Volume 10 Issue 7, July 2026

The latest budget proposal announced by the largest UK science funding agency threatens to severely affect UK astrophysics research. My colleagues and I are fighting back.

We can do more than defend science from a flood of AI-assisted papers. Used well, AI offers a historic opportunity to correct distortions in the publication system, help us publish fewer and better papers, and give scientists back the time to do their best work.

A simulation from sub-photosphere to the corona shows how our Sun forms prominences in a dipped magnetic arcade: through synergy between sporadic injection of chromospheric matter and a siphoned mass supply from thermal instability.

Recent work finds that an influx of microscopic interplanetary meteoric dust particles can provide a plausible explanation for two long standing mysteries of the Venus clouds: the identification of both the lower haze and the unknown UV absorber.

The second international workshop on “AI + Astronomy: Models, Data, Discovery”, held in Lausanne from 31 March to 1 April 2026 convened more than 50 scholars from the EPFL Laboratory of Astrophysics (Versoix, Switzerland) and Zhejiang Lab (Hangzhou, China). The event focused on the transformative impact of artificial intelligence (AI) on data-intensive astronomical research and provided a platform to address the complexities arising from its use.

This workshop examined the rapidly evolving role of language-based artificial intelligence in astronomy, discussing retrieval-augmented generation, agent-based systems, and multimodal models, alongside challenges around evaluation, trust, and responsible deployment.

The IAU’s first fully online symposium brought researchers together from around the globe to discuss technosignatures. Upcoming all-sky surveys, exoplanet missions and tools to search data archives offer new directions for the field.

Stellar occultation observations reveal that the trans-Neptunian object 2002 XV₉₃, which has a radius of only ~250 km, is surrounded by a very tenuous atmosphere. The result shows that even a small distant icy body can host a detectable atmosphere, at least transiently.

For a black hole X-ray binary system, the deflection of its radio jets by the stellar wind has been measured across an entire orbit. By balancing the momentum flux of the jets and the wind, the instantaneous kinetic power of the jet is determined and its value vindicates standard assumptions from cosmological simulations.

The mass spectrum of binary black-hole mergers has been expected to show a ‘mass gap’ above 45 solar masses, consistent with the physics of pair-instability supernovae. An extensive catalogue of gravitational-wave detections reveals a high-spin population above this threshold that probably results from repeated black-hole mergers in dense star clusters, and fits with the existence of the mass gap. The inferred boundary also constrains the carbon-to-oxygen fusion rate during helium burning in massive stars.

Multimessenger astronomy is set to provide a data deluge that will create a domain for physics-informed artificial intelligence to contribute transformative assistance and insight.

TAMBO will measure supra-petaelectronvolt tau neutrinos from the Universe by looking for the characteristic air shower of particles that are produced by the decay of a tau lepton that has interacted with Earth’s crust.

Self-consistent 3D simulations reveal how solar prominences can form and demonstrate dynamic structures that are consistent with observations. Plasma ejected from the surface creates cool structures in the hot solar atmosphere that continue to grow by coronal condensation.

Modelling demonstrates that the Venus’ lower haze originates from cosmic dust. These particles also provide a plausible source for the long-unidentified UV absorber, establishing cosmic dust as a vital component of the planet’s climate system.

Earth’s origin is debated. Meteorites formed in the inner and outer Solar System, but Earth represents an endmember, inner Solar System source alone. Venus and Mercury are predicted to be more extreme, implying a gradient in the inner Solar System.

A stellar occultation campaign reveals a thin atmosphere around the plutino 2002 XV₉₃, showing that even a modest-sized trans-Neptunian object can host an atmosphere, perhaps through cryovolcanism or a recent impact.

ALMA observations of the interstellar comet 3I/ATLAS provide a limit on the deuterium-to-hydrogen ratio in water. This ratio is a sensitive probe of temperature, suggesting that comet 3I formed in an ultracold environment with minimal thermal processing in its home system.

By counting single atoms of ⁶⁰Fe (which have a supernova origin), ²⁴⁴Pu and ²⁴⁷Cm (both formed via the r-process and independent of regular supernovae) in a deep-sea crust, the last actinide-producing r-process event near Earth is found to have occurred >100 Myr ago.

JWST observations of both the dayside and nightside of a hot exoplanet show that horizontal winds transport chemical species faster than the time they need to react chemically, which drives the nightside atmosphere out of chemical equilibrium.

Millisecond pulsar PSR J0435+3233, with a spin-down rate that is 100 times higher than those of other millisecond pulsars, occupies a unique position on the P–Ṗ diagram. This pulsar challenges classical recycling models and constrains spin-up theories.

LHAASO has detected γ-ray emission with a spectrum extending to 2 PeV from the pulsar wind nebula (PWN) powered by PSR J1849-0001, indicating an extreme particle acceleration efficiency and challenging the current particle acceleration theories.

An 18-year high-resolution radio image dataset of Cygnus X-1 has revealed that its radio jets are bent by the stellar wind. Modelling of jet–wind interactions strongly supports the standard assumptions of an accreting black hole in cosmological simulations.

Gravitational-wave data reveal evidence for the pair-instability mass gap and a population of high-mass black hole mergers formed in star clusters. The measurement also constrains the nuclear reaction that regulates carbon–oxygen production in massive stars.

A combined Bayesian model for photometric observations of type Ia supernovae and their host galaxies, together with a scalable neural-simulation-based inference pipeline, can provide constraints on cosmological parameters and host–type Ia supernova connections.