Overview
On September 10, 2026, multiple science outlets reported a Nature study: a team used an algorithm called kine to reconstruct 116 radio observations into a continuous video spanning 27 years, targeting the blazar 3C 345 in Hercules. In the video, the bright spots long treated as "shocks" move at speeds comparable to the surrounding plasma — directly conflicting with the core prediction of the shock model. Paper DOI: 10.1038/s41586-026-10988-5; first author is Caltech postdoc Marianna Foschi. A four-decade default explanation no longer holds for this source.
The Problem with VLBI: You Can Photograph, But Not Continuously
Very Long Baseline Interferometry (VLBI) links radio telescopes worldwide into an array with an effective aperture approaching Earth's diameter and milliarcsecond angular resolution. But telescopes record sparse visibility data, not images — imaging is an ill-posed inverse problem that algorithms must fill in. Since 1994, the MOJAVE monitoring program has imaged each observation independently; array configurations, beams, and the source itself change between epochs, so frames share no information. The result: only unresolved bright blobs seen moving roughly, with no view of internal dynamics.
What kine Does: Learning 27 Years as One Continuous Function
kine's core is a neural field: a neural network takes spatial and temporal coordinates as input and outputs radio intensity and polarization at that point. The entire video is learned as one continuous mathematical function, jointly optimized across all 116 observations with information shared across frames. The data come from 116 VLBA observations at 15 GHz between 1995 and 2022, part of the MOJAVE program.
Quantified Gains: ~4× Resolution, ~140× Dynamic Range
| Metric | kine | Traditional CLEAN imaging | |---|---|---| | Mean effective resolution | ~113 microarcseconds | ~4× coarser | | Dynamic range | ~5.1 × 10⁵ | ~1/140 of kine | | Synthetic-test resolution | 100–125 microarcseconds (mean 113) | — | | Max recoverable apparent speed | at least 23c in synthetic tests | — |
Important caveat: the 4× figure is reconstruction resolution, not a physical diffraction limit — the hardware did not change. The authors stress these gains depend on the quality and quantity of the data and are not dataset-independent improvements.
From "Tracking Bright Spots" to "Measuring Velocity Everywhere"
Because time is continuous in the reconstruction, the team differentiates with respect to time and performs optical flow analysis, estimating the local motion of emitting structures across the whole video — a per-point instantaneous velocity field, rather than center positions of a few discrete components. As validation, trajectories integrated from optical flow reproduce earlier Gaussian-model measurements of bright components, and synthetic tests show kine + optical flow can recover apparent speeds up to at least 23c on this dataset, above the largest measured values.
The Velocity Data: Bright Spots Move as Fast as the Fluid
| Region | Apparent speed | |---|---| | Fastest mean flow (1–3 mas from core, south side) | ~12c (±0.2c) | | Within 5 mas | ~9–11c | | Outer diffuse emission | ~5–8c | | Bright components (inner region) | ~10–13c | | Mean bulk flow, same region | ~9–12c |
At the jet's reported viewing angle, the maximum apparent speed corresponds to a physical speed of ~0.997c. Velocity standard deviations of 3–6c, roughly independent of position, suggest a turbulent plasma flow with large apparent-speed fluctuations.
The Conflict: Shock Models Require Shocks Faster Than the Fluid
The 1985 Marscher–Gear jet shock model holds that disturbances form shocks that compress plasma, amplify magnetic fields, and accelerate particles — making bright spots shock fronts. It makes a hard prediction: shock propagation speed relative to the surrounding fluid must differ.
The kine observations fail this test. Bright components move at 10–13c while the same region's bulk flow averages 9–12c — the same order of magnitude, with no "shock outrunning the flow." Corroborating evidence: bright-spot positions show no clear correlation with polarization-degree peaks, whereas shock-compressed, ordered magnetic fields should raise polarization in sync.
Foschi preserves nuance: the result does not invalidate the general shock model, but casts doubt on this specific source. The paper is more direct: there is no evidence that the moving bright components are strong shock regions. The team leans toward them being locally "re-magnetized" regions releasing magnetic energy — possibly via magnetic reconnection, turbulence, or local particle acceleration — though no conclusion is settled.
What Else the Video Shows
In the continuous reconstruction, material is continuously ejected from the core. Some bright components initially follow ballistic trajectories, bend toward the jet axis at ~2–3 mas from the core, then fade into diffuse plumes. The 3C 345 jet extends ~13 mas, exiting the bright compact core westward before bending north as a diffuse plume.
Significance: From Taking Photos to Measuring Velocity Fields
The deeper value is a change in observational capability. For three decades, jet kinematics could only locate bright spots and measure a few component speeds. With continuous reconstruction plus optical flow, researchers can measure the projected velocity at any point in the jet. As the paper puts it, applied to multi-epoch observations, kine enables measuring instantaneous local velocity fields from high-resolution time-continuous video — not just tracking model-fitted components.
Limits: One Source, Not All
Resist extrapolation. Foschi explicitly states the challenge to shock models applies, at minimum, to 3C 345 alone. Synthetic tests covered only scenarios modeled on 3C 345 data, and the 23c recovery limit was measured on this dataset. Notably, several co-authors previously worked on resolution enhancement for Event Horizon Telescope black hole images, and Katie Bouman is on the author list. The team plans to apply kine to other variable sources.
What to Watch Next
- kine applied to other sources. If multiple blazars show "bright spots at bulk-flow speed," the shock model faces a systematic challenge; if only 3C 345 does, it may be a source-specific quirk.
- Joint analysis of velocity fields, polarization, and magnetic fields. The reconnection/turbulence interpretation needs finer polarized time series to discriminate.
- Method spillover. Neural-field reconstruction applies beyond jets to any sparsely sampled, multi-epoch variable-source imaging.
References
1. Foschi M., Zhao B., Fuentes A., Bouman K. L., Gómez J. L., Levis A. Video reconstruction of variable VLBI observations with neural fields. Nature, DOI 10.1038/s41586-026-10988-5 2. Live Science: Watch: Scientists have revealed the most detailed video of a black hole jet ever taken, 2026-09 3. Daily Galaxy: AI Video Reveals A New Picture Of Black Hole Jets And Challenges Shock Wave Theory, 2026-09 4. Science News Today: Bright features in a black hole jet may not be shock waves after all, 2026-09 5. phys.org coverage: 116 snapshots spanning 27 years stitched into a video, 2026-09-10/11 6. MOJAVE monitoring program (VLBA 15 GHz, 1995–2022, 116 epochs)