didn’t notably change the activity of other genes in cultured human liver cells, suggesting it’s precise.
The data strengthens “the case for a new class of durable epigenetic medicines for large patient populations,” wrote the company in a press release.
PCSK9 isn’t the only gene involved in heart disease. CRISPR Therapeutics, headquartered in Switzerland, worked with the Cleveland Clinic Foundation to find another gene related to high cholesterol levels: ANGPTL3. Studies show people born with dysfunctional versions of the gene naturally have lower LDL levels and risk of heart disease.
The team used CRISPR-Cas9 to disable the gene and recruited 15 people with various blood lipid diseases to test the treatment’s safety profile. Two weeks after a shot, participants’ ANGPTL3 protein and LDL levels dropped significantly and remained low for at least 60 days. Results from the trial, also presented at AHA 2015, found that the treatment was well tolerated overall.
“This is really unprecedented,” said author Luke J. Laffin in a press briefing. “If confirmed in larger trials, this one-and-done approach could transform care for people with lifelong lipid disorders and dramatically reduce cardiovascular risk.”
Artery-blocking lipids beyond LDL are now also in CRISPR’s crosshairs.
Lipoprotein(a) is a mysterious nanoball of fat that’s somewhat similar to LDL in structure but with a more complex mix of components. The substance deposits cholesterol as it roams blood vessels—including smaller ones involved in healing and regeneration. An estimated 30 percent of people worldwide have abnormally high levels of lipoprotein(a). This is mainly due to genetic risks and is hard to reverse with dietary changes or medication.
Another CRISPR-based technology is showing promise here. At the conference, Scribe said its in-house CasXE gene editor inactivates a gene that makes Lp(a) in liver cells. In mice, a single injection slashed levels of the fatty balls by up to 95 percent, with no detectable off-target editing.
Finally, the company showcased a different CasXE gene editor that kneecaps a gene associated with lipid production. Like other genetic targets, people with naturally lower levels of the gene APOC3 have low levels of blood lipids and lower risk of heart disease. One shot edited over 75 percent of all liver cells in monkeys and almost completely reversed high blood lipid levels in mice.
These are all preliminary results, but they could lead to a quantum shift in managing a global chronic disease with a single shot instead of daily pills.
The post CRISPR Slashes ‘Bad Cholesterol’ Levels by 95 Percent in Early Results appeared first on SingularityHub.
2025-11-26 00:17:34
The tiny chips hitch a ride on immune cells to target inflammation in the brain. Scientists hope to kick off clinical trials within three years.
From restoring movement and speech in people with paralysis to fighting depression, brain implants have fundamentally changed lives.
But inserting implants, however small or nimble, requires risky open-brain surgery. Pain, healing time, and potential infections aside, the risk limits the technology to only a handful of people.
Now, scientists at MIT Media Lab and collaborators hope to bring brain implants to the masses. They’ve created a tiny electronic chip powered by near-infrared light that can generate small electrical zaps. After linking with a type of immune cell to form bio-electronic hybrid chips, a single injection into the veins of mice shuttled the devices into their brains—no surgery required.
It sounds like science fiction, but the injected chips easily navigated the brain’s delicate and elaborate vessels to zero in on an inflamed site, where the microchip reliably delivered electrical pulses on demand. The chips happily cohabitated with surrounding neurons without changing the cells’ health or behavior.
“Our cell-electronics hybrid fuses the versatility of electronics with the biological transport and biochemical sensing prowess of living cells,” said study author Deblina Sarkar in a press release.
The strategy, which the researchers call circulatronics, could radically change brain stimulation. Targeted electrical zaps have shown early promise for treatment of a variety of brain diseases, such as Alzheimer’s, depression, and brain tumors.
And because the devices can be engineered to dissolve after a certain amount of time, they could potentially collect neural signals from healthy people, providing an unprecedented look into our brain’s inner workings.
