In short: Her own company (founder) — talk-her-book caveat. "The most material aspect of our business is robot data": capturing "human priors" — the embedded knowledge of how people perform physical tasks — and packaging it "such that it's easy to train on for these large AI companies." Much of the work is the "grunt work" of cutting eight-hour messy capture videos into meaningful segments. Focus: tasks a human performs directly (assembly, sorting, manipulation), not machine-operator footage. Software that cost ~$200,000 to build last year could now be replicated "in like 3 weeks and $20,000."
Wilde's own private company. Robots learn tasks from examples, and the best examples are recordings of humans doing the job. Inheritance turns messy real-world footage — hours of video including breaks and chatter — into clean, labelled clips that AI labs can train robots on.
Because she profits from AI demand for training data, treat her bullishness on "demand for data continues to be underestimated" as informed but self-interested.
Nothing matches this filter.
Verbatim excerpts from the public transcripts (auto-pulled at each mention's timestamp, lightly cleaned). Timestamps link into the video; "source page" opens that commentator's full analysis at this row. Click a mention's header line to expand it (one open at a time). For personal study — not investment advice.