A reflection on ML4PS 2025, where researchers in physics and machine learning grappled with the role AI should play in scientific discovery, and what it would take to move from process acceleration toward deeper scientific insight.

The team at FirstPrinciples has reached the AI Physicist’s first autonomy milestone ahead of schedule, assembling the early loops of a system that will guide its own scientific reasoning from research to hypothesis and beyond.

Large Language Models have changed how we think, work, and do science, but can they truly reason like scientists? At FirstPrinciples, we’re exploring the limits of AI generalization and the promise of specialization through the development of the AI Physicist.

With a deep passion for physics, Shiraz Minwalla investigates the complexities of black holes, grey galaxies, and beyond.

FirstPrinciples presented an early demo of its AI Physicist at the Open Science Fair, held this year at CERN. The event sparked conversations on trust, openness, and the role of AI in research, underscoring how collaboration will shape the future of discovery.

In a research world defined by scale and AI-driven discovery, small colleges are proving impact isn’t measured in dollars. At places like Wellesley and Bowdoin, close mentorship, collaborative problem-solving, and a readiness to experiment with new tools are shaping the next generation of scientists to navigate a rapidly evolving research landscape.

Chain-of-thought reasoning has become a rare interface between human and machine logic. But a new paper warns that the window may be closing.

Physicist Neil Johnson has mapped the exact moment AI can flip from accurate to false, and he says understanding the physics could be the key to safer systems.

More scientists are leaving academia, trading tenure-track hurdles for the speed and flexibility of industry. For physicist Elizabeth Frank, that shift meant moving from mapping Mercury to mining the Moon — swapping publication bottlenecks for the fast, interdisciplinary problem-solving of space startups, and using AI to revive data gathered half a century ago.

At the intersection of artificial intelligence and fundamental physics, the NSF Institute for Artificial Intelligence and Fundamental Interactions (IAIFI) is preparing scientists to learn, think and understand at the deepest levels. In its latest week-long Summer School program, students explored frontier challenges through lectures, tutorials, and collaborative hackathons, testing how AI can shape the future of physics, and how physics can push the boundaries of AI.

At Columbia University, quantum breakthroughs emerge from a culture of shared labs, ideas, and materials. What began as informal collaboration has become a scalable model for cross-disciplinary science that powers advances in programmable quantum systems.

Z.ai's new open-source model and Harmonic's math-native chatbot highlight contrasting strategies for AI reasoning, while a new wave of increasingly specialized benchmarks invites us to rethink how we measure progress.

From prompt injection to physics simulators and open reasoning models, recent news shows that AI isn’t just accelerating science, it’s reshaping how it works. The question now, is will it deepen inquiry, or erode the principles on which credibility in science is built?

A new kind of network is emerging. One that could enable ultra-secure communication, safeguarded by the laws of physics. As the race to build a quantum internet accelerates, nations are vying for digital sovereignty and want to be the first to build the unbreakable internet.

She’s one of the internet’s sharpest scientific voices: equal parts physicist, critic, and communicator. In this candid conversation, Sabine Hossenfelder reflects on AI, the flood of low-impact theory papers, and how a scientific culture ripe for reform could finally be ready for change.

Hidden prompts buried in preprints show how easily large language models (LLMs) can be manipulated, exposing a deep vulnerability in science’s quality-control system. As artificial intelligence (AI) becomes part of the scientific review process, traceability and transparency must become new norms, not afterthoughts.

The Flatiron Institute stands out for its bold model: full-time scientists, open-source tools, and a mission to accelerate discovery through computation. Its success across fields—from astrophysics to neuroscience—reflects the power of that approach.

Large language models have advanced dramatically in recent years, yet when physicists gave them an undergraduate-level test, even the best models were only correct on around one out of every three questions. The new PhysUniBench benchmark exposes how far AI still has to go in mastering fundamental science.

String theory was once hailed as the “theory of everything” — a unified model of nature built on tiny vibrating strings. But after decades of expansion, the field has evolved beyond its namesake, embracing branes, dualities, and abstract geometry. Some physicists now wonder: is it time to rename the theory entirely?

What is AI? From symbolic logic to large-scale neural networks, we're unpacking how today’s systems learn, generate, and reason alongside the the misconceptions and challenges of applying AI in science and society.