Is It Real or Imagined? Here’s How Your Brain Tells the Difference - WIRED (Full Access)   

Those aren't just lyrics from the Queen song "Bohemian Rhapsody." They're also the questions that the brain must constantly answer while processing streams of visual signals from the eyes and purely mental pictures bubbling out of the imagination. Brain scan studies have repeatedly found that seeing something and imagining it evoke highly similar patterns of neural activity. Yet for most of us, the subjective experiences they produce are very different.

So "why are we not constantly hallucinating?" asked Nadine Dijkstra, a postdoctoral fellow at University College London. A study she led, recently published in Nature Communications, provides an intriguing answer: The brain evaluates the images it is processing against a "reality threshold." If the signal passes the threshold, the brain thinks it's real; if it doesn't, the brain thinks it's imagined.

Such a system works well most of the time because imagined signals are typically weak. But if an imagined signal is strong enough to cross the threshold, the brain takes it for reality.

Continued here

The VC Funding Party Is Over - WIRED (Full Access)   

“It might be the best time for any kind of business in any industry to raise money for all of history, like since the time of the ancient Egyptians,” an excitable Stuart Butterfield, CEO of Slack, told Farhad Manjoo in The New York Times in 2015.

This was no exaggeration. While interest rates remained close to zero, venture capital funds raised more money than ever and exited their investments at some of the highest valuations ever witnessed.

The glory days of VC are over, and if history is any guide, the tech bust should last through 2024 and beyond. In other words, the venture capital bust has only just started.

Continued here

A Mountain of Used Clothes Appeared in Chile’s Desert. Then It Went Up in Flames - WIRED (Full Access)   

This story originally appeared on Grist and was copublished withEl País.It's reproduced here as part of the Climate Desk collaboration. A Spanish-language version can be read here. Reporting was supported by theJoan Konner Program in the Journalism of Ideas.

On the morning of June 12, 2022, Ángela Astudillo, then a law student in her mid-twenties, grabbed her water bottle and hopped into her red Nissan Juke. The cofounder of Dress Desert, or Desierto Vestido, a textile recycling advocacy nonprofit, and the daughter of tree farmers, Astudillo lives in a gated apartment complex in Alto Hospicio, a dusty city at the edge of the Atacama Desert in northern Chile, with her husband, daughter, bunny, and three aquatic turtles.

They headed toward a mountain of sand known as El Paso de la Mula. Less than a mile from her home, squinting into the distance, Astudillo saw a thread of smoke rising from its direction. With her in the lead, the two vehicles caravanned toward the dune, the site of the second-largest clothes pile in the world.

Continued here

A Flaw in Millions of Apple, AMD, and Qualcomm GPUs Could Expose AI Data - WIRED (Full Access)   

As more companies ramp up development of artificial intelligence systems, they are increasingly turning to graphics processing unit (GPU) chips for the computing power they need to run large language models (LLMs) and to crunch data quickly at massive scale. Between video game processing and AI, demand for GPUs has never been higher, and chipmakers are rushing to bolster supply. In new findings released today, though, researchers are highlighting a vulnerability in multiple brands and models of mainstream GPUs—including Apple, Qualcomm, and AMD chips—that could allow an attacker to steal large quantities of data from a GPU’s memory.

The silicon industry has spent years refining the security of central processing units, or CPUs, so they don’t leak data in memory even when they are built to optimize for speed. However, since GPUs were designed for raw graphics processing power, they haven’t been architected to the same degree with data privacy as a priority. As generative AI and other machine learning applications expand the uses of these chips, though, researchers from New York–based security firm Trail of Bits say that vulnerabilities in GPUs are an increasingly urgent concern.

“There is a broader security concern about these GPUs not being as secure as they should be and leaking a significant amount of data,” Heidy Khlaaf, Trail of Bits’ engineering director for AI and machine learning assurance, tells WIRED. “We’re looking at anywhere from 5 megabytes to 180 megabytes. In the CPU world, even a bit is too much to reveal.”

Continued here