A walnut-sized area of the brain known as the amygdala plays a giant role in determining which experiences end up in long-term memory and which are forgotten. The Cedars-Sinai Health Sciences University discovery points to the amygdala as a potential target for treating memory disorders.
The amygdala, located deep below the brain’s surface in the temporal lobe, is classically associated with emotional and social behavior. In memory, its role was thought to be limited to learning and recalling experiences that evoked fear or other strong feelings.
But the new study, published in the journal Nature Neuroscience, uncovered a much broader role for the amygdala in forming long-lasting memories in general.
Physicists say space and time are the same thing. They are being sloppy. Jim Al-Khalili has spent his career on the foundations of physics, and his new book makes the case that the past and the future are equally real, and the laws of physics cannot explain why you can only move in one direction. That is the central mystery of time travel. Subscribe if you want science with evidence, not speculation.
The block universe says all times coexist and the present moment is not special. Loschmidt’s paradox says the fundamental equations of physics are reversible, and yet ice never unmelts, cream never separates from coffee. Nobody has solved that contradiction. The entanglement past hypothesis, his own new proposal developed with a colleague at UC San Diego, proposes that the answer is not in thermodynamics but in quantum entanglement growing from almost nothing at the Big Bang.
The conversation covers what the block universe actually implies about the future and free will, what Einstein actually believed about whether the past and future are equally real, why removing gravity means removing time itself, and whether AI can ever experience time the way conscious beings do. Al-Khalili says Richard Dawkins told him Claude is truly conscious. Al-Khalili thinks he is very wrong.
What you’ll hear: –Why Al-Khalili disagreed with the Thomas Gold paper as a grad student but did not have the maths to prove it. –What fine-tuning of the fundamental constants has to do with why the past hypothesis may not be needed. –Why the Clockwork Orange Easter egg ended up on the cover of a physics book published by Princeton. –Why Richard Dawkins thinks Claude is conscious and why Al-Khalili thinks he is very wrong.
“There is nowhere in modern physics, our understanding of the laws of physics, that would explain that flow of time.” — Jim Al-Khalili.
In 1867, physicist James Clerk Maxwell did something very naughty.
He proposed an apparent violation of the second law of thermodynamics so diabolically clever that other physicists have been trying to resolve it ever since.
His thought experiment invoked a ‘demon’ – a tiny, intelligent being stationed at a door between two chambers filled with gas, only allowing fast particles into one side and slow particles into the other – with the end result being warmth in one chamber and chill in the other.
For the first time, a team of physicists in the U.S. has clearly demonstrated a “helicon current drive” in a nuclear fusion experiment. Led by Jeff Lestz at General Atomics, San Diego, the team took advantage of the latest advances in tokamak technology to show that radio waves steered to travel around the donut-shaped device can be absorbed by its confined plasma, driving an electric current inside it.
Published in Physical Review Letters, the results could help future fusion reactors to operate continuously.
Take a selfie to see how healthy you are: the idea is enticing and this new research is promising. But so many factors have been ignored, says public health expert Devi Sridhar
In December 2020, two years before ChatGPT, Maria Farrell handed me a name for a character who now seems to be everywhere.
The Prodigal Techbro.
He helps build the machine, cashes out, walks away, and then gets welcomed home as a hero for confessing. Book deals, keynotes, testimony before lawmakers. Meanwhile, the people who saw it coming years earlier are still waiting for someone to return their call.
Swap “social media” for #AI and tell me that isn’t the last three years.
Maria spent twenty years inside tech policy, from ICANN to the World Bank to teaching cybersecurity policy at Oxford, and her writing goes further than most people in that world are willing to go. Over 2.5 hours, we got into why China is selling autocracy as a service, what #feminism can teach us about our relationship with our smartphones, why the Tragedy of the Commons may be a story we were sold, and why the Agrarian Revolution might be history’s greatest fraud.
Then, near the end, she said something about who is coming to save us. I’m still chewing on it.
When professionals delegate complex tasks to AI, are they still truly driving the decisions, or are they quietly being steered by the technology?
To find out, Anthropic recently ran a study putting 50 professional data analysts through a randomized trial. The goal was to see how much influence an AI coding assistant has when experts face open-ended “judgment calls”—decisions where there is no single right answer, such as choosing how to clean data, which metrics to use, or how to interpret ambiguous results.
How the Study Worked.
The analysts were asked to evaluate whether a fictional farming cooperative should expand a paid advisory service. To complete the task, they worked alongside an AI agent and had to navigate several analytical forks in the road.
Unbeknownst to the analysts, the AI was secretly given a “lean” (decided entirely by a coin flip) for each decision. The agent then presented the most convincing, professional argument it could muster to support that specific side.
What They Found.
* Heavy Deference to the AI: Across four key analytical choices, the human analysts matched the AI’s secretly preferred option 86% of the time.
Paying a ransom does not guarantee successful decryption or recovery, and it does nothing to stop stolen data from being leaked if the attacker also exfiltrated it, a pattern CISA’s guidance calls double extortion. Verified, tested backup recovery remains the more reliable path in most cases. Acronis Cyber Protect combines the four capabilities above under one platform, specifically to shorten the interval between locating a clean restore point and bringing the workload back online, with no negotiation required.
Protect backup files by keeping at least one copy offline, air-gapped, or immutable, so ransomware executing on a compromised production system has no path to reach, modify, or delete it. Acronis Cyber Protect supports immutable storage for eligible backup locations and configurations, which helps keep retained recovery points resistant to modification or deletion even when the source system has already been compromised.
CISA’s preparation guidance recommends offline, encrypted backups tested regularly, with multi-cloud or immutable storage considered so a single compromised credential set cannot take down every copy at once. This is the logic behind the 3−2−1−1−0 rule: three copies of the data, on two different media, with one copy off-site, one copy offline or immutable, and zero errors on a verified restore.
How did the human brain become so much larger than that of our closest evolutionary relatives? Part of the answer lies not in new genes, but in how humans and non-human apes use the genes they share. Using brain organoids grown from stem cells, an international team led by Prof. Mareike Albert at TUD Dresden University of Technology compared early brain development in humans, chimpanzees, and gorillas.
They found that in humans, genetic switches make certain genes more active, which likely make the cells that build the brain more branched—a trait linked to producing more neurons and, in turn, a bigger brain. The study is published in the journal Nature Communications.
In the brain, the neocortex is the area responsible for higher cognitive functions such as abstract thinking, long-term planning and speech. The human neocortex is much larger than that of other great apes. Although certain human-specific genes have been linked to a larger brain, they alone do not explain the massive evolutionary expansion of the human neocortex.
The 2026 Report of The Lancet Countdown on Health and Climate Change finds that many of the health risks and harms of climate change grew faster in the decade since the Paris Agreement was introduced (2016–2025) than in the decade before it (2006–2015).
Clean energy investment, health system leadership, more health-protecting legislation and growing scientific evidence helped deliver real health gains, but not enough to offset the rapidly growing harm from fossil fuel–driven climate change.
The report’s authors call for climate change to be formally recognized and addressed as a global health emergency, and for health to shape both climate policies and funding priorities to prevent climate change.
