Machines now a main audience for financial data, says SEC chief economist
Machines are playing an ever larger role in how financial data is accessed, processed and understood, according to Joshua White, Chief Economist of the US Securities and Exchange Commission (SEC). Speaking at the Investment Company Institute (ICI) conference in Nashville on 17 September, he argued that AI is changing the economics of information itself: how fast it can be processed, how widely it can be compared, and how much can be extracted from it.
The SEC’s own datasets show a dramatic shift. Through July, the Form N-PORT registered fund dataset was downloaded more than 14 times as often as in the same period last year. Money market fund data from Form N-MFP was downloaded nearly seven times as often, and business development company data almost 75 times as often. White called these increases extraordinary.
Most of this growth is likely to be from machines accessing and analysing the data. Financial information now, more than ever, has two audiences: people and machines. Or perhaps that should be machines and people. For most of the history of securities regulation, disclosure was designed around a human audience. Now, however, data flows directly into databases, AI models work with that information at a scale few could have imagined a decade ago, and they need consistent, structured information to do it well. AI models aren’t magic, they’re probabilistic, so every ambiguity raises the risk that information is misunderstood. Where investment decisions are on the line, it makes more sense to tell machines what a number means than force them to guess.
White also pushed back on the idea that AI makes data quality less of a concern. Structured data gives AI context, standardisation makes comparison easier, and consistent definitions allow thousands of filings to be analysed systematically instead of one document at a time. The AI revolution will run on better information far more than on sheer volume, and tagged, well-defined data is what makes that possible.
Well worth reading the full speech here.