Today’s brain implants are relatively bulky and struggle to reach deep into the brain. Most use batteries, either directly inside the device or in a battery pack affixed to the skull.
An ideal implant would be self-powered, controllable, and small enough to move through the smallest nooks and crannies of the brain and its vessels. A previous device, about the size of a grain of rice, used magnetic energy for power and generated electrical zaps in rodents while they actively roamed around. But because the device was controlled by magnetic fields, the setup required large and expensive hardware. Magnetic particles also tend to move in straight lines. This makes them terrible at navigating our brains serpentine vessels.
Near-infrared light offers an alternative to magnetic control. The wavelength easily penetrates the skull and brain with minimal scattering, suggesting it could control devices deep in the brain. Earlier this month, a team engineered an infrared-powered implant smaller than a grain of salt that could record from or stimulate neurons in mice. Although the device still required minimal surgery to implant, it reliably captured brain signals for a year, roughly half a mouse’s lifespan.
Infrared light has long been on Sarkar’s radar for an injectable brain implant. For six years, her team worked to solve multiple difficult roadblocks, eventually landing on circulatronics.
The team first had to make a chip so small it could easily flow through blood vessels without damaging them. The team turned to photovoltaic components that convert light into electricity, similar to the way solar panels work.
The chips are made of organic semiconductors that are biocompatible and flexible. This makes them suitable for navigation of our squishy bodies. Each one is like a tiny, light-powered battery sandwich, with a positive and negative metallic layer and an organic polymer inner filling.
Roughly 10 microns in diameter and smaller than a cell, these chips can be manufactured en masse with the same technology used to make computer chips. In tests with molds simulating the brain, the chips reliably generated electrical currents.
Then there was the problem of getting the chips to their target. The brain is protected by a wall of cells called the blood-brain barrier. The barrier is extremely selective of what molecules, proteins, and other materials can enter. Electronics, no matter how small, don’t make the cut. Some studies have tried to deliberately pry open the blood-brain barrier, but even a brief opening invites pathogens and other dangerous molecules inside.
The team’s solution was a cellular Trojan horse. When the brain experiences inflammation, the blood-brain barrier admits immune cells called monocytes. These cells roam the bloodstream equipped with chemical beacons to hunt down inflammatory sites. In theory, microchips could catch a ride on these cells through the blood-brain barrier without forcing it open.
To link monocytes to their tiny chip, the team used a Nobel Prize-winning technology called click chemistry. Think of it as Velcro. The researchers altered the surfaces of the monocytes in such a way that they formed Velcro-like “loops.” Then they added chemical “hooks” to the chips. When these components met, they clicked into place—but were still easily detachable—to form the final implant.
“The living cells camouflage the electronics so that they aren’t attacked by the body’s immune system, and they can travel seamlessly through the bloodstream. This also enables them to squeeze through the intact blood-brain barrier without the need to invasively open it,” said Sarkar.
To test their hybrid implants, the team tagged them with glow-in-the-dark trackers and injected them into the veins of mice. The critters had been given a chemical that triggered inflammation at a specific site deep in their brains.
Within 72 hours, the hybrid chips self-implanted into the inflamed area, whereas electronics lacking a cellular partner were barred from the brain. On average, around 14,000 hybrid implants latched onto the brain.
The devices worked as expected. After receiving pulses of near-infrared light for 20 minutes, neurons in the implanted region spiked with electrical activity at a magnitude similar to spikes trigged by current brain implants. Neighboring neurons were undisturbed.
The hybrid implants didn’t seem to affect the brain’s activity. Animals with the implant roamed around as usual. They showed no sign of changes to mood, memory, or other cognitive functions, happily sipping water and maintaining body weight for six months. Despite circulating in the blood after injection, the hybrid implants had no observable impact on other organs.
Although this study focused on brain inflammation, a similar strategy could be used to shuttle brain stimulation chips into stroke sites to aid rehabilitation. The system is relatively plug-and-play. Swapping monocytes for other cell types, such as T cells or neural stem cells, could allow them to act like cellular taxis for a wide range of other diseases.
