Market-structure
Tradeweb's CEO says the operational risk on Wall Street trading venues has shifted from a rogue trader's fat-finger to a rogue AI agent executing at machine speed, and the safeguards built for the old failure mode were not built for the new one
Billy Hult, Tradeweb's CEO since 2023 and with the venue since its founding, describes the emerging risk as not a single bad trade from a human error but an AI agent generating milliseconds of erroneous activity before anyone can intervene.
The fat-finger trade has a well-understood shape: one person, one bad order, one moment where a circuit breaker or a human on the other side of the desk has time to notice something is wrong before the damage compounds. Hult's point is that an AI agent operating on a venue like Tradeweb does not fail that way. It fails by generating a sustained stream of activity, correctly formed and syntactically valid at every step, that is nonetheless wrong in aggregate, and it can generate a lot of that activity before a human working at human speed even registers that something has gone off course.
That distinction matters for what kind of safeguard actually helps. A circuit breaker calibrated to catch an outlier single order does not necessarily catch an agent placing hundreds of individually unremarkable orders that only look wrong in combination. The defenses that worked for the fat-finger era, price bands, order-size caps, single-trade holds, assume the failure is a discrete event. An agent misbehaving at machine speed is closer to a process failure than an event, which argues for monitoring that watches trajectories and rates of change rather than just individual order validity.
It is also worth noting who is raising this. Tradeweb is not a retail brokerage worried about headline risk; it is fixed income and derivatives market infrastructure, the layer institutional desks route through. A venue operator flagging this in public, rather than treating it as an internal risk-management detail, suggests the industry conversation about AI agent safeguards on execution venues is still catching up to how far agent deployment has already gone on the buy side.
The uncomfortable part is that nobody has a proven answer yet. Kill switches and hard position limits are the obvious first move, but an agent that fails gradually rather than catastrophically can slip under limits calibrated for human-paced trading long before anyone pulls the switch. Venues built the current generation of safeguards for a threat model that assumed the fastest thing on the other end of an order was a person with a keyboard. That assumption is no longer safe, and the safeguards have not caught up to that yet.
Read the original: Yahoo Finance - Tradeweb's CEO says the operational risk on Wall Street trading venues has shifted from a rogue trader's fat-finger to a rogue AI agent executing at machine speed, and the safeguards built for the old failure mode were not built for the new one. Commentary is the independent editorial view of Share Trading; the original article is credited to its publisher.