Two figure skaters on a lake frozen to the horizon. They are so far apart that their silhouettes turn into dots, and their mutual attraction becomes almost an article of faith. By Newton's law, they should circle in an unimaginably slow waltz. But telescopes see otherwise. The pairs whirl faster, as if the ice itself breathes beneath them, imparting an acceleration that shouldn't be there. In space, it appears as a mysterious strengthening of gravity at incredibly small accelerations — hundreds of times weaker than Earth's pull.

Wide binary systems are an ideal testing ground for the nature of gravity. Unlike galaxies, there's almost no dark matter here to call upon to explain strange velocities. If gravity proves stronger than Newtonian here too, we'll have to acknowledge a crack in the theory's foundation. That's exactly what Vera Rubin suspected on galactic scales, now being tested on individual stellar pairs. In a fresh study, scientists sifted through Gaia data, passing it through three independent filters. They cleaned out spurious close encounters, demanded pristine photometric signals, and cut pairs where one partner could be an overlooked red dwarf. A clever numerical theatre also accounted for 'third wheels' — unseen stellar companions whose presence could mimic the anomaly. But all in vain.

After all the cleaning, the median speed at high accelerations, where Newton rules, perfectly matched the theory — a sign that the data are honest. But as soon as they stepped into the realm of ultra-weak fields, the speed jumped: gravity strengthened by a factor of one and a half. The statistical significance of the deviation soared past five sigma, and the version with unaccounted triple systems failed. To salvage the Newtonian picture, one would have to postulate an absurd number of hidden companions — so many that it contradicts observations under normal conditions.

What's next? The blow to the 'dark side' of cosmology is becoming ever more acute. A bold idea resurfaces, voiced in bypassing Einstein — that gravity itself changes its character on large scales. By the end of the decade, fresh Gaia data supplemented with 3D velocities from ground-based spectroscopy will allow us to reconstruct the orbits of thousands of such pairs and either confirm the breakdown of classical gravity or reveal unknown systematic effects. Along the way, we'll clarify how stellar evolution and metallicity influence the birth of triple systems. And we'll add an intriguing twist: the critical acceleration where Newton breaks down almost matches a fundamental constant derived from the speed of light and the age of the universe. As if nature itself is hinting where to look for new physics.