LeoLabs has detected another debris-generating event in orbit. This is the latest in a series of events we've seen over the last 6 months. Together, they point to a worrying trend for space safety and security, and underscore the critical need for persistent sensing.

At 2:30 am UTC on 27 August 2026, LeoLabs detected a fragmentation of a Chinese Long March rocket body, CZ-6C (NORAD ID 100472). The object’s orbit was 396 km perigee and 482 km apogee.

A few weeks ago, Novo Nordisk announced that its long-awaited ZEUS trial had failed. The study tested whether blocking IL-6, a key driver of inflammation, could reduce cardiovascular events -- a test of the hypothesis that inflammation is a treatable cause of heart disease. This was not a small undertaking, as the trial enrolled 6300 people over several years and must have cost in the hundreds of millions (I would estimate between $300M and $600M). @MariosGeorgakis wrote a great post-mortem trying to answer the question: why did the trial fail? I think there are two things that make his post great: 1. First, it goes through the evidence we had prior to this trial and shows just how hard it is to predict results in human trials for novel targets (and why companies often choose to avoid them). Based on the available evidence, the trial should have worked. The most important points are the data from the CANTOS trial (targeting IL-1), which strongly suggests that inflammation is promising as a treatable cause of cardiovascular disease. We also have convincing genetics data showing that IL-6 is causally implicated in cardiovascular disease. So far, genetic evidence remains the biggest booster of clinical trial success for a new target. 2. Second, we need to invest in better biomarkers, which could provide early biological signals of whether a drug is working. Today, we are living through what I consider a frankly unsustainable situation: a developer can spend half a billion dollars on a late-stage trial only for it to fail. In this case, the drug reduced inflammation as measured by C-reactive protein (CRP) in early trials, but didn't prevent heart attacks in later ones. CRP is a very poor biomarker as it is not predictive of clinical efficacy, but it's all we have. Having better early indicators of success would catch bad candidates years sooner and at a fraction of the cost, freeing that capital for therapies with a real chance of working. Unfortunately, as I explained in an article for @IFP last year, the incentives around developing clinically validated and useful biomarkers are skewed, since they function essentially as public goods: there is little possibility for any individual developer to make money from developing a clinically useful biomarker and making it available, even if doing so would benefit the field immensely. We still do not know why the trial failed. One thing I am worried about is that investors will see this as an invalidation of the inflammation hypothesis in cardiovascular disease and it will deter investment -- I think there are signs this is happening. This could just be a bad draw from the distribution of possible outcomes! But it is entirely rational that developers would avoid spending hundreds of millions of dollars to find out that their drug dailed. Again, my only prescription here is to get better biomarkers that indicate success earlier on. This would allow us to test many drugs targeting inflammation in earlier stage trials and progress only the actually promising ones, if such promising drugs exist at all.

Watch Starship's thirteenth flight test xitter.cf/i/broadcasts/1AJEmmYdM…

All systems are looking good, and weather is favorable for liftoff. Propellant load is about to begin for Falcon 9’s launch of the @SemperCitiusSDA's third Tranche 1 data transport mission from California → spacex.com/launches/sda-t1tl…

China just reached a near Earth asteroid. In a few months, we'll begin our own mission to one, at a fraction of the traditional cost. Imagine what would be possible if low-cost commercial deep space missions became a national capability.

Size estimation based on the official Tianwen-2 image: 469219 Kamo'oalewa appears to be ~20m -30m across in LOS projection, making it one of the tiniest natural rocks, even visited by a human spacecraft. Compare with the probe itself, the sampling operation will be exceptionally challenging.

Tianwen-2 had arrived. Here's asteroid Kamo'oalewa seen from 20 km away xinhuanet.com/20260706/eb8cb…

China's Tianwen-2 probe is matching orbits with space rock (469219) Kamo'oalewa, in a 0.90 x 1.10 AU x 7. 8 deg solar orbit. The rock is about 20 meters across and has a mass around 7000 tonnes, and is the smallest asteroid so far to have been visited by a spacecraft