The team hopes to kick off clinical trials of the technology within three years through MIT spinoff company, Cahira Technologies.
“This is a platform technology and may be employed to treat multiple brain diseases and mental illnesses,” said Sarkar. “Also, this technology is not just confined to the brain but could also be extended to other parts of the body in future.”
The post These Brain Implants Are Smaller Than Cells and Can Be Injected Into Veins appeared first on SingularityHub.
2025-11-25 08:43:40
Called 3D necroprinting, the sustainable system can print extremely intricate structures at high resolution.
Mosquitos are perhaps one of the most universally loathed creatures. Not only are their bites itchy and annoying, they carry diseases that kill nearly 600,000 people worldwide—making them the deadliest animal.
Yet they’ve thrived for millions of years, partly due to the female’s efficient “stinger.” Called a proboscis, the organ’s stiffness allows it to penetrate the skin and into the bloodstream with high precision, but its tiny size and structure don’t tip off the host until it’s too late.
These advantages caught Changhong Cao’s eye at McGill University. Inspired by mosquitos, the bioengineer and his team developed a high-resolution 3D printer using a mosquito proboscis as the nozzle. Called necroprinting, the system prints lines half the width of commercially available printers. In tests, it reliably completed multiple complex 3D structures in bioink, including honeycombs, a maple leaf, and a waffle-like housing encapsulating cancer and red blood cells.
“Repurposing dispensing structures from uninfected, laboratory grown, deceased organisms represents a new avenue for engineering applications, which not only reduces the cost of high-resolution dispense tip production but also minimizes environmental impact,” wrote the authors.
Engineers have long tapped Mother Nature for inspiration.
Early successes relied on mimicry, including self-cleaning surfaces inspired by lotus leaves or Velcro’s famous hook-and-loop structure derived from burdock burrs. More recent innovations combine soft, flexible biomaterials and living cells with plastics to form biohybrid robots capable of sensing, healing, and adapting to environments.
Another trend, perhaps more macabre, takes advantage of the complexity of animal anatomy. Mud eels, Madagascar hissing cockroaches, and beetle legs have been used to create biohybrid devices to monitor medical conditions and the environment. Necrobots repurpose spider legs into microgrippers that allow the legs to expand when activated and reverse to their natural state in a claw-like motion. The grippers can grasp randomly shaped objects up to 130 percent of their own weight, offering a low-cost, efficient, and biodegradable alternative to conventional grippers.
While biohybrid systems have mainly focused on robotics and sensing, Cao’s team had a different idea: Using animal materials in the manufacturing process, rather than the final product.
Nozzles were a favorable choice. For one, they’re widely used in 3D printing and in labs. Similar liquid-dispensing tips are currently made of nonbiodegradable materials—such as metals and plastics—with the US alone churning through over four billion annually.
They’re also costly, especially for high-resolution tips. The finest commercially available metal printer tips have an inner diameter of around 35 micrometers—roughly the size of a single human skin cell. A hefty price tag of over $80 per tip limits the technology’s use.
Cao’s team started their search for a natural printer head with a vast survey of animal appendages.
Among these were scorpion stingers, snake fangs, harpoons from cone snails, and claws from a variety of deadly bugs. Each had a unique shape, length, and inner diameter optimized for the animal—but not necessarily for a printer nozzle.
An ideal nozzle should be straight like a needle, with relatively high stiffness to keep its shape as fluid flows. A small inner diameter is also crucial for high printing resolution, with a length that’s easy to manipulate but not too long, as this leads to pressure buildup and failure.
In their search, the female mosquito proboscis stood out. Its biopolymer core helps maintain a straight structure, similar to a microneedle, as liquids flow through. The organ also boasts a tiny diameter of just 20 micrometers. It’s smaller than commercially available tips and has a stiffness similar to common plastics.
