The June US jobs report shows 57 thousand new jobs - sharply below the roughly 115 thousand expected. The temptation is to blame AI. Except the BLS report (the US Bureau of Labor Statistics) doesn't say that.
- BLS's June report (July 2): only +57 thousand jobs, below the ~115 thousand expected; unemployment at 4.2% because of a drop in participation to 61.5% (the lowest since March 2021).
- Household employment falls by 507 thousand; hospitality and restaurants lose 61 thousand jobs.
- The report doesn't attribute the weakness to AI - our honest read is 'we don't know yet,' no single-cause headlines.
57 thousand new jobs in June. Consensus was waiting for about 115 thousand. Anyone who writes about AI will feel the pull to draw an equal sign - 'here it is, the machines started eating the jobs.' Our job here is different: we stop and read what the data actually says, not what we want it to say.
No illusions - this is a discipline test. A weak labor market is the canvas everyone paints their favorite story on - both 'the machines are eating the jobs' and 'AI has nothing to do with it.' The data gives a duller picture - shrunk participation, seasonal weakness at restaurants, a collapsed household survey. None of these numbers carries a label saying 'cause: AI.'
The honest answer for now is one word. Too early. AI might be squeezing hiring at the margins, it might be rates, the season, demographics. If someone sells you a single explanation today, they're selling a confidence the numbers don't have yet. We're waiting for the signal, not the headline.