Wall Street Builds AI to Decode Warsh's Quiet Fed — SkimNews
SkimNews Take
When the Fed withdraws explicit forward guidance, decoding becomes the policy transmission mechanism itself — concentrating informational advantage in firms that can synthesize alternative signals, which risks fragmenting market expectations rather than anchoring them.
Get the Finance newsletter
Daily finance — markets, central banks, M&A, the prints that move money. Free.
- F/m Investments launched "WarshGPT," an AI chatbot built with Anthropic's Claude model for under $1,000 in roughly two weeks, parsing nearly 1,800 Warsh documents and transcripts to help clients predict Fed thinking — though it won't speak as Warsh or offer forecasts.
- Kevin Warsh, since becoming Fed Chair in May, has slashed forward guidance: the June FOMC statement ran ~130 words (down from 300+), and just 5% of his first press conference sentences covered policy-relevant topics versus a 27% average under Jerome Powell, per UBS.
- UBS strategist Elena Amoruso told clients Warsh's debut policy comments were "overwhelmingly hawkish" on labor, growth, and inflation, calling Fed-speak "the most high-value data set … in terms of how much one word can move dollars."
- JPMorgan Asset Management chief global strategist David Kelly said his team will lean more heavily on individual FOMC member speeches if the Fed scraps the "dot plot," though he expects major communication changes to take several months and may not be as drastic as feared.
- MacKay Shields senior macroeconomist Steve Friedman, a former New York Fed economist, warned that less clarity on the Fed's reaction function is "a negative for the economy" but could become "a source of alpha" for investors — and flagged Governor Christopher Waller as a "bellwether" for the broader committee.
- Markets are split on the September meeting: CME FedWatch shows Fed funds futures pricing a ~59% chance of a rate hike, while Kalshi traders expect rates to stay unchanged — a divide that reflects the new uncertainty Warsh's quieter posture has introduced.
Why it matters: Warsh has cut policy-relevant commentary by roughly 82% relative to Powell's average (5% vs. 27% of press conference sentences), institutionalizing information scarcity at the Fed. That gap will be filled by AI tools like WarshGPT, hires of Fed alumni, and fees for premium research — advantages that compound toward well-resourced firms while retail investors, as UC-Irvine's Richardson notes, will find it "much harder" to keep up.
Ask SkimNews