The mosquito proboscis previously inspired microneedles used in biopsies to diagnose cancer with minimal trauma to nearby tissues. Those needles use human-made materials. The new study used the actual organ itself from lab-farmed mosquitos into their 3D printing setup.
To harvest each proboscis, the team bathed frozen, lab-raised female mosquitoes in alcohol to sterilize them before removing the organ. They then designed a custom adapter to connect the proboscis to a metal tip attached to a mechanical extruder, which regulates the flow of fluids.
Resin seals the gap between the commercial and biological dispense tips to prevent leakage. The custom 3D bioprinter has a vertical arm that lifts the nozzle up and down and a horizontally moving “stage” at the bottom that acts as the printing canvas.
In tests, the team found the necroprinter could generally handle commercial inks used for bioprinting, although the proboscis tore if the liquid ran through too quickly. Similar to a tiny straw, the system also failed if the ink clogged at the bottom and ruptured that end. Balancing the speed of ink released from the nozzle and the speed of nozzle movement took calibration.
Imbalance of those forces generated a pressure buildup, “ultimately leading to either gushing of ink or catastrophic rupture of the mosquito proboscis,” the team wrote.
But once the parameters were dialed in, the necroprinter performed accurately and predictably. It readily printed lines roughly 20 micrometers in width, outperforming state-of-the-art nozzles. It also managed to precisely print more complex shapes such as honeycombs and maple leaves.
The microscopic structures surpassed the resolution capabilities of standard metal and plastic dispense tips, wrote the authors.
A third demo used a bioink containing cancer or red blood cells. The necroprinter generated structures laden with cells, which remained alive and healthy. Finally, the printhead showed promise for high-resolution drug delivery. Loaded with hydrogel, it deposited the material into pig skin at extremely low volumes that mimicked therapeutic delivery.
Compared to engineered 3D printheads, a mosquito proboscis is highly consistent in inner diameter and wall thickness. It’s also very affordable. According to the team, it costs just two cents to raise a single mosquito, and assembling a necroprinting dispense tip is less than a dollar.
However, because the tips contain biological tissue they may not be as long-lasting as plastic components. Initial tests found they last for about nine days on the counter and at least a year stored in a freezer. The nozzles also operate in temperatures comfortable for mosquitoes (roughly 20 to 30 degrees Celsius or 70 to 85 degrees Fahrenheit), but extreme shifts can cause catastrophic failure. The team is now mapping temperature boundaries.
Despite potential roadblocks, the system shows the promise of integrating biological material into advanced manufacturing.
The post Super Precise 3D Printer Uses a Mosquito’s Needle-Like Mouth as a Nozzle appeared first on SingularityHub.
2025-11-22 23:00:00
There Is Only One AI Company. Welcome to the BlobSteven Levy | Wired ($)
“Even the most panicked Cassandra of a decade ago likely didn’t imagine that advanced AI would be controlled by a single, interlocking, money-seeking behemoth. …This rococo collection of partnerships, mergers, funding arrangements, government initiatives, and strategic investments links the fate of virtually every big player in the AI-o-sphere. I call this entity the Blob.”
Europe Is Scaling Back Its Landmark Privacy and AI LawsRobert Hart | The Verge
“Under intense pressure from industry and the US government, Brussels is stripping protections from its flagship General Data Protection Regulation (GDPR)—including simplifying its infamous cookie permission pop-ups—and relaxing or delaying landmark AI rules in an effort to cut red tape and revive sluggish economic growth.”
Google DeepMind Hires Former CTO of Boston Dynamics as the Company Pushes Deeper Into RoboticsWill Knight | Wired ($)
“The hire is a key part of DeepMind CEO Demis Hassabis’ vision for Gemini to become a sort of robot operating system, similar to how Google supplies its Android software to an array of smartphone manufacturers. ‘We want to build an AI system, a Gemini base, that can work almost out-of-the-box, across any body configuration,’ Hassabis said in an interview with Wired.'”
