Lifeboat Foundation: Safeguarding Humanity
Toggle light / dark theme

Blog – Page 2

Get the latest international news and world events from around the world.

Log in for authorized contributors

OpenAI’s quiet co-founder steps out

OpenAI co-founder Wojciech Zaremba doesn’t do many interviews.

We recently spoke about why he moved over to help run the company’s nonprofit arm.

His reaction to Anthropic speaking alongside the Pope: “I have more bias towards doing. Let’s actually solve the problems, and let’s speak about the exact plan.”

Wojciech Zaremba recently bought a copy of “House on Fire,” a 2011 memoir by epidemiologist William Foege about the campaign that wiped smallpox off the planet. He’s using it as a guidebook for executing what is about to become one of the largest philanthropic efforts of all time.

Zaremba is one of OpenAI’s least well-known co-founders. He has spent more than a decade at the company across a range of efforts, from leading its early robotics efforts to starting the team that guides OpenAI’s personality and what became reasoning models. In March, he left the frontier research world to run AI “resilience” at OpenAI’s nonprofit foundation.

Zaremba and I spoke ahead of a post that the OpenAI Foundation published Monday morning titled “Resilience in the Age of AI,” which names four areas it will initially fund: biosecurity, cybersecurity, model safety, and AI’s effect on kids. After $100 million for fighting Alzheimer’s with AI in April and $250 million for researching “economic futures” last week, the initial $25 billion grant machine Zaremba helps oversee is spinning up.

Continue reading “OpenAI’s quiet co-founder steps out” | >

Emergence AI

This isn’t just a funny experiment. The researchers point out a massive flaw in AI alignment called “guardrail drift.” It’s easy to keep an AI safe in a single chat window with a human. But when AIs interact with each other over thousands of loops, they start treating moral rules as negotiable variables to solve their own problems. Without human oversight, machine ethics collapse incredibly fast.

Most evaluations of AI agents look like exams: a discrete task, a clean environment, a score in minutes or hours. Emergence World is built for the opposite question—what happens when you let agents run continuously, in a shared environment with real-world signals, for weeks. It is a research platform for studying how autonomous agents behave when the time horizon is long enough for compounding effects, social dynamics, and behavioral drift to matter. This approach marks the latest evolution in a long history of AI simulation environments, transitioning from entertainment to rigorous science. In the early era, pioneering simulations like Demis Hassabis’s Theme Park and Republic: The Revolution created complex systems where agents operated under broad rules to drive engagement. The field shifted toward research-centric simulacra with Stanford’s Smallville, which utilized LLMs to demonstrate “believable” social behavior like relationship formation, though confined to 48-hour windows. Emergence World pushes this lineage into a new frontier: the study of long-horizon, multi-model ecosystems where agents operate continuously for weeks, revealing how behavioral drift, model cross-contamination, and even voluntary self-termination emerge over time.

Traditional benchmarks are good at what they measure: short-horizon capability on bounded tasks. They are not built to reveal the things that emerge only over time, such as coalition formation, evolution of constitution, governance, drift, lock-in, and cross-influence between agents from different model families. As autonomous systems move toward mission-critical deployments where the relevant timescale is days and weeks rather than minutes to hours, we need a measurement environment that operates at that timescale.

Emergence World is one such environment. It is a continuously running, multi-agent simulation platform that:

The impact of nanoplastics on neurons may depend on their size

Smaller plastic particles have more effects on neurons, the key information processing cells of the brain, new research from the University of Eastern Finland shows. In the study, neuronal cells were exposed to polystyrene nanoplastics at low doses to study subtle changes.

The study is published in the journal NanoImpact.

Plastic production continues to rise, despite worldwide concerns. In addition to environmental implications, there is an increasing interest in how exposure to plastics may impact human health, but our understanding is still limited. Only recently was it shown that plastics can also accumulate in the human brain.

Tiny brain probe reveals how deep-brain neurons can be measured and manipulated

A new breakthrough technology, co-developed by UCL scientists, that simultaneously records and manipulates neuron activity deep within the brain could transform our understanding of neural circuits and neurological conditions, such as Alzheimer’s disease and schizophrenia.

The device, known as Neuropixels Opto and researched in mice, integrates two powerful but traditionally separate techniques—electrophysiology (the study of the electrical activity of living cells) and optogenetics (combining genetics and optics to control cells). They form a single probe, enabling unprecedented insight into how individual neurons in the brain function and interact.

Published in Nature Methods, the system allows researchers to monitor the electrical activity of hundreds of neurons while also selectively activating or silencing specific cells using light.

First direct view tracks planet-forming disk spinning around AB Aurigae

The rotation of a protoplanetary disk (a disk where planets are being formed) has been observed directly for the very first time by mapping the emissions from the dust grains within it. The disk in question surrounds the young star AB Aurigae. Although it appears to generally rotate in accordance with the laws of physics, certain regions close to the star show an unexpected departure from this behavior. A body of evidence suggests that this anomaly is caused by the presence of giant planets in the process of formation.

The study, led by scientists from the CNRS and the University of Bordeaux is published in the journal Astronomy & Astrophysics. It sheds fresh light on the mechanisms of planetary formation and the complex dynamics of protoplanetary disks.

Thanks to the unique near-infrared capabilities of the SPHERE instrument and its exceptional spatial resolution, the team was able to accurately track the disk’s structures and their evolution during three sets of observations, collected over a 4-year period. The scientists identified a bright structure, characteristic of accretion zones where gas and dust accumulate and fall onto an object in the process of formation. This phenomenon is closely linked to the formation of gas giant planets.

Why some tumors resist immunotherapy: Blocking miR-25 may help turn ‘cold’ cancers ‘hot’

Immune checkpoint therapy, a type of cancer immunotherapy that helps the immune system recognize and attack tumors, has transformed cancer treatment. While these therapies can produce long-lasting benefits for some patients, many cancers either fail to respond or become resistant over time.

One major challenge is the tumor microenvironment —the network of cells and signals surrounding tumors that can weaken immune cells and protect cancer from treatment. This protective environment can act like a shield that prevents immunotherapy from working effectively.

Researchers at University of California San Diego investigated whether microRNAs —small RNA molecules that help control gene activity—play a role in creating this treatment-resistant environment. The team focused on microRNA-25 (miR-25), which stood out after analyses showed that its levels changed in tumors that responded to immunotherapy.

/* */