Controlling quantum randomness

The empty space of vacuum is a shifting field of random fluctuations in electromagnetic fields. These fluctuations can be harnessed as a source of random numbers and, in turn, for the generation of random bit strings. By injecting a small bias field into their experimental system, an MIT team led by Professors Marin Soljačić and John Joannopoulos and featuring ISN Principal Research Scientist Steve Kooi demonstrated control over the probability distributions of bit strings; by attenuating the bias field, the team could manipulate the probability from truly random to deterministic. This control marks a massive, essential step toward the development of ultra-precise field sensing and photonic probabilistic computing.
A closeup image of a laser setup showing green light zig-zagging through a translucent rectangle

Experimental setup to generate tunable random numbers from vacuum fluctuations. Image: Courtesy of the researchers.

Schematic of the tuning of macroscopic probability distributions by biasing vacuum fluctuations.

Tuning macroscopic probability distributions by biasing vacuum fluctuations.
(A and C) Fluctuation snapshot, Hamiltonian, and associated probabilities for an (A) unbiased and (C) biased bistable system. In (C), a vacuum-level bias field is injected in the cavity. (B) Schematic of quantum optical multistable system. (D) Stochastic amplitude trajectories for a biased optical parametric oscillator (OPO). One hundred trajectories are plotted in each graph, and time is normalized to the cavity lifetime tcav. (E) Probability of state α(1) as a function of bias intensity b. Image: Charles Roques-Carmes et al., Biasing the quantum vacuum to control macroscopic probability distributions. Science 381, 205-209 (2023).

Dr. Charles Roques-Carmes adjusts the experiment setup.

Dr. Charles Roques-Carmes, one of the leaders of the research, adjusts the experiment setup. Image: courtesy of the researchers.

Dr. Yannick Salamin adjusts the experiment setup.

Dr. Yannick Salamin, one of the leaders of the research, adjusts the experiment setup. Image: courtesy of the researchers.