Sony AI’s autonomous robot “Ace” has achieved what researchers describe as the first instance of expert-level play by a machine in a commonly played physical sport. A study published in Nature on April 22, 2026, and accompanying demonstrations in Tokyo show Ace—an eight-jointed robotic arm with a network of high-speed vision systems including event-based sensors, nine synchronized cameras and three gaze systems—using model-free reinforcement learning to track spin, react in milliseconds and execute competitive shots under International Table Tennis Federation rules. In April 2025 Ace won three of five matches against elite human players; it initially lost to two professional players but later beat professional opponents in December 2025 and in early 2026. Tests were held on a full-size court with licensed umpires. The system returned a high proportion of high-spin shots, scored multiple direct points on serve and sometimes executed maneuvers human players had not expected. Authors and Sony AI say the architecture demonstrates that simulated learning can transfer to fast, precise, real-world interaction tasks and point to broader applications in manufacturing, service robotics and other time-critical domains.

A study led by Nicholas Makris at the Massachusetts Institute of Technology, published in Scientific Reports on April 23, 2026, finds that rice seeds exposed to the sound and vibration of falling raindrops germinate substantially faster than seeds not exposed to droplet impacts. Researchers submerged roughly 8,000 Oryza sativa seeds in shallow water and subjected some trays to streams of rain-like droplets while keeping controls in identical water without droplet impact. Depending on conditions and seed position in puddles, germination was accelerated by roughly 24–40%. The team attributes the effect to pressure waves transmitted through water that jostle microscopic starch-rich statoliths inside seed cells; those movements plausibly trigger early growth processes. The study also measured very high underwater sound pressures produced by single drops in shallow puddles. Authors caution that other factors — such as aeration, pressure changes or extra material from drops — might contribute, and they call for further biological and cross-species testing to confirm mechanisms and field relevance.

A NASA Office of Inspector General audit released in April 2026 warned that development and testing delays for next-generation spacesuits could push demonstrations — and therefore readiness for lunar surface operations — to 2031, potentially slipping Artemis landing plans well beyond NASA’s current target of 2028. NASA contracted Axiom Space (after Collins Aerospace exited in 2024) under firm-fixed-price, service-based xEVAS deals to supply lunar and microgravity suits; the OIG said that acquisition approach and overly optimistic schedules increased technical and schedule risks. NASA leadership and Axiom dispute the 2031 projection: Administrator Jared Isaacman and Axiom CEO Jonathan Cirtain say they remain committed to a 2027 in-space demonstration and supporting a 2028 landing. The report also flagged interoperability problems (different suit interfaces across lander designs), the lack of a common suit standard, and the danger that the International Space Station’s retirement around 2030 could leave insufficient testbeds. NASA has inserted a low-Earth-orbit demo flight into Artemis sequencing to de-risk systems, but other pressures — Human Landing System readiness, procurement protests and budget uncertainty — add to schedule vulnerability.

NASA unveiled the completed Nancy Grace Roman Space Telescope at Goddard Space Flight Center on April 21, 2026 and is targeting an early September 2026 launch, with a contractual latest launch date of May 2027. The observatory, built for wide-field infrared surveys, pairs a 300-megapixel Wide Field Instrument with a high-contrast coronagraph to hunt exoplanets and map the large-scale structure of the cosmos to probe dark matter and dark energy. NASA expects Roman to amass an enormous archive during its five-year primary mission — on the order of tens of thousands of terabytes — and to identify hundreds of millions of galaxies and tens of thousands of exoplanets. The telescope will lift off on a SpaceX Falcon Heavy from Launch Complex 39A at Kennedy Space Center and is managed by Goddard with participation from JPL, Caltech/IPAC and the Space Telescope Science Institute. The project, named for astronomer Nancy Grace Roman, was completed ahead of schedule and under budget after more than a decade and about $4 billion of investment.

Researchers led by Maya Bach and Jennifer Jacquet published a PLOS Climate paper on April 22, 2026, analysing 1,233 environmental claims from 33 of the world’s largest meat and dairy companies across 2021–2024. Using a validated greenwashing framework, the team found 98% of claims were deceptive or misleading; 68% were climate‑related, only 356 (29%) had company‑provided supporting evidence, and scholarly literature backed just three claims. Seventeen companies now hold net‑zero pledges, but these largely rely on offsets rather than direct emissions reductions. The authors highlight token initiatives — a regenerative pilot covering 24 farms (0.0019% of operations) and trivial packaging tweaks — that contrast with the sector’s disproportionate emissions role: animal agriculture accounts for at least 16.5% of global greenhouse gases and a large share of food‑system emissions. The paper warns that vague, unverifiable promises risk misleading consumers and policymakers and may delay substantive action needed to meet climate targets.