Materials, energy and the research frontier shaping the next decade.
13 articles published
Stanford researchers used AI models to generate functional bacteriophage genomes with no natural counterpart. The result is a genuine scientific milestone and a pointed test of how DNA-synthesis screening and biosecurity rules are meant to work.
The open questions catalogued from Paul Erdős's vast output are unusually well suited to the kind of search that language models are good at, and unusually poorly suited as evidence about deep mathematical reasoning.
A long-standing vulnerability in widely used Thermo Fisher forensic DNA software let researchers add or remove profiles from evidence files, exposing the fragility of digital chain of custody.
Tens of thousands of tracked objects and a much larger population of untrackable fragments now share low Earth orbit with an expanding satellite population. The maths of that crowding is getting harder to ignore.
Thirteen deaths confirmed, hundreds of thousands under evacuation instruction and a rescue window measured in days. A clear account of the situation and the response.
A new claim of ambient-pressure, room-temperature superconductivity has generated familiar excitement — and familiar scepticism. The replication process itself is the more interesting story.
Heat domes, vapour pressure deficit and fuel moisture sound technical. They are also the three numbers that decide whether a spark becomes a headline.
Most of the extra energy trapped by greenhouse gases ends up in the ocean, not the air. That measurement is unusually steady, and unusually informative about what comes next.
Similar magnitudes produce wildly different death tolls. The difference is almost entirely engineering, drills and building codes enforced over decades.
A more precise descendant of the original CRISPR technique has moved from laboratory demonstration to authorised treatment. The gap between the two remains a useful lesson in how gene editing actually gets to patients.
Fusion has cleared the physics threshold of net energy gain more than once now. Turning that into a power station is a separate, and much harder, problem.
NASA's return to the Moon has slipped repeatedly. Here is a sober look at which delays are routine engineering caution and which point to deeper structural problems.
Fabs caught up. Grids did not. The binding constraint on computing has quietly moved from silicon to electricity.