MAROKO133 Breaking ai: Oxford Researcher Warns That AI Is Heading for a Hindenburg-Style D

📌 MAROKO133 Breaking ai: Oxford Researcher Warns That AI Is Heading for a Hindenbu

Is the AI bubble going to burst? Will it cause the economy to go up in flames? Both analogies may be apt if you’re to believe one leading expert’s warning that the industry may be heading for a Hindenburg-style disaster.

“The Hindenburg disaster destroyed global interest in airships; it was a dead technology from that point on, and a similar moment is a real risk for AI,” Michael Wooldridge, a professor of AI at Oxford University, told The Guardian.

It may be hard to believe now, but before the German airship crashed in 1937, ponderously large dirigibles once seemed to represent the future of globe-spanning transportation, in an era when commercial airplanes, if you’ll permit the pun, hadn’t really taken off yet. And the Hindenburg, the largest airship in the world at the time, was the industry’s crowning achievement — as well as a propaganda vehicle for Nazi Germany.

At over 800 feet long, it wasn’t far off the length of the Titanic — another colossus whose name became synonymous with disaster — and regularly ferried dozens of passengers on Trans-Atlantic trips. All those ambitions were vaporized, however, when the ship suddenly burst into flames as it attempted a landing in New Jersey. The horrific fireball was attributed to a critical flaw: the hundreds of thousands of pounds of hydrogen it was filled with were ignited by an unfortunate spark. 

The inferno was filmed, photographed and broadcasted around the world in a media frenzy that sealed the airship industry’s future. Could AI, with its over a trillion dollars of investment, head the same way? It’s not unthinkable. 

“It’s the classic technology scenario,” Wooldridge told the newspaper. “You’ve got a technology that’s very, very promising, but not as rigorously tested as you would like it to be, and the commercial pressure behind it is unbearable.”

Perhaps AI could be responsible for a catastrophic spectacle, such as a deadly software update for self-driving cars, or a bad AI-driven decision collapsing a major company, Wooldridge suggests. But his main concern are the glaring safety flaws still present in AI chatbots, despite them being widely deployed. On top of having pitifully weak guardrails and being wildly unpredictable, AI chatbots are designed to affect human-like personas and, to keep users engaged, be sycophantic.

Together, these can encourage a user’s negative thoughts and lead them down mental health spirals fraught with delusions and even full-blown breaks with reality. These episodes of so-called AI psychosis have resulted in stalking, suicide and murder. AI’s ticking time bomb isn’t a payload of combustible hydrogen, but millions of potentially psychosis-inducing conversations. OpenAI alone admitted that ChatGPT that more than half a million people were having conversations that showed signs of psychosis every week.

“Companies want to present AIs in a very human-like way, but I think that is a very dangerous path to take,” Wooldridge told The Guardian. “We need to understand that these are just glorified spreadsheets, they are tools and nothing more than that.”

If AI has a place for us in the future, it should be as cold, impartial assistants — not cloying friends that pretend to have all the answers. A shining example of this, according to Wooldridge, is how in an early episode of “Star Trek,” the Enterprise’s computer says it has “insufficient data” to answer a question (and in a voice that is robotic, not personable.)

“That’s not what we get. We get an overconfident AI that says: yes, here’s the answer,” he told The Guardian. “Maybe we need AIs to talk to us in the voice of the ‘Star Trek’ computer. You would never believe it was a human being.”

More on AI: It Turns Out That Constantly Telling Workers They’re About to Be Replaced by AI Has Grim Psychological Effects

The post Oxford Researcher Warns That AI Is Heading for a Hindenburg-Style Disaster appeared first on Futurism.

🔗 Sumber: futurism.com

📌 MAROKO133 Eksklusif ai: Shape-conforming 3D bioelectronic mesh maps 91% of lab-g

Scientists have built a soft, three-dimensional electronic device that can wrap around lab-grown human mini brains and record activity across nearly their entire surface.

The system captures electrical signals from 91 percent of a neural organoid, overcoming a major limitation in how researchers study these brain-like tissues.

Human neural organoids, grown from stem cells, are increasingly used to model brain development and disease. They form interconnected neural circuits and produce coordinated electrical rhythms.

But most existing recording tools are flat and rigid, allowing scientists to sample signals from only a handful of locations.

That mismatch has prevented researchers from observing how activity spreads across the entire network. Without full coverage, important synchronized patterns and large-scale communication between neurons can go undetected.

The new device, developed by researchers at Northwestern University and Shirley Ryan AbilityLab, changes that by conforming to the organoid’s curved shape. It begins as a flat, flexible lattice and transforms into a 3D framework that gently envelops the spherical tissue.

From fragments to full networks

“Human stem cell-derived organoids have become a major focus of biomedical research because they enable patient-specific studies of how tissues respond to drugs and emerging therapies,” said John A. Rogers, who led the device development.

“A key missing component is hardware technology that can interrogate, stimulate and manipulate these tiny analogs to organs in the human body.”

The mesh-like scaffold contains up to 240 individually addressable microelectrodes. Each electrode measures just 10 microns in diameter, roughly the size of a single cell. The porous design allows oxygen and nutrients to flow through while maintaining stable electrical contact.

“Human neural organoids are living 3D tissues that contain active neural circuits communicating through electrical signals,” said Dr. Colin Franz, who led the organoid development.

“By creating soft, shape-matched electronics that conform to the organoid’s geometry, we can now record from and stimulate hundreds of locations across its surface at once. This allows us to study neural activity at the level of whole networks rather than isolated signals.”

When the team tested systems with only eight or 32 electrodes, recordings were limited and localized. With the full 240-channel array, researchers observed synchronized oscillatory waves spanning the organoid.

Because each electrode’s position is mapped in three dimensions, the system can generate detailed activity maps.

Pop-up electronics, real brains

The transformation from flat sheet to 3D structure relies on controlled mechanical buckling, similar to how a pop-up book folds into shape. This approach allows the electronics to match the organoid’s curvature without damaging it.

“The device’s structure needs to support these metabolic processes to sustain the viability of the tissue,” Rogers said.

“Basically, the organoid needs to breathe. The hardware must not significantly constrain or suffocate it.”

Beyond recording signals, the platform can also deliver electrical stimulation. In drug tests, the team observed clear changes in neural activity.

Exposure to 4-aminopyridine increased signaling, while botulinum toxin disrupted coordinated firing. These responses demonstrate that the system can detect meaningful changes in living neural networks.

The researchers also showed that altering the lattice design can shape how organoids grow, producing non-spherical geometries such as cubes and hexagons. That capability could enable modular assembly of different tissue types for future research.

As organoids become more central to neuroscience and drug development, tools that provide near-complete monitoring may help determine whether experimental therapies truly restore coordinated brain function.

The study was published in Nature Biomedical Engineering.

🔗 Sumber: interestingengineering.com