Pfizer’s mRNA Flu Shot Outperforms Standard Flu Vaccine in Late-Stage TrialBerkeley Lovelace Jr. | NBC News
“The Phase 3 trial found Pfizer’s mRNA shot cut flu-like illness by 34.5% compared with a standard flu shot. …Developing an mRNA shot is typically faster [than a traditional flu vaccine], which could allow those decisions [about what strains to target] to be made later in the year—and give scientists more flexibility to pivot if the circulating strain changes.”
AI Race Cars Are Catching Up to Human DriversRachyl Jones | Semafor
“Former Formula 1 driver Daniil Kvyat drove against an AI-powered race car in Abu Dhabi, where he clocked a faster time but failed to catch up with the autonomous vehicle’s head start. Only 1.6 seconds separated the best laps of the human and the vehicle, compared to last year’s 10-second gap, indicating significant performance improvements in the AI.”
It’s Too Soon to Call an End to the AI BoomKen Brown | The Information ($)
“I’ve spent the past few weeks talking to the bankers and investors leading the financing of AI. I’m convinced a crack in the market isn’t coming anytime soon. Investor demand is very strong, and it’s too soon for any real problems in the financing machine to show up.”
Hugging Face CEO Says We’re in an ‘LLM Bubble,’ Not an AI BubbleSarah Perez | TechCrunch
“‘I think all the attention, all the focus, all the money, is concentrated into this idea that you can build one model through a bunch of compute and that is going to solve all problems for all companies and all people,’ said Delangue. ‘I think the reality is that you’ll see in the next few months, next few years, kind of like a multiplicity of models that are more customized, specialized, that are going to solve different problems.'”
New Gene-Editing Strategy Could Help Development of Treatments for Rare DiseasesPam Belluck and Carl Zimmer | The New York Times ($)
“A study published on Wednesday outlines a new approach that could make the process more efficient and less costly. Writing in the journal Nature, researchers presented a path toward a gene-editing strategy that could eventually be standardized for many different rare diseases, instead of personalized edits for each one.”
Waymo Enters 3 More Cities: Minneapolis, New Orleans, and TampaSean O’Kane | TechCrunch
“In 2026, Waymo [also] plans to expand to Dallas, Denver, Detroit, Houston, Las Vegas, Miami, Nashville, Orlando, San Antonio, San Diego, Seattle, and Washington, DC. It’s also testing in New York City, and plans to offer commercial rides internationally starting with London and Tokyo.”
Blue Origin Revealed Some Massively Cool Plans for Its New Glenn RocketEric Berger | Ars Technica
“One week after the successful second launch of its large New Glenn booster, Blue Origin revealed a roadmap on Thursday for upgrades to the rocket, including a new variant with more main engines and a super-heavy lift capability.”
What Google Has That OpenAI Doesn’tMartin Peers | The Information ($)
“All this points up a reality that should have been obvious. While we in the news media breathlessly report on every step Sam Altman takes to make OpenAI a vertically integrated AI giant, Google is already there. “
We Finally Know the Birthplace of the Mars-Sized Rock That Spawned Our MoonMargherita Bassi | Gizmodo
“In a study published today in the journal Science, researchers investigated the isotopic fingerprints—the ratio of isotopes, or versions, of elements in a material—of iron in rocks from the moon, Earth, and meteorites (meteoroids that reach the ground). Their results bolster the theory that the impactor was born in the inner solar system and closer to the sun than where Earth originated.”
We Can Now Track Individual Monarch Butterflies. It’s a Revelation.Dan Fagin | The New York Times ($)
“The breakthrough is the result of a tiny solar-powered radio tag that weighs just 60 milligrams and sells for $200. Researchers have tagged more than 400 monarchs this year and are now following their journeys on a cellphone app created by the New Jersey-based company that makes the tags, Cellular Tracking Technologies.”
The post This Week’s Awesome Tech Stories From Around the Web (Through November 22) appeared first on SingularityHub.
