AI moves from experimentation to governed deployment as productivity gains grow
The Derivatives Management Insight Report is free to read for qualifying executives, to request your copy visit: https://www.acuiti.io/derivatives-management-insight-report-q3-2026/
London – 22 September 2026: Firms across the derivatives market have moved beyond the discovery phase of artificial intelligence and into governed deployment, with 44% now reporting significant, quantifiable time and cost savings from the technology but taking very different approaches to governance, the Q3 Acuiti Derivatives Management Insight Report has found.
The quarterly Derivatives Management Insight Report, which was released today, consolidates Acuiti’s longstanding derivatives focused buy-side, proprietary trading, sell-side execution and clearing reports, into a single market-wide publication.
The report is based on a survey of the Acuiti Derivatives Expert Network, a group of senior executives from across the sell-side, hedge funds, asset managers and proprietary trading firms. Each quarter, members of the Expert Network suggest topics and questions for the report that are then circulated across the Network via an anonymised survey.
This quarter’s report looks at the governance around AI deployment and found that the overwhelming majority of firms have in place formal frameworks around AI usage among employees. As part of those frameworks, two thirds of firms limit which AI models employees can use while a third limit both the models and the use cases.
There were significant differences between company types in terms of AI controls. All banks reported that they controlled employees’ usage of AI and two thirds had established a specific AI risk committee to monitor AI policy.
Prop firms were at the other end of the spectrum with around half allowing relatively free use by employees and third currently having no formal governance policy in place. Hedge funds were more likely to govern AI within existing risk committees, rather than establishing a new committee, while around a fifth allowed relatively free usage among staff.
AI is having a tangible improvement on productivity. Forty four percent of network members reported significant, quantifiable time and cost savings from AI, with a further 42% noticing benefits they have not yet been fully quantified.
This quarter’s report also asked about 24/7 trading and found relatively consistent scepticism across the market towards extending trading in traditional, institutional derivatives contracts to a 24/7 cycle.
Sixty-five percent of proprietary trading firms were opposed to weekend trading in contracts such as oil futures, with just 5% definitely in favour and the remainder potentially supportive.
That view is mirrored on the sell-side, where 48% of members said that none of their institutional clients were interested in 24/7 trading functionality and just under a third reported only a handful were.
Other key findings in this quarter’s report include:
- Two-thirds of firms in the network performed better than in H1 2025, with only 12% seeing any deterioration year on year, while 64% came in ahead of a typical year
- CME’s cloud migration is likely to be a challenge for sell-side firms that do not have an existing relationship with Google Cloud: almost a third of sell-side execution desks believe onboarding a new cloud provider would take longer than 12 months
- Institutional demand for perpetual futures is building: only a small minority of sell-side execution desks offer perps today, but 53% are either working on an offering or considering one, with risk management and modelling around the funding rate the most commonly cited challenge
- The buy-side is sceptical on BoE gilt repo plans: 88% expect the Bank of England’s proposed minimum haircut requirements on gilt repo to increase overall funding costs and 63% believe the measure will reduce liquidity in the gilt market
- Clearing relationships are stable but tilted towards expansion: 75% of buy-side and prop firms plan no change to their number of FCM/GCM relationships, and among those that do, far more are looking to increase than reduce, primarily to cut reliance on a single clearer
- Crypto data infrastructure remains self-managed: three-quarters of prop firms active in digital assets source data directly from centralised exchange APIs and 63% use decentralised exchange or on-chain data, a marked departure from the vendor-led models of traditional asset classes
“The early phase of AI adoption was defined by experimentation and a degree of caution,” says Ross Lancaster, head of research at Acuiti. “What this quarter’s report shows is a market that is now building the governance and controls around its usage.”
Looking ahead to the final quarter of the year, 68% of network members were optimistic about the performance of their derivatives business, with buy-side firms the most positive and prop trading firms slightly more pessimistic than the market average.
Download the full report here: https://www.acuiti.io/derivatives-management-insight-report-q3-2026/
#######
For more information, contact Will Mitting at Acuiti
Tel.: +44 (0) 203 998 9190
Email: willmitting@acuiti.io
To join the Acuiti Derivatives Expert Network, contact Marco Pisanello at marcopisanello@acuiti.io
About Acuiti
Acuiti is a management intelligence platform serving senior executives across the global derivatives, ETFs and fixed income markets. Through its exclusive expert networks, Acuiti gathers anonymised insight from banks, brokers, proprietary trading firms and buy-side institutions, publishing benchmarking data, quarterly sentiment analysis and bespoke research on the trends shaping the industry. Founded in 2019, Acuiti brings greater transparency to operations across global markets. Visit acuiti.io.
#######